Study and Analysis of User Desired Image Retrieval

Background: In the present digital world, Content Based Image Retrieval (CBIR) has gained significant importance. In this context, image processing technology has become the most sought one, as a result its demand has increased to a large extent. The complex growth concerning computer technology offers a platform to apply the image processing application. Well-known image retrieval techniques suitable for application zone are 1. Text Based Image Retrieval (TBIR) 2. Content Based Image Retrieval (CBIR) 3. Semantic Based Image Retrieval (SBIR) and etc. In the recent past, many researchers have conducted extensive research in the field of content-based image retrieval (CBIR). However, many research related studies on image retrieval and characterization have exemplified to be an immense issue and it should be progressively developed in its techniques. Hence, by putting altogether the research conducted in recent years, this survey study makes a comprehensive attempt to review the state-of-the art in this field. Aims: This paper aims to retrieve similar images according to visual properties, which are defined as shape, color, texture and edge detection. Objective: This study investigates the CBIR to achieve the task because of the essential and fundamental problems. The present and future trends are addressed to show contributions and directions and can inspire more research in the CBIR methods. Result: We present an in-depth analysis of state of the art on CBIR methods; We explain the methods based on color, texture, shape and edge detection with performance evaluation metrics. In addition, we have discussed some significant future research directions reviewed. Methods: This paper has quickly anticipated the noteworthiness of CBIR and its related improvement, which incorporates Edge Detection Techniques, various sorts of Distance Metric (DM), performance measurements and various kinds of Datasets. This paper shows the conceivable outcomes to overcome the difficulties concerning re-ranking strategies with an improved accuracy. Conclusion: At last, we have proposed another technique for consolidating different highlights in a CBIR framework that can give preferred outcomes over the current strategies.