Joint Local Contrast and First-Order Directional Derivative Methods for Detecting Infrared Small Targets in Complex Backgrounds

Infrared target detection technology is one of the most key technologies of current detection systems, which is not only applied to automatic navigation, security protection, and surveillance devices in daily life but also widely used in the military. However, due to the influence of complex background, the infrared detection system is often difficult to achieve accurate and fast detection of small targets, so the current small target detection methods still need to be improved. In this article, an effective algorithm is proposed by complementally combining local contrast and first-order directional derivative (FODD) algorithms, incorporating human visual characteristics for enhancement. Then according to the different characteristics of the background and small targets in infrared images, for the problem of excessive redundant information in infrared weak small target images, the high real-time and low leakage rate of local contrast method is used to find out the information-rich suspicious blocks, and then the low false alarm rate of FODD is used to find out the targets in the suspicious area. Through a lot number of experiments, and comparing with a variety of other algorithms. The results show that the detection algorithm proposed in this article is significantly better than other algorithms in terms of low false alarm rate and can maintain good real-time performance.