A moving object deblurring method based on distributed camera array

Abstract

The invention discloses a moving target deblurring method based on a distributed camera array, which is implemented according to the following steps: Step 1, building a distributed camera array system; Step 2, associating objects in the array, and then moving the associated image sequence Target 3D field flow estimation; Step 3. Sampling the 3D field flow sequence respectively, and using the triangular transformation method to estimate the motion point of the fuzzy target according to the sampling points; Step 4. Estimating the motion point of the fuzzy target by linear interpolation method according to the motion point Fuzzy kernel function; step 5, segment the fuzzy moving target to obtain a target sequence, and use the fuzzy kernel function obtained in step 4 to deconvolute to obtain a deblurred image of the target. Multi-frame rate and multi-exposure strategies are adopted between different camera arrays to overcome the contradiction between the exposure time of a single camera and high-resolution image acquisition; the configuration of the camera array is used to improve the detection range of the field of view and achieve the detection of moving targets under the requirement of a large detection field of view. Vague.

Description

technical field [0001] The invention belongs to the technical field of optical detection and relates to a method for deblurring a moving target based on a distributed camera array. Background technique [0002] In the multi-camera array optical detection system, due to the fast moving speed of the moving target in the detection field of view and the long exposure time of the camera, the relatively large displacement of the target between the beginning and the end of the camera exposure will result in the formation of target overlap. Shadow, ie, motion blur effect. A clear target image is an important prerequisite for target detection, tracking, and recognition. Therefore, research on high-precision, high-reliability moving target deblurring methods is an important means to improve the effects of optical detection system target detection, tracking, and recognition. Existing moving target deblurring methods are mainly divided into two categories: one is a software-led method ...

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to provide a moving target deblurring method based on a distributed camera array, to solve the problem that the existing method has a small detection range and cannot make full use of the advantages of the baseline length between camera arrays to provide the necessary multi-view target motion information

Images