Resnet50 architecture for multiclass brain tumor classification using fine-tuning technique
Technology Developments in Computer Intelligence and their Applications in the era of Industry 5.0, 2025, DOI Link
Swarm intelligence: Theory and applications in fog computing, beyond 5G networks, and information security
Swarm Intelligence: Theory and Applications in Fog Computing, Beyond 5G Networks, and Information Security, 2025, DOI Link
View abstract ⏷
This book offers a comprehensive overview of the theory and practical applications of swarm intelligence in fog computing, beyond 5G networks, and information security. The introduction section provides a background on swarm intelligence and its applications in real-world scenarios. The subsequent chapters focus on the practical applications of swarm intelligence in fog-edge computing, beyond 5G networks, and information security. The book explores various techniques such as computation offloading, task scheduling, resource allocation, spectrum management, radio resource management, wireless caching, joint resource optimization, energy management, path planning, UAV placement, and intelligent routing. Additionally, the book discusses the applications of swarm intelligence in optimizing parameters for information transmission, data encryption, and secure transmission in edge networks, multi-cloud systems, and 6G networks. The book is suitable for researchers, academics, and professionals interested in swarm intelligence and its applications in fog computing, beyond 5G networks, and information security. The book concludes by summarizing the key takeaways from each chapter and highlighting future research directions in these areas.
An empirical analysis of evolutionary computing approaches for IoT security assessment
Kumar Sahu V., Pandey D., Singh P., Haque Ansari M.S., Khan A., Varish N., Khan M.W.
Journal of Intelligent and Fuzzy Systems, 2025, DOI Link
View abstract ⏷
The Internet of Things (IoT) strategy enables physical objects to easily produce, receive, and exchange data. IoT devices are getting more common in our daily lives, with diverse applications ranging from consumer sector to industrial and commercial systems. The rapid expansion and widespread use of IoT devices highlight the critical significance of solid and effective cybersecurity standards across the device development life cycle. Therefore, if vulnerability is exploited directly affects the IoT device and the applications. In this paper we investigated and assessed the various real-world critical IoT attacks/vulnerabilities that have affected IoT deployed in the commercial, industrial and consumer sectors since 2010. Subsequently, we evoke the vulnerabilities or type of attack, exploitation techniques, compromised security factors, intensity of vulnerability and impacts of the expounded real-world attacks/vulnerabilities. We first categorise how each attack affects information security parameters, and then we provide a taxonomy based on the security factors that are affected. Next, we perform a risk assessment of the security parameters that are encountered, using two well-known multi-criteria decision-making (MCDM) techniques namely Fuzzy-Analytic Hierarchy Process (F-AHP) and Fuzzy-Analytic Network Process (F-ANP) to determine the severity of severely impacted information security measures.
Assessing Different Tools Employed in Auto-segregation of Plastic Waste
Kumar P., Jones H., Pathak P., Mallampalli S.N., Konathala P.R., Vasireddy P.V., Mugdha M., Singh P.
Plastic Footprint: Global Issues, Impacts and Solutions, 2025, DOI Link
View abstract ⏷
Segregation and recycling of waste have been recognized to be vital for both economic and ecological reasons, with industries demanding high efficiency. However, current studies on automatic waste detection lack benchmarks and widely accepted standards, making comparisons difficult. This chapter addresses the issue by employing deep learning for waste classification into two categories: recyclable and non-recyclable materials. The dataset explained in this chapter has been compiled from various sources, ensuring a diverse representation of waste types. The garbage classification model is trained on the MobileNetV2 deep neural network architecture, enabling rapid and accurate classification of domestic waste. The model achieved an impressive 94% absolute accuracy, in 15 epochs, showcasing its efficiency and effectiveness. The applications of this research aim to provide better waste categorization and encourage more widespread recycling practices. By leveraging deep learning techniques, the proposed model streamlines the waste-sorting process, potentially saving substantial labor, material, and time costs associated with manual sorting. The development of efficient and accurate waste classification systems can significantly contribute to environmental sustainability efforts and promote a circular economy. The proposed waste classification model automates plastic segregation by making the process more efficient and faster.
Secure Image Transmission in UAV Applications using 2D Chaos and Hybrid Watermarking Method
Shaik V., Dixit K., Bhandari V., Bhargav R., Singh P.
6th IEEE International Conference on Recent Advances in Information Technology, RAIT 2025, 2025, DOI Link
View abstract ⏷
The rapid growth of Unmanned Aerial Vehicles (UAVs) has been largely driven by various fields, ranging from agriculture to entertainment. In most of these fields, UAVs serve as platforms for capturing images and transmitting data to other UAVs or ground stations. However, this process is vulnerable to unauthorized access and various security risks. Digital Image Watermarking (DIW) has emerged as a potential solution to strengthen the authentication of digital image transmission within UAV systems. To address this issue, a hybrid DIW technique employing Discrete Wavelet Transform (DWT), Singular Value Decomposition (SVD), 2D chaotic system, row permutation, column permutation, and Arnold transform encryption is proposed. The binary watermark image containing authentication/copyright information is encrypted using 2D Chaos, row-column permutation and Arnold transform. Then encrypted watermark is embedded in the original image using DWT-SVD. The reverse process is carried out at the receiver end to extract the watermak information. The proposed method enhances the authenticity of watermark images during transmission. Experimental results demonstrate that the proposed method achieves high imperceptibility, with Peak Signal-to-Noise Ratio (PSNR) values exceeding 45 dB, Structural Similarity Index (SSIM) values close to 1, Normalized Correlation (NC) of 0.99, and a Bit Error Rate (BER) of 0.0054. This indicates that the method exhibits strong robustness against various attacks, ensuring the reliability and authenticity of the transmitted images.
On the Design of Unstructured Student Feedback Summarization Model using Transformer Architecture for Quality Education
Jiru A.W., Tuyisenge F., Thakkar H.K., Singh P.
Procedia Computer Science, 2025, DOI Link
View abstract ⏷
In the past few years, the Universities across the world has increased their focus on teaching-learning process improvisation. In this regards, the Universities periodically collect the students' feedback through pre-defined questionnaire using online tools such as Google forms. However, such pre-defined questionnaire does not allow the students to express their opinion outside the questions. Therefore, students unable to convey their true feedback and the overall purpose of the feedback mechanism collapses. The alternative mechanism is to allow students to write a free style feedback within certain word limits. However, it is highly challenging to summarize such freestyle feedback manually to know the overall conclusion of the entire class. In this paper, we have engaged a Bidirectional Encoder Representations from Transformers to design an extractive summarization model for unstructured student feedbacks to improve the quality of educations. The proposed model allows the students to provide feedback about the course and also allow the course faculties to be able to summarize received student feedbacks in its true spirit by extracting the key ideas from students' feedback. The experimental results show that the if produced summaries are too small then it unable to include all the important aspects. However, when produced summaries are sufficient in size then it successfully captures all the important aspects with PrecisionBERT, RecallBERT, and FscoreBERT of 78%, 61%, and 0.61%, respectively. The proposed model is compared with three existing schemes and it provides the improved results for PrecisionBERT, RecallBERT, and FscoreBERT.
OpenCV Algorithm for IoMT-Based Patient Emotion Pattern Analysis
Jadeja D., Singh P., Thakkar H.K., Kumar V.D.R.
