A Rapid Extraction System of Small Infrared Targets

Abstract

The invention discloses a system for quickly extracting small infrared targets, comprising: an image acquisition module for acquiring an image of a target to be detected; at least one image scaling module connected to the image acquisition module for performing an image acquisition on the image of the target to be detected Narrowing down; the target extraction module is connected to the image scaling module, and is used to extract the target point from the reduced target image to be detected, and extracts the feature value of the target point; the target judgment module is connected to the target extraction module for Judging whether the target point is a target to be extracted according to the feature value, if so, recording the position of the target to be extracted; a data cache module connected to the image acquisition module for storing the image of the target to be detected. The invention adopts an image scaling module to shrink the image of the target to be detected, so that it can be adapted to a preset template, thereby improving calculation efficiency, realizing rapid extraction of small infrared targets, and meeting real-time requirements.

Description

technical field [0001] The invention belongs to the field of infrared image processing, and in particular relates to a fast extraction system for small infrared targets. Background technique [0002] Because the infrared small target fast extraction algorithm of the multi-directional circular gradient method requires a large amount of calculation to simulate and process one frame on a PC, the processing speed is slow, and it cannot meet the requirements of real-time processing, so the application range is limited. FPGA, on the other hand, can perform interface expansion and high-speed parallel computing, can collect and capture signals of various interfaces, realize massive data storage, and simultaneously perform large data volume operations with high parallelism and low logic complexity to realize hardware acceleration. On the basis of the infrared small target fast extraction algorithm based on the multi-directional circular gradient method, the advantage of FPGA parallel...

AI Technical Summary

Problems solved by technology

[0003] In the process of transplanting the infrared small target fast extraction algorithm directly to the FPGA hardware platform, its operation logic is relatively complicated, and the entire operation cannot be well realized. Its resource occupancy rate and hardware cost are high, and fast data processing cannot be realized. Therefore, it is urgently needed Design an image extraction system with high detection efficiency and good real-time performance

Images