Novel Precision Vision Measurement Method Between Area-Array Imaging and Linear-Array Imaging Especially for Dynamic Objects
Novel Precision Vision Measurement Method Between Area-Array Imaging and Linear-Array Imaging Especially for Dynamic Objects
复制标题
DOI：
10.1109/tim.2022.3207826
复制
发表时间：
2022
影响因子：
5.6
通讯作者：
Yang Gao;H. Cui;Xu Wang;Zhe Huang
中科院分区：
文献类型：
--
作者：
Yang Gao;H. Cui;Xu Wang;Zhe Huang
本文提出了一种新颖的高性能视觉测量方法。通过单向像素合并，将传统面阵相机转换为等效的窄面阵相机，然后将这三个相机合理组合，实现精确的空间位置和姿态测量。该方法成像介于面阵成像和线阵成像之间，能够克服基于面阵相机的双目立体视觉测量系统（BSMS）采集速度有限、高帧率下灵敏度不足、数据量大的缺点。同时还克服了采用柱面透镜的多线阵相机测量系统（MLMS）易受环境光干扰、多目标能力弱的缺点。仿真和物理实验验证，当采样速度提高8倍，数据量减少到3/16时，本文方法的测量精度与BSMS处于同一水平。
A novel high-performance vision measurement method is proposed in this article. Via single-direction pixel binning, a traditional area-array camera is converted into an equivalent narrow area-array camera, and then a reasonable combination of these three cameras is used to realize precise spatial position and attitude measurements. This method, whose imaging is between area-array imaging and linear-array imaging, is able to overcome the shortcomings of the binocular stereovision measurement system (BSMS) based on area-array cameras, which has limited acquisition speed, insufficient sensitivity at a high frame rate, and a large amount of data. And it also overcomes the disadvantages of the multiple linear-array-camera-based measurement system (MLMS) using cylindrical lenses, which are easy to be disturbed by ambient light, and has weak multi-target capabilities. Simulation and physical experiments validate that, the measurement accuracy of the method in this article is at the same level as BSMS, when the sampling speed is increased by up to 8 times, and the amount of data is reduced to 3/16.