Health 5.0: Concepts, Challenges, and Solutions, 2025, DOI Link
View abstract ⏷
In the era of the Internet of Medical Things (IoMT), providing excellent patient care requires an understanding of patient emotions and how they are feeling. The method described in this abstract, which makes use of OpenCV algorithms, is new for examining patient emotional patterns. This technology uses IoMT devices in conjunction with physiological signals from the body and facial expressions to ascertain the emotional states of patients. OpenCV’s powerful image processing methods, such as feature extraction, emotion recognition, and facial identification, are used to do real-time facial cue analysis. In order to contextualize emotions, IoMT devices also collect data on physiological traits including skin conductance, body temperature, and heart rate variability. Machine learning models are given the processed data in order to find connections linking emotional states to bodily reactions. Convolutional neural networks and recurrent neural networks are two examples of deep learning algorithms that are used to extract complex patterns from merged data. An important asset of the system is its versatility; it can be tailored for a variety of medical situations, such as chronic pain management, mental health disorders, and post-operative care. In the end, real-time analysis improves patient well-being and the general standard of healthcare services by enabling prompt responses, such as warning healthcare personnel of distress signals.
Secure Digital Image Transmission in Unmanned Aerial Vehicles Using Lightweight Encryption and Watermarking Approach
Singh P., Shaik V., Bhargav R.
Lecture Notes in Electrical Engineering, 2025, DOI Link
View abstract ⏷
The rapid evolution of Unmanned Aerial Vehicles (UAV) has been largely driven by its extensive deployment in surveillance, agriculture, forestry, archaeology, military operations, environmental monitoring, and other similar domains. Across majority of these applications, UAV serve as image capture platforms, relaying captured data to other UAV or ground stations. However, transmission of digital image through UAV channels remains susceptible to unauthorized access and a spectrum of deliberate or inadvertent security breaches. Digital Image Watermarking (DIW) has emerged as a promising solution for fortifying the security of digital image transmission within UAV frameworks. To address this issue, a secure digital image watermarking scheme using RDWT-SVD transform domain is proposed. At the sender node, an encrypted watermark (authentication or copyright information) is interleaved in the host digital image. The receiver node extracts the watermark and decrypts to validate the authenticity of the received image. A novel light weight symmetric cryptographic-based four-level encryption approach is proposed for encryption/decryption of the watermark for high UAV image security. This encryption technique utilizes chaotic map and XOR operation to ensure high security at minimal computational expense. Experimental evaluations demonstrate the effectiveness of the proposed scheme, showcasing high visual quality and robustness (with an average Peak Signalto-Noise Ratio (PSNR) of 42.93 dB, Average Structural Similarity Index (SSIM) of 0.99, Normalized Correlation (NC) of 0.99, and Bit Error Rate (BER) of 0.0014 calculated across 100 images captured by UAVs. Both subjective and objective analyses affirm the heightened security and computational efficiency of the proposed encryption scheme. Furthermore, the scheme exhibits resilience against various image processing attacks, as validated through comparative assessments against recent state-of-the-art approaches. The presented DIW framework holds promise for applications such as copyright protection.
Ensuring integrity and security of medical image transmission in IoMT using highly imperceptible and robust watermarking approach
Singh P., Devi K.J., Nizami T.K., Prakash C.S., Thakkar H.K., Hussain S.A., Mallik S.
Scientific Reports, 2025, DOI Link
View abstract ⏷
With the technological revolution, the Internet of medical things (IoMT) has developed to be of immense benefit. In IoMT, medical images and patients’ data are widely transmitted through private/public network. An ideal transmission should not jeopardize the security, confidentiality, authenticity, authorization, or integrity of medical data/images. To ensure effective transmission and address the aforementioned issues, this paper proposes a blind region based medical image watermarking approach where a medical image is partitioned into region of interest (ROI) and region of non-interest (RONI). To ensure ROI intergrity, localized tamper detection and recovery bits (LTDRB) are generated. For precise diagnosis, patient’s electronic health record (EHR) and LTDRB are embedded in RoNI using hybrid DWT-SVD. No embedding is done in RoI to maintain its integrity and high visual quality. To ensure the security and confidentiality of EHR, a novel encryption scheme using Magic Square technique with low computational cost is proposed. Experimental results demonstrates that the proposed scheme provides high imperceptibility (Avg. PSNR>55 dB, SSIM ≈ 1 and BER ≈0), robustness, security at low computational cost and high accuracy in tamper detection and recovery. A comparative study with some of the latest related research shows that the proposed scheme provides imperceptibility and robustness at par. However, the proposed scheme shows superior performance by providing higher EHR security at low computational cost and higher accuracy in ROI tamper detection and recovery, which other schemes have overlooked.
EfficientNet B0 Model Architecture for Brain Tumor Detection and Classification Using CNN
Ratna Kumar V.D., Muchina F.E., Hussain M.M., Singh P.
Advancements in Artificial Intelligence and Machine Learning, 2025, DOI Link
View abstract ⏷
Brain tumors are a life-threatening disease, and a lot of people are losing their lives. These brain tumors are abnormal cells that develop in and around the brain. This research explores the cutting edge of medical imaging processing, focusing on enhancing the detection and categorization of brain tumors. EfficientNetB0 is the most advanced deep learning architecture that has been thoroughly compared with other deep learning models in order to improve brain tumor classification accuracy using the Kaggle MRI image dataset with 7023 images. The drawbacks of manual tumor identification techniques are discussed, and precise classification using deep neural networks is proposed, with special attention to the transition from binary to multiclassification. This chapter's primary focus is on improving and optimizing the EfficientNetB0 model through the addition of trainable layers on top of its basic architecture. Several techniques are used like global average pooling for spatial and dimensionality reduction with reduced parameters, dropout to drop layers, and dense net with softmax for multiclass classification. Concurrently, strategic layer freezing is used to refine the deep learning models for foundation design. The results show that the finetuned EfficientNetB0 model with hyper-parameter optimization guarantees exceptional brain tumor accuracy. EfficientNetB0 has achieved a good accuracy of 99.7% and a precision of 99.5% compared to Resnet50, VGG16, InceptionV3 and Xception. This work presents a unique deep-learning method in accordance with a transfer learning strategy for assessing brain cancer categorization accuracy using the enhanced ResNet50 model. As we advance the state-of-the-art, this chapter offers researchers, medical professionals, and patients a solid foundation for accurate and timely brain tumor diagnoses, thus contributing to the research community.
Efficient UAV-Based Forest Fire Detection Using CNN and YOLOv8 Integration
6th IEEE International Conference on Recent Advances in Information Technology, RAIT 2025, 2025, DOI Link
View abstract ⏷
Wild forest fires pose a significant threat to ecosystems, causing habitat destruction, biodiversity loss, and severe air pollution. The rapid spread of uncontrolled fires leads to irreversible environmental degradation and increased greenhouse gas emissions. Timely and accurate fire detection is therefore essential to minimize risks to both human life and economic assets. This paper presents a UAV-based wildfire detection framework that integrates Convolutional Neural Networks (CNNs) for image classification with YOLOv8 for real-time object localization. The system was trained on 10,000 UAV-acquired images and achieved a classification accuracy of 92.6%, with an F1-score of 0.93, precision of 0.95, and recall of 0.92. The YOLOv8 model demonstrated strong detection capabilities, achieving a mean Average Precision (mAP) of 44.5% and an image processing time of 0.75 seconds per frame. Comparative evaluations reveal a 15% improvement in accuracy and a 20% reduction in computational cost over traditional methods. The proposed framework also exhibits robustness to noise and compression artifacts, attaining a Normalized Cross-Correlation (NCC) of 0.98 under Gaussian noise and a Peak Signal-to-Noise Ratio (PSNR) of 38.4 dB under compression. These results highlight the system's potential for effective real-time UAV-based wildfire surveillance, delivering reliable detection and rapid response in challenging environments.
