Using Relevance Feedback in Bridging Semantic Gaps in Content-Based Image Retrieval

Abstract

Content-based image retrieval (CBIR) is a difficult area of research in multimedia systems. The research has proved extremely difficult because of the inherent problems in proper automated analysis and feature extraction of the image to facilitate proper classification of various objects. An image may contain more than one objects and to segment the image in line with object features to extract meaningful objects and then classify it in high-level like table, chair, car and so on has become a challenge to the researchers in the field. The latter part of the problem, the gap between low-level features like color, shape, texture, spatial relationships and highlevel definitions of the images is called the semantic gap. Until we solve these problems in an effective way, the efficient processing and retrieval of information from images will be difficult to achieve. In this paper we explore the possibilities of how relevance feedback can help us solve this problem of semantic gap although lot of works have already been done using the concepts of relevance feedback in this area. This would enable efficient image retrieval for internet of the future.

Narrowing Semantic Gap in Content-based Image Retrieval

“…Relevance feedback (RF) is an interactive supervised learning technique that has been proposed to bridge the semantic gap between the low-level image features used and the semantic content of the images and, thus, to improve the retrieval results [48][49][50][51][52][53]. In particular, RF attempts to insert the subjective human perception of image similarity into a CBIR system.…”

Using Emergence Phenomenon in Meaningful Image Segmentation for Content-based Image Retrieval

“…These include Latent Semantic Indexing (LSI), contextual search, user feedback, data clustering in the extraction of perceptual concepts, content-based soft annotation (CBSA), image classifications, ontology, top-down, ontologically driven approaches and bottom-up, automatic-annotation approaches, using machine learning methods to associate low-level features with query concepts, using relevance feedback to learn users' intention, generating semantic template to support high-level image retrieval, fusing the evidences from HTML text and the visual content of images for WWW image retrieval , use of ontology which represent task-specific attributes, objects, and relations, and relate these to the processing modules available for their detection and recognition, use of context-awareness for identifying image semantics and relationships to contribute to closing the semantic gap between user information requests and the shortcomings of current content-based image retrieval techniques , enhanced ICBIR system which allows users to input partial relevance which includes not only relevance extent but also relevance reason for a multi-phase retrieval where partial relevance can adapt to the user's searching intention in a from-coarse-to-fine manner [2]. Although these are good, constructive progresses in solving the problem of semantic gap in CBIR, they cannot define the semantic meanings of an image specifically.…”

The SinBiota 2.0 Biodiversity Information System

“…One interesting alternative is to construct the multimodal search architecture based on the works of Torres et al [12] and Fedel [65], with the inclusion of sound functionality, and to implement the content-based image search component as an extension of the LIRE [66] framework. Given that the multimodal search feature must be flexible enough to handle images and sounds from unrelated species, the Relevance Feedback [67], [68] technique can be adopted to improve search quality.…”