Study of Image Retrieval Method Based on Salient Points and Comprehensive Characteristics

Technology has been a very good development in the past twenty or thirty years, content-based image retrieval, many low-level visual features is proposed for image retrieval, real-time problem in image retrieval has got great attention of researchers, content-based image retrieval technique has been widely used in medical, education, digital library, industrial and commercial fields and based on the military field. This paper describes the image retrieval technology research background and significance, introduces the current research situation and research hotspot in content-based image retrieval, the basic method of image retrieval based on content and key problems are explained in detail. The image can cause visual attention point, known as the significant point. The literature and presents a new method for automatic extraction of salient points, and on this basis to achieve significant point based image retrieval. Find the analysis of experimental results: the foreground and background are distinct and image background color of a single, can extract salient points effectively, the recall rate and correct rate was higher; the background image is not obvious, is not conducive to the significant point. Extraction, the retrieval precision rate and recall rate is low. Copyright © 2013 IFSA.