Lightweight 2D Chaos-Based Encryption Scheme for Secure Image Communication in UAV Applications
Bhargav R., Singh P., Vardhan P.V.
International Conference on Communication Systems and Networks, COMSNETS, 2025, DOI Link
View abstract ⏷
The rapid increase in Unmanned Aerial Vehicles (UAVs) usage in various sectors has highlighted the need for secure image transmission. UAVs act as image-capturing platforms, transmitting data to ground stations and other UAVs over various networks, making the images vulnerable to potential attacks. Secure transmission of these images is crucial to prevent unauthorized access, ensure data integrity, and maintain the confidentiality of sensitive information. To address these challenges, this paper presents a 2D chaos-based encryption method designed to balance security and computational efficiency. The proposed method utilizes a 2D logistic map, reducing computational load while maintaining strong encryption, making it suitable for UAVs with limited resources. The scheme demonstrates high resistance to differential attacks, achieving NPCR and UACI values of 99.6973% and 32.4387%, respectively. Additionally, the high entropy value of 7.97 confirms enhanced randomness and security. The encryption method is evaluated under various attacks, including filtering, compression, and noise, demonstrating robust protection for transmitted images. A comparative analysis was also conducted, which supports the effectiveness of the proposed scheme in terms of reduced computational time and ensure encryption strength.
Automatic Vehicle Number Plate Recognition: A Narrative review of End-to-end Workflow
Raj A., Kumar Thakkar H., Singh P.
2025 International Conference on Pervasive Computational Technologies, ICPCT 2025, 2025, DOI Link
View abstract ⏷
In the current era, technology helps a lot to enhance human experiences leading to smart city development, which includes optimized energy usage, better healthcare availability, minimal governance, and intelligent mobility. However, considering the increase in day-to-day traffic, intelligent mobility is crucial, and its effective implementation is ensures flawless vehicle movements. The primary goal of is to design and develop the Intelligent Transportation Systems (ITS). The effective ITS ensures automatic vehicle movement management, and automatic traffic rule violation detection, which heavily relies on Automatic Vehicle Number Plate Recognition (AVNPR). In this paper, we described end-to-end workflow of AVNPR including the technologies used in each stage, the benefits and drawbacks of each technology followed by the limitations. The proposed paper exclusively reviews the modern, state-of-the-art models and architectures, and pioneered works carried out within the research area of AVNPR and the recent advancements.
Enabling secure image transmission in unmanned aerial vehicle using digital image watermarking with H-Grey optimization
Devi K.J., Singh P., Bilal M., Nayyar A.
Expert Systems with Applications, 2024, DOI Link
View abstract ⏷
Drone technology, also known as Unmanned Aerial Vehicles (UAVs), has advanced rapidly in the last decade owing to the huge number of users. This approach has an immense opportunity in areas like healthcare, agriculture, and forestry. However, the transmission of digital images via drone technology is associated with vulnerabilities and risks, due to a lack of efficient security solutions. Digital Image Watermarking (DIW) has emerged as a viable solution for digital image transmission in UAV applications. However, achieving robustness, security, and imperceptibility in a watermarking system at the same time is a difficult task. To address the aforementioned concerns, this paper proposes a secure digital image watermarking scheme in the hybrid DWT-SVD domain. The scaling factor is chosen adaptively from the hybrid Harmony-Grey Wolf (H-Grey) optimization algorithm to maintain a balance between imperceptibility and robustness. In addition, a novel symmetric cryptographic-based four-level encryption approach PSMD (Partitioning, Substituting, Merging, Division) is proposed to address UAV image security concerns by using Fibonacci and prime number series to generate two random keys. The proposed encryption scheme provides high security at low computational cost. The experimental results show that the visual quality and robustness are both high (Avg. PSNR = 41.27 dB and Avg. SSIM = 0.99, NC = 0.99, and BER = 0.0008 calculated for 100 images). The subjective and objective experimental analysis indicates that the proposed encryption scheme is highly secure and the computational cost is also low. The average embedding and extraction time is 0.25 s and 0.09 s It is resistant to various image processing attacks. A comparison with some of the most recent popular schemes confirms the scheme's effectiveness. The presented DIW can be used for copyright protection, UAV image transmission applications, military applications, and other purposes.
A framework of hybrid metaheuristic H-Grey optimization for embedding factor decision-making in digital image watermarking on social media
Optimization and Computing using Intelligent Data-Driven Approaches for Decision-Making: Optimization Applications, 2024, DOI Link
View abstract ⏷
Nowadays, social media platforms are a great way to connect and communicate with people all around the world. Digital images are the primary media of communication. Transmission of digital images over social media applications comes across various security issues like unauthorized accessing, owner identity theft, tarnishing metadata, and so on. To address all these stated issues, in this chapter, a secured digital image watermarking scheme is proposed in the hybrid redundant discrete wavelet transform (RDWT) - singular value decomposition (SVD) domain to provide high visual quality and robustness of the images. Also, embedding strength parameter selection decision is made using a hybrid metaheuristic Harmony-Greywolf (H-Grey) optimization approach to balance trade-off watermarking characteristics. To ensure higher security, embedding positions can be selected using a cross-layer approach in RDWT-SVD domain, and before embedding, watermark is encrypted using a Latin square sequence. Experiments are conducted on the proposed scheme in terms of watermarking characteristics like imperceptibility, robustness, and security. The outcomes of the experimental results show higher performance.
DOMINER: Domain Feature Mining from Unstructured Data for Effective Text Summarization
Thakkar H.K., Singh P., Kumar Y.
Procedia Computer Science, 2024, DOI Link
View abstract ⏷
In the current era of internet, a huge volume of opinionated sentiment reviews is generated on a daily basis. Normally, such reviews are enriched with deep sentiments and express the opinions on various features of a product or service under consideration. Identification of a set of important domain features opens an opportunity to build automatic product and service summarization system. It is observed that the common phenomena is to consider the frequently appearing words as the domainb features. However, not all frequently appearing words can be domain-specific features. In this paper, a novel Domain Feature Miner (DOMINER) approach is proposed for robust extractive summarization. The entire domain feature set mining problem is modelled as a clustering problem. Firstly, the bond energy based clustering technique is employed to cluster the domain features based on their frequency and co-appearance counts. Later, relevant clusters are extracted for the final set of domain feature set retrieval. The Proposed DOMINER scheme is extensively evaluated against quantitative performance metrics such as precision, recall, and F-score for six diversified domains such as Cellphone, Camera, Laptop, Tablet, Television, and Hotel. Experimental results on benchmark data sets reveal that proposed DOMINER scheme mines high quality of domain feature from unstructured reviews evident from precision, recall, and F-score values as 78.10%, 64.21%, and 70.48%, respectively against state of the art existing schemes. The high quality domain features extraction from DOMINER help generate the robust extractive summaries for product and services.
Integration of statistical parameters-based colour-texture descriptors for radar remote sensing image retrieval applications
Varish N., Reddy S.R., Sashi Varma N.G., Singh P.
International Journal of Computational Science and Engineering, 2024, DOI Link
View abstract ⏷
In this paper, a novel image retrieval method based on colour-texture contents for radar remote sensing applications is proposed, where global properties-based colour contents are extracted from different numbers of groups of histograms of colour image planes, and local properties-based texture contents have been derived from block level GLCM of an image plane. The integration of colour-texture contents represents the low dimensional feature which reduces overall computational overhead and increases the retrieval speed. To give importance to the feature components, suitable weights are imposed to both colour-texture contents appropriately. The obtained feature information describes the radar image effectively and also similarity measures play a significant role for better performance. This work compares eight similarity metrics to select the best one in the retrieval process. To validate the suggested method, experiments on two image datasets are performed and good retrieval results have been attained with rich colour-texture contents.
