Infrared Small Target Detection And Tracking Based On Background Suppression

Infrared imaging technology has many advantages such as well-concealment and strong anti-interference ability,it gets a widely use in the military and civilian applications fields.In practical applications,because of the long imaging distance,the infrared targets are small and lack of shape and texture features,thus making target detection and tracking very difficult.Therefore,achieving accurate detection and tracking of infrared small targets is an important and difficult task.The major wok and results of research are as follows.In the image preprocessing,this thesis studied the characteristics of infrared image and classical algorithms of image preprocessing.To overcome the limitations of traditional Robinson Guard filter,this thesis proposes an improved Robinson Guard filter algorithm for background suppression by combining with visual contrast mechanism.Experiments confirm that this algorithm can adjust the filter window size adaptively,and it is adaptive to small targets with different sizes and effectively suppress different backgrounds.In terms of target detection,aiming at the shortcoming that traditional Robinson Guard filter is easily influenced by clutter and noise,this thesis presents a kind of infrared small target detection algorithm based on the combination of Top-hat transformation and Robinson Guard filter.This algorithm effectively improves target detection performance and robustness.Meanwhile,according to frequency characteristics of the infrared image,this thesis adopts Gaussian low-pass filter and spectral residual approach to obtain saliency image in frequency domain,and then uses Robinson Guard filter to suppress the background and noise in spatial domain.Experimental results demonstrate that this algorithm has good ability to process images with different background and low signal-to-noise ratio(SNR).On target tracking,this thesis studied Kalman filter algorithm and SIFT algorithm.Because Kalman filter algorithm cannot accurately track the small targets with high mobility,this thesis adopts Kalman filter algorithm to estimate velocity and acceleration of targets.Combining the true positions of targets in the current frame with the estimation of velocity and acceleration of targets in the next frame,we can indirectly obtain the estimation of positions of targets in the next frame.When the targets cannot be detected accurately by the detection algorithm,we adopt SIFT algorithm to match the estimated positions of undetected targets with previously detected targets,in order to make up for disadvantages of detection algorithm and improve the tracking performance.