A Smart Image Background Recognition Model and Its Application to Content-Based Image Retrieval

A smart image background recognition model is proposed, in which inverse square smoothness model (ISSM) and canny edge algorithm are combined. Compared with methods such as method using background difference based on edge extracting, this method is well performed in processing image with complex background. ISSM is practical and robust model, as it makes full use of the relationship between image pixels. It is proved in experiments that the performance of image retrieval is substantially improved when using image background recognition model proposed in this paper. The average accuracy of image retrieval increases greatly by adding image background recognition model into image retrieval system.