Radar Remote Sensing Image Retrieval Method Using Fusion of Handcrafted and Deep Features
Varish N., Sujana D.L., Akhil V., Greeshma K., Singh P.
Lecture Notes in Networks and Systems, 2024, DOI Link
View abstract ⏷
In this paper, a novel feature extraction method based on deep and handcrafted features is proposed for radar remote sensing image retrieval. The main purpose is to capture the low-level and high-level image characteristics to enhance the representation of visual patterns. Initially, we applied a CNN to the RGB color image, which extracts high-level features and creates a feature descriptor known as FVCNN with a dimension of 128. We further divide the RGB image into its red (R), green (G), and blue (B) components to compute handcrafted features. Then, to find patterns within each component, we use sparse local ternary pattern (LTP) operators in vertical, horizontal, and diagonal directions. The LTP-based features are then combined to create an additional feature descriptor known as the ternary feature descriptor (FVT). The high dimension of FVT is reduced by Principal Component Analysis (PCA) to the top 128 features. The final feature descriptor (FVFinal), with a dimension of 256, is created by combining the feature descriptors FVCNN and FVT respectively. In order to capture a wider range of visual characteristics, this feature fusion aims to take advantage of the complementary strengths of both CNN and handcrafted-based features. To choose the most effective metric for the retrieval process, this paper evaluates seven similarity metrics including Bray–Curtis, Canberra, Chebyshev, City block, Correlation, Cosine, and Euclidean. The proposed method is validated by trials on the UCM dataset, which produced satisfactory retrieval outcomes.
Mask Wearing Detection System for Epidemic Control Based on STM32
Luoli, Yadav A., Khan A., Varish N., Singh P., Thakkar H.K.
Lecture Notes in Networks and Systems, 2024, DOI Link
View abstract ⏷
This paper designs an epidemic prevention and control mask wearing detection system based on STM32, which is used to monitor the situation of people wearing masks. Tiny-YOLO detection algorithm is adopted in the system, combined with image recognition technology, and two kinds of image data with and without masks are used for network training. Then, the trained model can be used to carry out real-time automatic supervision on the wearing of masks in the surveillance video. When the wrong wearing or not wearing masks are detected, the buzzer will send an alarm, so as to effectively monitor the wearing of masks and remind relevant personnel to wear masks correctly.
Advancing Digital Image Forensics: Enhancing Image Forgery Detection Through Error Level Analysis and Convolutional Neural Networks
Mohammed K.B., Agrawal I., Kandimalla M.D., Govathoti P.F., Prakash C.S., Singh P.
Lecture Notes in Networks and Systems, 2024, DOI Link
View abstract ⏷
Digital image forgery detection is crucial in image forensics, aiming to identify manipulated regions and preserve visual integrity. Our framework combines Error Level Analysis (ELA) with prominent Convolutional Neural Network (CNN) architectures (VGG-16, VGG-19, ResNet-50, and Xception) to detect forgeries. ELA exploits error-level inconsistencies from manipulation, while CNN architectures extract features. We compare ELA with patch-level techniques, demonstrating its superior accuracy in capturing subtle artifacts. Experiments on the CASIA1 dataset evaluate the framework using metrics such as loss, accuracy, recall, precision, F1-score, and computational time. Results confirm the framework’s effectiveness in accurately detecting forgeries. Computational time analysis highlights its efficiency for real-world applications. In conclusion, our research presents a comprehensive framework using ELA and CNN architectures, showcasing ELA’s superiority and the potential of integrating it with CNNs for efficient forgery detection. This work advances image forensics, benefiting researchers and practitioners.
Coronavirus Herd Immunity Optimization-Based Control of DC-DC Boost Converter
Lecture Notes in Networks and Systems, 2023, DOI Link
View abstract ⏷
This paper presents a novel coronavirus herd immunity optimization (CHIO) algorithm for tuning the proportional-integral-derivative (PID) controller for the DC-DC boost converter. The closed-loop control action using the PID controller is designed to regulate the output voltage of DC-DC boost converter across the load end. CHIO is a nature-inspired meta-heuristic optimization algorithm formulated based on the way humankind handled the coronavirus pandemic (COVID-19) in recent years. This optimization algorithm exploits the herd immunity and social distancing concepts. The optimization algorithm has been developed on MATLAB/Simulink software for obtaining the optimum PID controller gains. Extensive simulations are conducted under (i) start-up response, (ii) reference voltage change (iii) load resistance change, and (iv) input voltage change to find the performance of the proposed controller. The obtained results indicate a successful convergence and satisfactory dynamic response of the output voltage under wide variation in the operating points.
Robust and Secure Medical Image Watermarking for Edge-Enabled e-Healthcare
Singh P., Devi K.J., Thakkar H.K., Bilal M., Nayyar A., Kwak D.
IEEE Access, 2023, DOI Link
View abstract ⏷
Advancements in networking technologies have enabled doctors to remotely diagnose and monitor patients using the Internet of Medical Things (IoMT), telemedicine, and edge-enabled healthcare. In e-healthcare, medical reports and patient records are typically outsourced to a server, which can make them vulnerable to unauthorized access and tampering. Therefore, it is crucial to ensure the authorization, security, confidentiality, and integrity of medical data. To address these challenges, this paper proposes a novel reversible watermarking approach with a high payload and low computational cost. First, the input medical image is divided into a Border region (BR) and a Non-Border region (NBR). The NBR region is upscaled using Neighbour Mean Interpolation (NMI) to ensure reversibility. The Electronic Patient Record (EPR) is encrypted using a pseudorandom key, which is generated adaptively from the host medical image and the Enigma machine. The encrypted EPR is then embedded in the medical image using NMI. Two levels of tamper detection (global and local) are performed at the receiver's end for higher accuracy. A Global Integrity Code is generated and embedded in BR using LSB embedding technique for global tamper detection. The experimental results show that the visual quality and robustness are both high (Avg. PSNR = 41.03 dB and Avg. SSIM = 0.99, NC = 0.99, and BER = 0.0019 calculated for 100 images). The subjective and objective experimental analysis indicates that the proposed scheme is highly secure and the computational cost is also low. The average embedding and extraction time (including embedding, encryption and decryption, extraction process respectively) is 0.88 s and 0.83 s. It is resistant to various image processing attacks. A comparison with some of the most recent popular schemes confirms the scheme's effectiveness.
A Comparative Analysis of Ten Classifiers and their Impact on Early Detection of Stroke Prediction
Dinkar S., Nikhil Y., Bhargav B., Gopichand G., Singh P.
IEEE Region 10 Humanitarian Technology Conference, R10-HTC, 2023, DOI Link
View abstract ⏷
Stroke represents a significant global health challenge, leading to substantial disability and mortality across the world. Countries in the developing regions, including India, bear a substantial burden of stroke cases, with the most prevalent type being ischemic stroke. The ability to predict the likelihood of stroke is of utmost importance for effective prevention and early intervention. Therefore, this conference paper aims to assess and compare the performance of various machine learning algorithms in predicting stroke. The study employs a dataset containing diverse input variables, such as age, blood pressure, diabetes status, and smoking habits. To achieve this, ten machine learning classifiers, namely Logistic Regression, K-Nearest Neighbors, Support Vector Machines, Gaussian Naive Bayes, Decision Tree, Random Forest, XGBoost, Stochastic Gradient Descent, and AdaBoost are implemented and evaluated for their predictive capabilities. The outcomes of this research offer valuable insights into the effectiveness of each algorithm, serving as a valuable reference for future investigations in the field of stroke prediction.
