Infrared weak and small target detection method based on integrated fusion features

Abstract

The invention discloses an infrared weak and small target detection method based on integrated fusion features, which relates to the field of machine learning, and comprises the following steps: acquiring an initial image of an infrared weak and small target as a training set, and establishing a classifier to obtain a trained model; obtaining a to-be-detected image and filtering the to-be-detected image through a high-pass filter; performing constant false alarm threshold segmentation on the filtered image; carrying out candidate target area marking on the segmented binary image, and calculating the center coordinate of the binary image; according to the center coordinate of each candidate target, image blocks are taken from the to-be-detected image; extracting feature parameters of each to-be-detected image block; and classifying the feature parameters of the to-be-detected image block through the trained model, obtaining and outputting the center coordinate of the target, and completing target detection. According to the method, the classification capability of the fusion features is higher, the classification precision can be improved, the convergence speed can be increased, and thus the purpose of reducing classifier parameters is achieved; the method has enough adaptability to complex application scenes, and is convenient for engineering application.

Description

technical field [0001] The invention relates to the field of machine learning, in particular to a method for detecting small and weak infrared targets based on integrated fusion features. Background technique [0002] Infrared weak target detection technology is one of the core technologies of the airborne optoelectronic system, and is the basic premise of target surveillance and reconnaissance and precise strike. With the development of technology, the photoelectric detection distance is required to be farther and farther. Infrared targets have a small imaging size under long-distance conditions, and even the human eye is difficult to distinguish them, and are interfered by clutter in various complex scenes in the air, ground, and sea, making it difficult for weak and small targets to be accurately detected. [0003] At present, infrared small target detection technology can be divided into methods based on multi-frame and single-frame images from the technical route. The...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the above-mentioned shortcomings in the prior art, the infrared dim and small target detection method based on integrated fusion features provided by the present invention solves the problems that the existing methods have insufficient adaptability to complex application scenarios or are difficult to apply in engineering due to high computational complexity

Images