Yuvaraj Tankala and Joseph K Paul, 5th-semester B Tech Computer Science and Engineering students of SRM University-AP, Andhra Pradesh, India has worked with Dr Manikandan V M, Assistant Professor in Computer Science and Engineering Department on a research project and their research paper “A Content-based Image Retrieval Scheme with Object Detection and Quantized Color Histogram” got accepted for publication in the International Journal of Computational Science and Engineering.

Content-based image retrieval (CBIR) is an active area of research due to its wide applications. Most of the existing CBIR schemes are concentrated to do the searching of the images based on the texture, colour, or shape features extracted from the query image. In this manuscript, we propose an object detection based CBIR scheme with quantized colour histograms. In the proposed scheme, the meaningful objects will be identified from the query image by using you only look once (YOLO) object detection techniques and the quantized histograms of each of the object categories. The object lists, their count, and the area covered by the objects along with quantized colour histograms will be used during feature matching to retrieve the related images from the large image pool. The experimental of the proposed scheme is carried on the Corel 1K and Caltech image dataset. We have observed an average precision of 0.96 during the experimental study which is quite high while comparing the precision from the well-known existing schemes.

To retrieve relevant images from a large image pool, we use content-based image retrieval (CBIR) schemes. In a CBIR scheme, the properties of the query image will be matched with the properties of the images in the image pool. The images which are very close to the given query image will be returned by the CBIR scheme. Most of the existing CBIR schemes use colour, shape and texture properties for image comparison. In the proposed scheme, we use an object detection-based approach with quantized colour histograms to retrieve the relevant images from the image pool.

The real-life applications of the proposed scheme are listed below:
● In the fashion designing and textile industry, CBIR systems can be used to find the existing designs.
● The CBIR systems are useful in crime prevention by retrieving similar crime scenes or the images of criminal persons based on the query image.
● Professional web designers or poster designers want to retrieve relevant images depends on the specific context that they are working.
● To retrieve similar medical images with the relevant treatment details in a computer-assisted diagnosis system.

The team currently continue their research work in the same domain to come up with a content-based image retrieval system that will return the relevant images by understanding the relationships among the objects in the image. The classes of the objects in the scene and their properties also will be considered along with the relationship between the objects in the scene.