Research Of The Infrared Image Enhancement Technology And Detection Methods

With the rapid development of high-tech, infrared imaging technology is progressing ceaselessly. At present, the application of infrared thermal imaging device has gradually developed from the military field to the civilian field which shows a deeper trend. However, due to the restriction of optical system in infrared thermal imaging device and the influence of working environment, the effect of infrared imaging sometimes can’t meet people’s needs well. Especially in the applications of image target recognition, a series of enhancement algorithms and detection algorithms in the course of image processing is also needed.This paper first introduces the research background and significance, analyzes the research status of infrared imaging technology home and abroad and introduces the main applications of infrared imaging technology briefly; Secondly, the paper introduces some preliminary knowledge, including the principle of infrared imaging, the noise characteristics and the overview of infrared image processing technology; Again, the paper discusses infrared image enhancement algorithms in detail, the discussion is carried on from the aspects of image smoothing, image sharpening, histogram equalization, the paper emphasizes the median filtering and high-pass filtering method, and makes some beneficial improvement; The paper discusses the nonlinear enhancement algorithm in infrared images based on wavelet analysis, puts forward image enhancement treatment methods for weak infrared image and does some simulation experiments to verify these methods; Then the paper studies infrared image target detection algorithm, introduces mathematical morphology method, discusses lateral inhibition network method in detail and adaptively confirms the suppression factor of the lateral inhibition network through anisotropic Gaussian kernel function, which changes image information and takes advantage of image information effectively. Doing do comparative experiments with traditional methods, the results show that the method in this article is superior to traditional methods, suppresses the background clutter effectively and highlights the target. We applies this method to the automatic monitoring system of railway turnouts in fifth part. Results show that the design scheme of this article is more environment-adaptable, has higher detection accuracy, higher stability and is easy to maintenance. Finally, the paper makes some summary and outlook.