The Research Of Image Segment In The Content-Based Image Retrieval

The explosion in the amount of images being generated in many different fields and the growth of the internet emphasizes the need for retrieving images based on their contents. Therefore, CBIR(content-based image retrieval) has attracted researchers from different fields like computer vision, image processing, database, etc. Furthermore, CBIR has demonstrated its wide usefulness in many fields such as biochemistry, medical diagnosis, etc.Many researches have proposed different algorithms to represent and retrieve images based on global features, such as the color histogram of the whole image. Thus, it is essential to segment and then represent the image by the local features of its composing region. As a consequence, CBIR from the vast collection of image will be more effective. Therefore, image segmentation aims at dividing an image into several segments where each segment is visually coherent. Image depends on many features like intensity, blurring, and even number of segmentation.Many efforts towards image segmentation have lead to many different techniques. Most existing automatic image segmentation methods can be classified into these approaches:1. Threshold techniques.2. Boundary-based techniques.3. Region-based techniques.4. Clustering techniques.In this paper the network image segmentation universality is attained by studying separately gray image segmentation and color image segmentation. Gray segments improved over diverse traditional image segmentation and based on OTSU. It is characters was found by formula derivation, and by its characters the results amended latterly. This paper also discussed the early necessity of sub-image segmentation and the quantity of images to segment.About color segments, we select an algorithm named hill-climbing in HSV which is used to explain the color space by people to segment the image. However, It didn't up to the standard, so we quantified the input, modified the output and merged the small area by analysis of the problem to make the output perfect.