Security Issues in Deep Learning
Patel S., Bakaraniya P.V., Mishra S., Singh P.
Studies in Computational Intelligence, 2023, DOI Link
View abstract ⏷
Deep learning has created substantial improvements for industries and set the tempo for a destiny constructed on artificial intelligence (AI) technology. Nowadays, deep learning is turning into an increasing number of vital in our everyday lifestyles. The appearance of deep learning in many applications in life relates to prediction and classification such as self-driving, product recommendation, classified ads and healthcare. Therefore, if a deep learning model causes false predictions and misclassification, it may do notable harm. This is largely a critical difficulty inside the deep learning model. In addition, deep learning models use big quantities of facts inside the training/learning phases, which in corporate touchy facts can motivate misprediction on the way to compromise its integrity and efficiency. Therefore, while deep learning models are utilized in real-world programs, it's mile required to guard the privateness facts used inside the model. The countless opportunities and technological abilities that system learning has added to the arena have concurrently created new safety dangers that threaten development and organizational development. Understanding system learning safety dangers is one of every of our contemporary technological time's maximum vital undertakings due to the fact the results are extraordinarily high, mainly for industries along with healthcare in which lives are at the line. We talk about the forms of system mastering safety dangers that you may stumble upon so you may be higher organized to stand them head-on.
Secure transmission of medical images in multi-cloud e-healthcare applications using data hiding scheme
Devi K.J., Singh P., Dash J.K., Thakkar H.K., Tanwar S., Alabdulatif A.
Journal of Information Security and Applications, 2023, DOI Link
View abstract ⏷
In recent years, medical image transmission using a multi-cloud system has played a significant role in e-Healthcare infrastructure. It allows medical practitioners to easily store, retrieve, and share patients’ medical information across multiple stakeholders. However, multi-cloud image transmission may be vulnerable to multiple security breaches, such as authentication, confidentiality, and security issues. Motivated by these issues, this paper proposes a data-hiding scheme for secure medical image transmission in a multi-cloud environment. The proposed scheme ensures imperceptible robustness and watermark security at a low computational cost. Here, the medical image is divided into a number of shares using Neighbor Mean Interpolation (NMI). To achieve confidentiality, Electronic Patient Healthcare Record (EPHR) is encrypted using Double Scan Pixel Position Shuffling (DSPPS) approach. Then, the encrypted EPHR is divided into shares and embedded in the cover medical image shares. Finally, a minimum of 50% of watermarked image shares are utilized to retrieve the original medical image and encrypted EPHR, consequently reducing multi-cloud latency and computational burden. Experimental results show that the proposed scheme shows high imperceptibility, robustness, and watermark security at a low computational cost. Comparative analysis with some of the recent popular data hiding schemes shows that the proposed scheme has improved imperceptibility and robustness by 10%–15% (approximately) with higher watermark security at a low computational cost.
Robust and Secured Reversible Data Hiding Approach for Medical Image Transmission over Smart Healthcare Environment
Jyothsna Devi K., Singh P., Santamaria J., Patel S.
Studies in Computational Intelligence, 2023, DOI Link
View abstract ⏷
With the rapid progress of cloud computing, there has been a marked improvement in the development of smart healthcare applications such as Internet of Medical Things (IoMT), Telemedicine, etc. Cloud-based healthcare systems can efficiently store and communicate patient electronic healthcare records (EHR) while allowing for quick growth and flexibility. Despite the potential benefits, identity violation, copyright infringement, illegal re-distribution, and unauthorized access have all been significant. To address all these breaches, in this paper, a reversible medical image watermarking scheme using interpolation is proposed. The medical image is partitioned into Border Region (BR), Region of Interest (ROI), and Region of Non-interest (RONI) regions. BR is used for embedding integrity checksum code generated from ROI for tamper detection. RONI is used for embedding watermark. To ensure complete recovery of ROI and high embedding capacity, ROI is compressed before embedding. To ensure high-security compressed ROI, hospital emblem and EHR merged and then encrypted using a random key generated from Polybius magic square to get higher security. The proposed scheme is proved to take less computational time as there are no complex functions used in the embedding. The experiments performed on the proposed scheme is proved to have high imperceptibility, robustness, embedding capacity, security, and less computational time. All these confirm that the proposed approach is a potential candidate for suitable in smart healthcare environment.
Application of Distributed Back Propagation Neural Network for Dynamic Real-Time Bidding
Desai A., Thakkar H.K., Singh P., Bhargavi L.S.
Communications in Computer and Information Science, 2022, DOI Link
View abstract ⏷
Programmatic buying popularly known as real-time bidding (RTB) is a key ascendancy in online advertising. The Ad Tech industry is experiencing sustained growth, especially due to the increased use of mobile devices. While data has become essential for targeting and ad performance, data businesses have become difficult to differentiate due to their proliferation, as well as limitations of attribution. This provides an opportunity for Big Data practitioners to leverage this data and use machine learning to improve efficiency and make more profits. Taking such an opportunity we came up with an application of a machine learning algorithm, distributed back propagation neural network (d-bpnn) to predict bid prices in a real-time bidding system. This paper depicts how d-bpnn is used to achieve less eCPM (effective Cost Per Mille) for advertisers while preserving win rate and budget utilization.
Exploiting deep and hand-crafted features for texture image retrieval using class membership
Pattern Recognition Letters, 2022, DOI Link
View abstract ⏷
In the modern digital era, with the availability of low-cost hardware like sensors and cameras, a huge amount of image databases are being created for diverse applications. These databases give rise to the need of developing efficient content-based image retrieval (CBIR) systems. Major efforts have been put over the past two decades to develop different global and low-level texture features to build efficient CBIR systems. However, designing texture features that are suitable for distance-based retrieval is always a challenging task. Recently, Convolution Neural Networks have shown promising results for object detection and classification. CNNs are also applied to build classifier-based retrieval systems. However, the classifier-based retrieval methods can retrieve images only from the predicted class. Therefore, the performance of such system greatly depends on classification performance of the classifier. This paper proposes a method that exploits the strength of the Convolutional Neural Networks for predicting the class membership of the query image for all output classes and retrieve images using a modified distance function in the wavelet feature space. The performance of the proposed method is evaluated using three popular texture datasets of varying complexity and found to be superior to all competing methods considered.
PRMS: Design and Development of Patients’ E-Healthcare Records Management System for Privacy Preservation in Third Party Cloud Platforms
Zala K., Thakkar H.K., Jadeja R., Singh P., Kotecha K., Shukla M.
IEEE Access, 2022, DOI Link
View abstract ⏷
In the current digital era, personal data storage on public platforms is a major cause of concern with severe security and privacy ramifications. This is true especially in e-health data management since patient's health data must be managed following a slew of established standards. The Cloud Service Providers (CSPs) primarily provide computing and storage resources. However, data security in the cloud is still a major concern. In several instances, Blockchain technology rescues the CSPs by providing the robust security to the underlying data by encrypting data using the unique and secret keys. Each network user in Blockchain has its own unique and secret keys linked directly to the transaction keys as a digital signature to protect the data. However, Blockchain technology suffers from the latency and throughput issues in high workload scenarios. To overcome e-healthcare records privacy issues in a third-party cloud, we designed a Patient's E-Healthcare Records Management System (PRMS) that focuses on latency and throughput. A comprehensive performance analysis of PRMS is carried out on different third-party clouds to validate its applicability. Moreover, the proposed PRMS system is compared with Blockchain platforms such as Hyperledger Fabric v0.6 and Etherium 1.5.8 against latency and throughput by adjusting the workload for each platform up to 10,000 transactions per second. The proposed PRMS is compared to the Secure and Robust Healthcare-Based Blockchain (SRHB) approach using Yahoo Cloud Serving Benchmark (YCSB) and small bank datasets. The experimental results indicate that deploying PRMS on Amazon Web Services decreases System Execution Time (SET) and the Average Delay (AD) time by 2.4%, 8.33%, and 25.15%, 15.26%, respectively. Additionally, deploying PRMS on the Google Cloud Platform decreases System Execution Time (SET) and Average Delay (AD) by 2.27%, 2.4%, and 2.72%, 4.73% AD, respectively. The experimental results confirm the superiority of the PRMS under the high workload scenario over SRHB and its applicability in cloud data centers.
