The Detection And Recognition Of Infrared Weak And Small Targets

With the rapid development of computer and digital image processing technology, the infrared target detection technique has been used in many applications, such as astronomy prognosticates, remote sensing, target tracking, guiding etc. But with the characteristics of far distance, small area, weakness of shape of image, lose of detail of image and low SNR etc, the infrared weak and small target detection technique is one of the key and difficult point of infrared guidance system.Because of the characteristics of infrared weak and small targets, the research of infrared target detection of theories and methods is concentrated on raising of SNR, the detection ability of algorithm, coverage, efficiency of computation and reliability of detection algorithm. The research of infrared weak small target detection is still a issue with great challenge.This paper first analysed the common algorithm of image pre-processing and thresholding, concluded the advantage and short coming of different algorithm, then, brought forward a infrared weak and small targets detection method. The most important characteristic is: At first, a method to classify different kinds of noise points and the points belong to background was issued, in order to reduce the impress which was caused by the misuse of filtering algorithm; second, put forward and used fuzzy weighted mean filtering algorithm based on gray-level and space-level, to process the pixel points which is sufferd gauss noise; At last, extracting the targets using 2D-level ostu algorithm. Improving the performance by correcting the weak point of this algorithm - the computation amount is huge, which deduced the time complexity evidently, but the space complexity had not been changed so much, so it is good at real time computation.At last, this target detaction alghrothm was programmed under the platform of VC6. The experiment showed that the infrared weak and small target detection algorithm the paper suggested is rational in process, runs quickly, viable, and especially good in real-time application, and the ratio of detection is high.