Improved U-Net Network for Infrared Small Target Detection

Infrared small target detection is a key technology in infrared search and tracking systems. Despite the rapid development of computer vision in recent years, infrared small target detection based on deep learning is still few and far between due to the lack of texture information and detailed feature of infrared small target. In this paper, we propose an improved U-Net for infrared small target detection, which adds a multi-scale feature fusion module on top of the U-Net. In order to better capture the key information of infrared small target, the multi-scale feature module fuses the high-level semantic information in the convolutional neural network and the low-level small target detail feature across layers. Experimental results show that our proposed method is able to segment infrared small target well and reduce false alarms caused by complex backgrounds compared to traditional methods. In addition, our proposed method performs better than the original U-Net, demonstrating the effectiveness of our proposed improved U-Net.