Performance Analysis of Image Retrieval Method Using Quantized Bins of Color Histogram
Varish N., Singh P., Yaser S., Surapaneni A., Reddy B.V.
Lecture Notes in Networks and Systems, 2022, DOI Link
View abstract ⏷
Direct histogram to histogram matching in content-based image retrieval is not proficient due to its large number of bins. The total number of bins of an original histogram represents the large dimensional feature descriptor which requires high computational overhead during the retrieval process. To address this issue, in the proposed scheme image histogram is quantized into a different number of bins which represents the low dimensional feature descriptor effectively. Since, the global and local features play an important role in image retrieval, therefore, considering any single feature for image retrieval is not adequate, so in this paper, a quantized histogram-based global and local features have been considered for feature representation. To avoid variations among the feature components, suitable weights are assigned to the local and global features effectively. To check the efficacy of the proposed method, performance analysis using a different number of bins has been evaluated based on two standard similarity distances for corel-1 K image dataset. The presented work has achieved satisfactory retrieval results in terms of precision, recall, and F-score metrics.
Robust, Reversible Medical Image Watermarking for Transmission of Medical Images over Cloud in Smart IoT Healthcare
Jyothsna Devi K., Jayanth Krishna M.V., Singh P., Santamaria J., Bakaraniya P.
Predictive Analytics in Cloud, Fog, and Edge Computing: Perspectives and Practices of Blockchain, IoT, and 5G, 2022, DOI Link
View abstract ⏷
There has been a significant advancements in the field of e-healthcare. E-healthcare applications are changing and adapting to improve smart healthcare domain employing IoT devices, making working in the healthcare domain more convenient. For diagnosis and treatment, medical images are transferred through cloud from radiology center to smart healthcare centres. Among the contentions that causes problems in medical image transmission is the integrity and confidentiality of medical images on the smart healthcare platform. Another major challenge with ensuring protection for electronic patient health records (EHR) in the cloud infrastructure is the confidentiality of patient records. Hence medical image watermarking helps to maintain their authenticity, confidentiality and integrity. Here proposing a reversible medical image watermarking scheme in the spatial domain that embedded EHR in the medical cover image at the sender and the same watermark is retrieved and original image is restored at the receiver end. The embedding is done using the quadratic difference expansion and the data encryption is done using the magic polybius music square to ensure high embedding capacity and security. From the obtained experimental results satisfied all watermarking characteristics.
Clairvoyant: AdaBoost with Cost-Enabled Cost-Sensitive Classifier for Customer Churn Prediction
Thakkar H.K., Desai A., Ghosh S., Singh P., Sharma G.
Computational Intelligence and Neuroscience, 2022, DOI Link
View abstract ⏷
Customer churn prediction is one of the challenging problems and paramount concerns for telecommunication industries. With the increasing number of mobile operators, users can switch from one mobile operator to another if they are unsatisfied with the service. Marketing literature states that it costs 5-10 times more to acquire a new customer than retain an existing one. Hence, effective customer churn management has become a crucial demand for mobile communication operators. Researchers have proposed several classifiers and boosting methods to control customer churn rate, including deep learning (DL) algorithms. However, conventional classification algorithms follow an error-based framework that focuses on improving the classifier's accuracy over cost sensitization. Typical classification algorithms treat misclassification errors equally, which is not applicable in practice. On the contrary, DL algorithms are computationally expensive as well as time-consuming. In this paper, a novel class-dependent cost-sensitive boosting algorithm called AdaBoostWithCost is proposed to reduce the churn cost. This study demonstrates the empirical evaluation of the proposed AdaBoostWithCost algorithm, which consistently outperforms the discrete AdaBoost algorithm concerning telecom churn prediction. The key focus of the AdaBoostWithCost classifier is to reduce false-negative error and the misclassification cost more significantly than the AdaBoost.
Robust and secured watermarking using Ja-Fi optimization for digital image transmission in social media
Jyothsna Devi K., Singh P., Thakkar H.K., Kumar N.
Applied Soft Computing, 2022, DOI Link
View abstract ⏷
Widespread transmission of digital image in social media has come up with security, confidentiality and authentication issues. Ensuring copyright protection of digital images shared through social media has become inevitable. To address these issues, a robust and secure digital image watermarking scheme using Redundant discrete wavelet transform (RDWT) - Singular value decomposition (SVD) hybrid transform is proposed in this paper. In the proposed scheme, digital image is divided into 4 × 4 non-overlapping blocks, and low information blocks are selected for embedding to ensure higher imperceptibility. For watermark embedding 1-level RDWT is applied on the selected blocks followed by SVD decomposition to make the proposed scheme highly robust against common attacks. One watermark bit is embedded in each left and right singular SVD matrices by adjusting the coefficients. This makes the proposed scheme free from false positive error and achieve high embedding capacity. Before embedding, watermark encryption is done by using a pseudo random key. The pseudo random key is generated adaptively from the cover image by using discrete wavelet transform saliency map, block mean approach and cosine functions. High imperceptibility and robustness is indispensable for the digital images shared through social media. But, these watermarking characteristics are in trade-off. In the proposed scheme, the trade-off is balanced by using optimized scaling factor (embedding strength). Scaling factor is optimized by using the proposed JAYA-Firefly (Ja-Fi) optimization. Experimental results demonstrate that the proposed scheme provides high imperceptibility, robustness, embedding capacity and security. Furthermore, performance comparison with the recent state-of-the-art schemes affirms that the proposed scheme has superior performance.
Region-Based Hybrid Medical Image Watermarking Scheme for Robust and Secured Transmission in IoMT
Singh P., Devi K.J., Thakkar H.K., Kotecha K.
IEEE Access, 2022, DOI Link
View abstract ⏷
With the growth in Internet and digital technology, Internet of Medical Things (IoMT) and Telemedicine have become buzzwords in healthcare. A large number of medical images and information is shared through a public network in these applications. This paper proposes a region-based hybrid Medical Image Watermarking (MIW) scheme to ensure the authenticity, authorization, integrity, and confidentiality of the medical images transmitted through a public network in IoMT. In the proposed scheme, medical images are partitioned into Region of Interest (RoI) and Region of Non-Interest (RoNI). To ascertain integrity of RoI, tamper detection and recovery bits are embedded in RoI in the medical image. RoI is watermarked using adaptive Least Significant Bit (LSB) substitution with respect to the hiding capacity map for higher RoI imperceptibility and accuracy in tamper detection and recovery. Electronic Patient Record (EPR) is compressed using Huffman coding and encrypted using a pseudo-random key (secret key) to provide higher confidentiality and payload. QR code of hospital logo, Encrypted EPR, and RoI recovery bits are interleaved in RoNI using Discrete Wavelet Transform-Singular Value Decomposition (DWT-SVD) hybrid transforms to achieve a robust watermark. The proposed scheme is tested under various geometric and non-geometric attacks such as filtering, compression, rotation, salt and pepper noise and shearing. The evaluation results demonstrate that the proposed scheme has high imperceptibility, robustness, security, payload, tamper detection, and recovery accuracy under image processing attacks. Therefore, the proposed scheme can be used in the transmission of medical images and EPR in IoMT. Relevance of the proposed scheme is established by its superior performance in comparison to some of the popular existing schemes.
