A review on content-based image retrieval system: present trends and future challenges. (19th August 2021)

• Record Type: Journal Article
• Authors: Rout, Narendra Kumar; Atulkar, Mithilesh; Ahirwal, Mitul Kumar
• Is Part Of: International journal of computational vision and robotics. Volume 11:Number 5(2021)
• Page Start: 461
• Page End: 485
• Publication Date: 2021-08-19
• Subjects: content-based image retrieval -- CBIR -- low level features -- feature extraction methods -- weight assignment method -- image database; Computer vision -- Periodicals; Robotics -- Periodicals; Artificial intelligence -- Periodicals; 006.3705
• Languages: English
• ISSNs: 1752-9131

Abstract

The issues of getting similar images with better accuracy is now a challenge in content-based image retrieval (CBIR) system due to exponential rising volume of image databases. In CBIR, first of all image features are extracted. Importance of each low level feature is graded by their repute based on citations in various comparable studies. With this, different weight assignment methods for features like individual weightage, equal assignment of weights and other assignment methods employed in the CBIR systems have been reported. However, the weight assignment to the features of the image is calibrated manually based on its importance in doing accurate searches on particular databases. This paper presents a review on CBIR systems and frequently used features with different weight assignment methods. The future challenge identified from this study is, to make the CBIR system automated for assigning weights to image features. Solutions for reported challenge are also suggested.