SVM-based radar RD image weak target detection method and system, storage medium and electronic terminal

Abstract

The invention provides a radar RD image weak target detection method based on an SVM, and the method comprises the steps: obtaining radar echo data, carrying out the preprocessing of the data, and generating a radar distance Doppler dimension image; preprocessing and labeling the original image to obtain a data set; constructing a support vector machine (SVM) network for weak target detection under a low signal-to-noise ratio; performing target detection and cross validation on the constructed support vector machine (SVM) network; obtaining a test set target detection accuracy result. According to the radar RD image target detection method, the target detection network is obtained through repeated training of the SVM network based on a large amount of target-containing radar RD image data under different signal-to-noise ratios; the target detection network obtained by the method has the advantages of high detection accuracy, good practical effect and the like under a low signal-to-noise ratio, and the method is suitable for popularization and application in practical application.

Description

technical field [0001] The invention relates to the technical field of image target detection, in particular to a SVM-based radar RD image weak target detection method, system, storage medium and electronic terminal. Background technique [0002] In recent years, target detection has been widely used in many fields. Radar, as an important means of target detection, can analyze and process the echo in the irradiation area, detect target information from clutter, interference, noise and other signals, and determine its distance, speed, angle and other parameters, but in the complex background, especially in the low signal-to-noise ratio, the echo information often contains a lot of noise and clutter information in addition to the target, and the amplitude of the noise is close to the target, and the clutter often appears Non-linear, non-Gaussian, non-uniform, non-stationary characteristics, which greatly limit the radar target detection performance. [0003] The existing rada...

AI Technical Summary

Problems solved by technology

[0002] In recent years, target detection has been widely used in many fields. Radar, as an important means of target detection, can analyze and process the echo in the irradiation area, detect target information from clutter, interference, noise and other signals, and determine its distance, speed, angle and other parameters, but in the complex background, especially in the low signal-to-noise ratio, the echo information often contains a lot of noise and clutter information in addition to the target, and the amplitude of the noise is close to the target, and the clutter often appears Non-linear, non-Gaussian, non-uniform, non-stationary characteristics, which greatly limit the radar target detection performance

[0003]Existing radar target detection methods include constant false alarm detection algorithm, etc., constant false alarm algorithm is based on statistical model, it is often difficult to accurately describe the background model, in the non-uniform background , especially under the low signal-to-noise ratio with various types and shapes, there will be serious constant false alarm loss, and the detection performance will decline

To sum up, the existing radar RD image target detection methods have problems such as simple model, low universality, and weak learning ability, and it is difficult to fundamentally solve the problem of weak detection ability of small and small targets in radar RD images.

[0004] The publication number is CN111913158A, which provides a radar signal processing method for detecting low, slow and small targets in a complex clutter background. It obtains the target track in the reference window through the constant false alarm detection algorithm, and then obtains the speed and pitch of the target according to the track information. , orientation, distance, etc. The constant false alarm algorithm is based on a statistical model, and it is often difficult to accurately describe the background model. In a non-uniform background, especially under a low signal-to-noise ratio with various types and shapes, serious constant false alarms will appear. Therefore, the existing radar RD image target detection methods have problems such as simple models, low universality, and weak learning ability, and it is difficult to fundamentally solve the problem of weak detection ability of small and small targets in radar RD images.

Images