ReLearner: A Reinforcement Learning-Based Self Driving Car Model Using Gym Environment
Thakkar H.K., Desai A., Singh P., Samhitha K.
Communications in Computer and Information Science, 2022, DOI Link
View abstract ⏷
In the recent past, Artificial intelligence and its sister technology such as Machine Learning, Deep Learning, and Reinforcement learning have grown rapidly in several applications. The self-driving car is one of the applications, which is the need of the hour. In this paper, we describe the trends in autonomous vehicle technology for the self-driving car. There are many different approaches to mathematically formulate a design for the self-driving car such as deep Q-learning, Q-learning, and machine learning. However, in this paper, we propose a very basic and less compute-intensive simplistic self-driving car model called “ReLearner” using the Gym environment. To simulate the self-driving car model, we preferred to create a simple small environment OpenAi gym which is a deterministic environment. The OpenAi gym provides the virtual simulation environment and parameter tuning to train and test the model. We have focused on two methods to test our model. The basic approach is to compare the performance of the car when tested using Q-Learning and another using a random action agent, i.e., No reinforcement learning. We have derived a theoretical model and analyzed how to use Q-learning to train cars to drive. We have carried out a simulation and on evaluating the performance and found that Q-learning is a more optimal approach to solve the issue of a self-driving car.
A Decisive Metaheuristic Attribute Selector Enabled Combined Unsupervised-Supervised Model for Chronic Disease Risk Assessment
Mishra S., Thakkar H.K., Singh P., Sharma G.
Computational Intelligence and Neuroscience, 2022, DOI Link
View abstract ⏷
Advanced predictive analytics coupled with an effective attribute selection method plays a pivotal role in the precise assessment of chronic disorder risks in patients. Traditional attribute selection approaches suffer from premature convergence, high complexity, and computational cost. On the contrary, heuristic-based optimization to supervised methods minimizes the computational cost by eliminating outlier attributes. In this study, a novel buffer-enabled heuristic, a memory-based metaheuristic attribute selection (MMAS) model, is proposed, which performs a local neighborhood search for optimizing chronic disorders data. It is further filtered with unsupervised K-means clustering to remove outliers. The resultant data are input to the Naive Bayes classifier to determine chronic disease risks' presence. Heart disease, breast cancer, diabetes, and hepatitis are the datasets used in the research. Upon implementation of the model, a mean accuracy of 94.5% using MMAS was recorded and it dropped to 93.5% if clustering was not used. The average precision, recall, and F-score metric computed were 96.05%, 94.07%, and 95.06%, respectively. The model also has a least latency of 0.8 sec. Thus, it is demonstrated that chronic disease diagnosis can be significantly improved by heuristic-based attribute selection coupled with clustering followed by classification. It can be used to develop a decision support system to assist medical experts in the effective analysis of chronic diseases in a cost-effective manner.
Time Series Analysis of COVID-19 Waves in India for Social Good
Nallam L.S.D., Sankati S., Thakkar H.K., Singh P.
Studies in Computational Intelligence, 2022, DOI Link
View abstract ⏷
In the past decade, the world has seen rapid advancements in the field of healthcare services due to the state of the arts in technologies. Several real-time health monitoring applications and products are designed to assist the human to take the timely precautionary measures to avoid the unseen abnormalities. However, current healthcare monitoring infrastructures are not ready to provide efficient health services during the sudden and unknown pandemic situations such as COVID-19. The COVID-19 started in the later part of 2019, rapidly spread across the countries and labeled as a pandemic in the very early part of the 2020. Several people died due to the lack of the healthcare infrastructure and lack of access to health facilities. This book chapter explores the various technologies such as augmented reality, connected e-health along with the time series analysis of COVID-19 waves in India to know the implication of COVID-19 on society for a social good.
5G Enabled Smart City Using Cloud Environment
Bakaraniya P., Patel S., Singh P.
Predictive Analytics in Cloud, Fog, and Edge Computing: Perspectives and Practices of Blockchain, IoT, and 5G, 2022, DOI Link
View abstract ⏷
In the recent past, cloud supported applications are dominating the market due to the flexibility, affordability, conveniency, and its ubiquitous access. Past few years have witnessed significant focus on enabling the cities with Internet of things supported devices for efficient monitoring and management. The recent advancement in the wireless communications such as 4G have contributed a lot and there is a growing expectation in wireless communications with 5G technology. In this book chapter, 5G enabled smart city framework, possible challenges, and future work is described. The chapter also focuses on how 5G technology can be integrated with cloud environment for overall development of smart city application.
Dual Secured Reversible Medical Image Watermarking for Internet of Medical Things
Devi K.J., Singh P., Thakkar H.K.
Studies in Computational Intelligence, 2022, DOI Link
View abstract ⏷
The Internet of Medical Things (IoMT) plays a big role in today’s healthcare industry. Smart healthcare has been a great use in sharing all kinds of medical data from Electronic Health Records (EHR) to hospital management. This advancement in healthcare made the diagnosis easy for the doctors and patients by saving their time. But in recent years there are a lot of threats in IoT and IoMT frameworks which motivated many researchers to find a way to overcome these challenges. In this paper, reversible data hiding technique using a linear and quadratic difference expansion to embed the EHR on the image is proposed. The medical image is divided into Border Region (BR) and Non Border Region (NBR). The hospital logo is watermarking in the BR of the image. Using LSB (Least Significant Bit) method hospital logo is embedded to ensure autauthentication. LSB approach is an easy and fast way to perform. To ensure confidentiality NBR is used for EPR embedding. Selecting a pair of pixels in NBR for embedding 2 bits of EPR pixels at each time with the linear and quadratic difference expansion. For ensuring dual security, EPR is encrypted using a Pseudo random key. Then encrypted EPR is partitioned into Odd and Even position pixels parts. Further, Odd and Even position pixels are watermarking in NBR using linear and quadratic difference expansion. Pseudo random key is generated adaptively from the mean of Medical image and Divide and Conquer algorithm to provide higher security. This proposed method is proved to take less computational time as there are no complex functions used in the algorithm. The test performed on this technique is proved to have high imperceptibility, robustness, security, and less computational time. All this confirms that proposed approach is a potential candidate for the security of data in the IoMT frameworks.
A New Robust and Secure 3-Level Digital Image Watermarking Method Based on G-BAT Hybrid Optimization
Devi K.J., Singh P., Dash J.K., Thakkar H.K., Santamaria J., Krishna M.V.J., Romero-Manchado A.
