Long-distance weak and small target visual detection method based on self-adaptive space-time fusion

Abstract

The invention discloses a long-distance weak and small target visual detection method based on self-adaptive space-time fusion. The method comprises the following steps: 1, filtering a ground background below a horizon in a video frame image to obtain a spatial domain map; 2, processing the spatial domain graph by using a dark target inter-frame difference method to obtain a time feature graph; 3,processing the spatial domain graph by using a row-column decoupling bottom cap morphological filtering method to obtain a spatial feature graph; 4, designing an adaptive switching space-time featuremap fusion mechanism to fuse the time feature map and the space feature map to generate an adaptive space-time fusion map; and 5, local adaptive threshold segmentation.Noise and clutter are suppressed while the contrast of the target and the background is enhanced, so that long-distance weak and small target detection with high accuracy, low false detection and less missing detection is realized.

Description

technical field [0001] The invention relates to the field of computer vision, in particular to a method for visually detecting long-distance weak and small targets based on adaptive spatio-temporal fusion. Background technique [0002] With the gradual opening of low-altitude airspace and the increase of non-cooperative small targets such as small unmanned aerial vehicles, model airplanes, gliders, delta wings, and kites, the difficulty of airspace control has greatly increased. In particular, small multi-rotor UAVs have been widely used in industrial and civil applications, including personal aerial photography, entertainment, etc. , agriculture, forestry, logistics, meteorology, security, etc. But at the same time, in recent years, incidents of "black flying" and "indiscriminate flying" of drones have occurred frequently all over the world, seriously endangering personal privacy, public places, aviation and national security. In order to cope with the challenges to indiv...

AI Technical Summary

Problems solved by technology

Due to the fact that the target is far away from the surveillance camera and the monitoring environment is complex, the obtained observation target has the characteristics of weak signal, small imaging area, weak shape or texture feature, no fixed motion law, and lack of relative motion. There are many problems in the current weak target visual detection. challenge

Images