A Method for Recognition of 2D Chaff Distribution Characteristics Based on Support Vector Machine Algorithm

Abstract

The invention belongs to the technical field of identification of two-dimensional chaff cloud distribution characteristics, and discloses a two-dimensional chaff distribution characteristic identification method based on a support vector machine algorithm, which performs test preprocessing and database building operations, and is within the viewfinder range of a high-speed camera Throw known types of chaff, and properly take and save multiple photos of chaff diffusion as the data source of the database; formally conduct experiments, and take photos of the diffusion process of chaff clouds containing unknown types of chaff; Based on the support vector machine algorithm and the HOG feature extraction method, the edge search is performed on the collected test pictures, and the pictures are segmented according to the edge continuity; the chaff length and posture recognition are performed on each segmented sub-object, and the length is regarded as the value to be obtained. The length of the chaff and the attitude angle of the chaff can be obtained by calculating the slope of the line connecting the upper and lower endpoints. The invention effectively improves the accuracy of establishing the chaff cloud diffusion model, and at the same time improves the data processing efficiency in the chaff test.

Description

technical field [0001] The invention belongs to the technical field of two-dimensional chaff cloud distribution characteristic recognition, and in particular relates to a two-dimensional chaff distribution characteristic recognition method based on a support vector machine algorithm. Background technique [0002] At present, the closest existing technology: in electromagnetic warfare, the use of chaff cloud has a long history, simple production, obvious effect, and low cost, and has been widely used in the battlefield. As the electronic confrontation between the enemy and the enemy, the radar as the main reconnaissance equipment plays a vital role in the defense of countries all over the world. Therefore, in modern warfare, especially in future warfare, chaff jamming technology releases a large amount of chaff in space, resulting in a large difference between the echo obtained by the enemy radar and the real echo of the target, so that the enemy radar Cannot identify target...

AI Technical Summary

Problems solved by technology

[0004] To sum up, there are several significant deficiencies in the establishment of the chaff cloud diffusion model: First, the chaff cloud diffusion model based on the ideal model has no practical significance in the battlefield environment, the battlefield environment or the wake environment are intricate, and the chaff cloud The diffusion of chaff generally does not follow a uniform distribution, normal distribution or exponential distribution, and using these ideal distributions to establish a chaff cloud model will not be able to simulate the real environment; secondly, it is difficult to establish a complex and smooth chaff cloud diffusion model through theoretical means. The reason is that there are still some unresolved problems in the aerodynamics in the complex flow field. At the same time, using traditional numerical methods (such as using the finite time domain difference algorithm to solve the diffusion equation) to simulate the flow field and chaff displacement faces many difficulties. A large amount of calculation; again, in the existing chaff cloud movement and diffusion experiments, image recognition technology is rarely introduced into it, and the collected pictures are analyzed and chaff feature extraction is performed, and relevant data are processed manually (such as chaff Attitude analysis and chaff center point distribution characteristics analysis) will face a huge workload and consume unnecessary manpower and material resources

[0006] (1) The chaff cloud model based on the common distribution in the existing technology is quite different from the actual one, especially in the chaff diffusion modeling in the complex flow field, there is no unified theoretical method at present

[0007] (2) In the existing chaff cloud diffusion experiments, image recognition technology has not been used to extract the attitude and distribution characteristics of chaff, which makes the modeling of chaff cloud diffusion based on pictures inefficient

Images