Mathematics, 2022, DOI Link
View abstract ⏷
This contribution applies tools from the information theory and soft computing (SC) paradigms to the embedding and extraction of watermarks in aerial remote sensing (RS) images to protect copyright. By the time 5G came along, Internet usage had already grown exponentially. Regarding copyright protection, the most important responsibility of the digital image watermarking (DIW) approach is to provide authentication and security for digital content. In this paper, our main goal is to provide authentication and security to aerial RS images transmitted over the Internet by the proposal of a hybrid approach using both the redundant discrete wavelet transform (RDWT) and the singular value decomposition (SVD) schemes for DIW. Specifically, SC is adopted in this work for the numerical optimization of critical parameters. Moreover, 1-level RDWT and SVD are applied on digital cover image and singular matrices of LH and HL sub-bands are selected for watermark embedding. Further selected singular matrices (Formula presented.) and (Formula presented.) are split into (Formula presented.) non-overlapping blocks, and diagonal positions are used for watermark embedding. Three-level symmetric encryption with low computational cost is used to ensure higher watermark security. A hybrid grasshopper–BAT (G-BAT) SC-based optimization algorithm is also proposed in order to achieve high quality DIW outcomes, and a broad comparison against other methods in the state-of-the-art is provided. The experimental results have demonstrated that our proposal provides high levels of imperceptibility, robustness, embedding capacity and security when dealing with DIW of aerial RS images, even higher than the state-of-the-art methods.
Blind and secured adaptive digital image watermarking approach for high imperceptibility and robustness
Singh P., Devi K.J., Thakkar H.K., Santamaria J.
Entropy, 2021, DOI Link
View abstract ⏷
In the past decade, rapid development in digital communication has led to prevalent use of digital images. More importantly, confidentiality issues have also come up recently due to the increase in digital image transmission across the Internet. Therefore, it is necessary to provide high imperceptibility and security to digitally transmitted images. In this paper, a novel blind digital image watermarking scheme is introduced tackling secured transmission of digital images, which provides a higher quality regarding both imperceptibility and robustness parameters. A block based hybrid IWT-SVD transform is implemented for robust transmission of digital images. To ensure high watermark security, the watermark is encrypted using a Pseudo random key which is generated adaptively from cover and watermark images. An encrypted watermark is embedded in randomly selected low entropy blocks to increase the security as well as imperceptibility. Embedding positions within the block are identified adaptively using a Blum–Blum–Shub Pseudo random generator. To ensure higher visual quality, Initial Scaling Factor (ISF) is chosen adaptively from a cover image using image range characteristics. ISF can be optimized using Nature Inspired Optimization (NIO) techniques for higher imperceptibility and robustness. Specifically, the ISF parameter is optimized by using three well-known and novel NIO-based algorithms such as Genetic Algorithms (GA), Artificial Bee Colony (ABC), and Firefly Optimization algorithm. Experiments were conducted for the proposed scheme in terms of imperceptibility, robustness, security, embedding rate, and computational time. Experimental results support higher effectiveness of the proposed scheme. Furthermore, performance comparison has been done with some of the existing state-of-the-art schemes which substantiates the improved performance of the proposed scheme.
Image Retrieval Scheme Using Efficient Fusion of Color and Shape Moments
Lecture Notes in Networks and Systems, 2021, DOI Link
View abstract ⏷
Due to the tremendous increase in the digital image data, the efficient and effective image content-based search scheme for retrieving desired images from a large image repository is highly required. The biggest challenge in image retrieval scheme is to retrieve the desired multimedia images from the digital image repository with minimum time. Extracting significant image features with low dimensional feature descriptor play a significant role in improving retrieval outcomes. In the presented paper, an image retrieval scheme is proposed using fused low dimensional feature descriptor which is obtained by fusion of probability histogram-based HSV color moments and multiresolution based shape moments. The color moments and shape moments are extracted from the Laplacian filter based preprocessed image. The suggested scheme is implemented on a standard Corel-1K image dataset and the retrieval accuracy is measured using precision, recall, and F-score metrics. The experimental outcomes are also validated and compared with some existing state of the art image retrieval schemes and it outperforms over the existing ones.
Reversible and Secured Image Watermarking Technique for IoMT Healthcare
Devi K.J., Singh P., Yadav R.K., Gafaru M.Z.
2021 12th International Conference on Computing Communication and Networking Technologies, ICCCNT 2021, 2021, DOI Link
View abstract ⏷
In recent days, IoMT based medical treatment and diagnosis is so popular in health - care domain. The IoMT collects the medical data from patients on real time and sends over insecure channels. This leads to various authentic and security threats. To address these concerns, in this paper a region based reversible Medical Image Watermarking (MIW) approach is proposed in spatial domain to ensusing low real time computational complexity. First the cover image is divided into Border Region (BR) and Non Rorder Region (NBR). Binary hospital logo image is embedded in BR using LSB to achieve authentication. Encryted Electronic Patient Record (EPR) information is embedded in NBR to achieve confidentiality. A reversible quadratic difference expansion approach with low computational cost and high embedding capacity is proposed for EPR embedding. To achive high EPR security, a Pseudo Random key is used in encryption. A Pseudo random key is generated using Playfair cipher approach. A 16 × 16 Playfair is used for generation of Pseudo Random key. The obtained experimental results show the proposed scheme is superior as compared to previous works in terms of security, embedding capacity and have low computational time complexity. Thus the scheme can be used in real time healthcare industries for IoMT.
False-Positive-Free and Geometric Robust Digital Image Watermarking Method Based on IWT-DCT-SVD
Lecture Notes in Electrical Engineering, 2021, DOI Link
View abstract ⏷
This paper presents a new hybrid image watermarking method based on IWT, DCT, and SVD domains, to solve the problem of false-positive detection and scale down the impact of geometric attacks. Properties of IWT, DCT, and SVD enable in achieving higher imperceptibility and robustness. However, SVD-based watermarking method suffers from a major flaw of false-positive detection. Principal component of watermark is embedding in the cover image to overcome this problem. Attacker cannot extract watermark without the key (eigenvector) of the embedded watermark. To recover geometrical attacks, we use a synchronization technique based on corner detection of the image. Computer simulations show that the novel method has improved performance. A comparison with well-known schemes has been performed to show the leverage of the proposed method.
Secured Cross Layered Watermark Embedding for Digital Image Authentication Using IWT-SVD
2020 IEEE 15th International Conference on Industrial and Information Systems, ICIIS 2020 - Proceedings, 2020, DOI Link
View abstract ⏷
In this paper, we are proposing a novel Integer Wavelet Transform(IWT)-Singular Value Decomposition(SVD) blind digital image watermarking scheme using hybrid transform to achieve higher imperceptibility, robustness, security and authenticity. In the proposed scheme, pseudo-random Latin square sequence is used for watermark encryption, encrypted watermark is divided and cross implant technique is used for embedding to overcome the problem of False Positive Problem (FPP). Further watermarking strengthening parameter is optimized using nature-inspired Artificial Bee Colony(ABC) algorithm, Simulation results shows that the proposed scheme has higher imperceptibility and robustness with different image modalities(gray-scale and colored). Performance comparison with some popular schemes shows that the proposed scheme surpass them in terms of robustness, imperceptibility and confidentiality.
MEDICAL IMAGE WATERMARKING for AUTHENTICATION, CONFIDENTIALITY, TAMPER DETECTION and RECOVERY
2019 10th International Conference on Computing, Communication and Networking Technologies, ICCCNT 2019, 2019, DOI Link
View abstract ⏷
This paper presents a region based blind medical image watermarking (MIW) scheme for ensuring authenticity, integrity and confidentiality of medical images. Medical image is segmented into region of interest(ROI) and region of non interest (RONI). ROI is watermarked for tamper detection and recovery in the spatial domain. For providing confidentiality and authenticity, electronic patient record (EPR) and hospitals logo is embedded as a robust watermark in RONI using IWT-SVD hybrid transform. Various experiments were carried out on different medical imaging modalities for performance evaluation of the proposed scheme in terms of imperceptibility, robustness, tamper detection and recovery. Evaluation results show that the visual quality of watermarked image is good and it is robust under common attacks. A comparison with well known schemes has been performed to show superiority of the proposed method.
