The Research Of Image Recognition Technology In Mobile Applications

Image recognition technology becomes moreand more mature with the rapid development of computer technology and information technology. The applications of image recognition technology are increasingly penetrated into our daily lives, such as license plate recognition, two-dimensional code recognition, fingerprint recognition and handwriting recognition, and so on. With the continuous improvement of smartphone capabilities, the image recognition technology has been increasingly applied in mobile phones, such as two-dimensional code recognition and fingerprint recognition, and so on. The fact that mobile is a necessity of our daily lifeand use the camera of mobile to identify is convenient make the research prospect of image recognition technology on intelligent terminals relatively broad.Quickly identify the target image from the background image, that is to say the image retrieval technology that based on local characteristics is currently a hot topic in the visual field. With the rapidly growing popularity of smartphones and increasing living standards, image recognition technology currently used on the computer can no longer meet the requirements of the people. Now people need fast and accurate recognition technology on smart phone. In this paper, the study is to make the efficiency and performance of SURF (Speeded Up Robust Features, robust features that fast) algorithm based on feature description on smart phone optimal. Compared with the computer, the defects that low processing speed and small memory resources of mobile phone result in limited image processing capability and which must be improved by improving and simplification handling of the existing algorithms; Directed at the defects of time-consuming feature points extracting and large memory-occupying, this paper proposes an improved SURF algorithm. The algorithm pretreatment for processing video frames, simplify the original algorithm, avoid extraction of useless information and use an adaptation threshold when extracting feature points. Experimental results show that using the improved SURF algorithm in mobile applications has achieved the purpose of real-time processing. It has certain values in both theory and practice.