An Adaptive Spatial-Temporal Local Feature Difference Method For Infrared Small-Moving Target Detection

A significant challenge is correctly detecting tiny moving objects in a sequence of infrared (IR) photographs. we provide a brand-new technique termed spatial-temporal local feature difference (STLFD) to address this problem. In order to successfully increase the contrast between the target and the background, the approach adopted in our work comprises the application of spatial and temporal filters, followed by pixel-level adaptive background suppression (ABS). The proposed method is broken down into three different steps. Initially, the current frame image is used to gather three temporal frame images. Two feature maps are then retrieved from the data using filters from the spatial and temporal domains that were specifically created for this job. After combining the two domains, the output feature is produced. The pixel-level ABS module is further employed to successfully reduce noise interference. Utilizing a threshold allows for the creation of the segmented binary map. The experimental findings presented in this paper reveal that the suggested strategy outperforms current state-of-the-art approaches in the detection of IR tiny-moving object detection.