Multiple Frames Based Infrared Small Target Detection Method Using CNN

In the field of infrared, the detection of dim small target has been a challenging topic. Especially in the complex background, a large number of false alarms might appear when using traditional detection methods, this cannot meet the dual requirements of high detection rate and low false alarm rate. Therefore, in this paper, a CNN based target detection method is proposed for infrared sequence. For an image to be detected, firstly, aiming at the noise and interference which are spatially non-stationary but temporally stable, the method of registration and frame difference is used to suppress the background and extract the target, and the motion information between nearby frames is obtained; next, the extracted information is fed into the light CNN network for the extraction and fusion of different level features; finally, based on the fusion features the small target is detected. The proposed multiple frames based method relies on neural network to extract the motion information in infrared sequence to enhance the distinguishability of the target features under complex condition. Compared with the traditional methods, experimental results show that the proposed CNN method can achieve higher detection rate and lower false alarm rate, especially for the detection of dim small targets in complex background, the advantage is more obvious.