Burst point detection system and method based on FPGA

Abstract

The invention discloses a burst point detection system and method based on an FPGA. The system includes a signal control and data processing module, a decoding circuit, a data caching circuit, an Ethernet circuit, and a peripheral circuit. The decoding circuit obtains original image data and caches the original image data to the large-capacity data caching circuit, an improved three-frame difference method in the signal control and data processing module is used to perform burst point target detection to obtain a burst point area, then a gray-scale weighted algorithm is used to obtain center pixel coordinates of a burst point after the connected components labeling, and finally the detected burst point image and the center pixel coordinates of the burst point are sent to a PC via the Ethernet circuit. The system and method of the invention implement the pipeline design of the algorithm steps, obtain high-speed operation results, and can realize the real-time detection of the results ofthe burst point on an image with a resolution of 1024*1024; can accurately obtain the burst point area and can quickly and accurately label the center point coordinates of the burst point; and can still maintain a high detection rate under the premises of light change and background movement.

Description

technical field [0001] The invention relates to the field of target detection, in particular to an FPGA-based explosive spot target detection system and method. Background technique [0002] With the development of social economy and science and technology, the requirements for the speed and accuracy of target detection are getting higher and higher. The explosion point target of explosive body is a rapidly changing target. In the civilian field, the explosion point detection of fireworks and firecrackers is helpful for urban supervision and fire prevention. In the field of military and national defense, shell blast point detection is an important link in the weapon testing and evaluation system, which can be used in shooting range shell shooting verification, damage effect evaluation, and local fire position estimation in counter-fire strikes. At present, the main methods used in the detection of explosive spots include the following two aspects. One is to use laser instru...

AI Technical Summary

Problems solved by technology

Among the above three methods, the adjacent frame difference method is simple to calculate and can achieve high real-time performance, but the segmented target is not complete enough to extract all the feature pixels. The background subtraction method can provide the most complete feature data and obtain a relatively complete target, but it is not robust to changes in the background. The optical flow method is suitable for accurate analysis and can obtain the motion parameters of the target, but its calculation is complicated and it is difficult to meet the real-time requirements.

[0005] In recent years, scholars at home and abroad have conducted in-depth research on the use of image information for explosive point detection, and formed corresponding results, but there are still the following problems: 1. Papers [Qin Xiaoyan, Wang Xiaofang, Chen Ping, etc. based on Explosion point detection of shells based on Adaboost algorithm[J].Acta Ordnance Engineering, 2012,33(6):682-687. A detection algorithm of explosion point was constructed based on Adaboost algorithm, but the selection of the explosion point in this algorithm depends on Based on prior information, it is difficult to meet this requirement for the battlefield environment or some other military environments; 2. The paper [Qin Xiaoyan, Yuan Guanglin. Detection of bomb explosion points based on principal component tracking and analysis[J]. Acta Armaments, 2013 ,34(10):1266-1272.] Proposed a detection method based on Principal Component Pursuit (PCP) and Principal Component Analysis (PCA), which is relatively complicated and cannot be transplanted into hardware

3. Patent [CN201710023813.6] According to the different characteristics of explosive spots and noise spots or moving targets, the explosive spot targets can be identified from the original traces detected by radar, but this method requires the help of a radar system, which is complex and expensive more expensive

Images