Content-based image retrieval, namely CBIR (Content-based image retrieval), is a research branch focusing on large-scale digital image content retrieval in the field of computer vision. A typical CBIR system allows users to enter a picture to find other pictures with the same or similar content. Traditional image retrieval is text-based, that is, the query function is realized through the name of the picture, text information and index relationship.

This concept was put forward by T.Kato in 1992. In this paper, he established an image database based on color and shape, and provided some retrieval functions for experiments. Since then, the image retrieval process based on image feature extraction and the concept of CBIR have been widely used in various research fields, such as statistics, pattern recognition, signal processing and computer vision.