Infrared weak and small moving target detection method based on track point space-time consistency

Abstract

The invention discloses an infrared weak and small moving target detection method based on track point space-time consistency. The infrared weak and small moving target detection method comprises thesteps: firstly extracting dense tracks with different lengths for continuous video frame images, and generating a dynamically changing track list; secondly, detecting candidate targets in the single-frame image by using a track point space aggregation model, and eliminating pseudo targets by using motion characteristics; providing a contour number coding mechanism for aggregation representation oftrack points; and finally, for continuous video frames, proposing a backward target detection algorithm based on contour code word consistency. According to the infrared weak and small moving targetdetection method, the spatial aggregation and the time consistency of the target track points are considered at the same time, so that the maximization of the detection rate and the minimization of the false alarm rate are solved.

Description

technical field [0001] The invention belongs to the technical field of infrared image processing methods, and in particular relates to a method for detecting small infrared moving targets based on the spatiotemporal consistency of track points. Background technique [0002] Infrared detectors have the advantages of excellent concealment ability, super environmental adaptability and anti-interference, small size, light weight and low power consumption. In recent decades, infrared detection technology has been widely used in people's lives. , medical, military, engineering and other fields have more and more extensive applications. However, infrared video is generally obtained by long-distance shooting of the detector. The imaging area of ​​the target on the imaging system is small, and the number of pixels occupied in the current video frame image is limited; the brightness is low, and there is no clear shape and texture information; the contrast is poor, and the infrared We...

AI Technical Summary

Problems solved by technology

However, infrared video is generally obtained by long-distance shooting of the detector. The imaging area of ​​the target on the imaging system is small, and the number of pixels occupied in the current video frame image is limited; the brightness is low, and there is no clear shape and texture information; the contrast is poor, and the infrared Weak and small targets are often submerged in background noise; the direction and size of motion are difficult to predict, so how to efficiently detect infrared weak and small targets in video under complex background environments has important military and theoretical significance

Images