自动调焦在单目测距方法中的研究

详细信息    查看全文 | 推荐本文 |

英文篇名：Study on Monocular Distance Measurement Based on Auto Focus 作者：任工昌 ; 刘崇展 ; 刘朋 ; 张戈 英文作者：REN Gong-chang;LIU Chong-zhan;LIU Peng;ZHANG Ge;School of Mechatronics Engineering,Shaanxi University of Science and Technology; 关键词：单目测距 ; 图像处理 ; 自动对焦 ; 搜索算法 ; 变焦镜头 ; 定量关系 英文关键词：Monocular Distance Measurement;;Image Processing;;Auto Focus;;Search Algorithm;;Zoom Lens;;Quantitative Relationship 中文刊名：JSYZ 英文刊名：Machinery Design & Manufacture 机构：陕西科技大学机电工程学院; 出版日期：2019-04-08 出版单位：机械设计与制造 年：2019 期：No.338 基金：国家自然科学基金项(51175314);; 陕西省教育厅基于摄像技术的智能化项目(15JK1079);; 陕西省农业科技创新项目(2010NKC-32) 语种：中文; 页：JSYZ201904040 页数：4 CN：04 ISSN：21-1140/TH 分类号：154-157

摘要

针对传统测距技术和自动调焦方法的局限性,从光学系统的成像理论出发,在图像处理的基础上,提出一种以变焦镜头为图像采集设备、基于物距与电机脉冲数定量关系的单目视觉测距新方法。该方法根据光学成像原理,计算不同物距下目标图像的清晰度值,根据改进的粗精变步长爬山搜索策略实现自动对焦,记录步进电机转向及脉冲个数,建立脉冲数与物距的定量关系,实现测距。以粮仓为例验证了该方法的可行性和有效性,且相对误差小于3%、方差小于0.702,满足精确度和稳定性要求,避免了复杂的摄像机标定,计算简单且易于实现。
        In view of the limitations of traditional ranging technology and automatic focusing method,based on the imaging theory of optical system,a new method of monocular vision ranging based on the quantitative relationship between object distance and motor pulse number and a zoom lens for image acquisition equipment is proposed. According to the principle of optical imaging,the sharpness value of the target image is calculated,and the auto focus is recorded according to the improved coarse and fine step-by-step climbing search strategy. The stepping motor steering and the number of pulses is recorded, the quantitative relationship between the pulse number and the material distance is established, distance measurement is achieved. The feasibility and validity of the method are verified by the granary as an example. The relative error is less than 3% and the variance is less than 0.702,which satisfies the accuracy and stability requirements. The complicated camera calibration is avoided and is simple and easy to implement.

引文

[1]林鹰,付洋.散装仓粮食数量识别关键-矩形标尺图像识别[J].浙江工业大学学报,2007,41(10):1643-1646.(Lin Ying,Fu Yang.Key of bulk warehouse grain quantity recognitionRectangular benchmark image recognition[J].Journal of Zhejiang University:Engineering Science,2007,41(10):1643-1646.)
    [2]王文波.基于单目视觉的实时测距方法研究[D].大连:大连理工大学,2014.(Wang Wen-bo.Study on real-tmie distance detection based on monocular vision technique[D].Dalian:Dalian University of Technique,2014.)
    [3]周天弋.基于计算机视觉的车辆与前方车距检测[D].杭州:浙江工商大学,2010.(Zhou Tian-yi.Vehicle and forward distance detection based on computer vision[D].Hangzhou:Zhejiang Gongshang Univerdity,2010.)
    [4]郭磊,徐友春,李克强.基于单目视觉的实时测距方法研究[J].中国图象图形学报,2006(1):74-81.(Guo Lei,Xu You-chun,Li Ke-qiang.Study on realtmie distance detection based on monocular vision technique[J].Journal of Image and Graphics,2006(1):74-81.)
    [5]陈国金.数字图像自动聚焦技术研究及系统实现[D].西安:西安电子科技大学,2007.(Chen Guo-jin.Study on the auto-focusing technology of digital image and realization for an auto-focusing system[D].Xi’an:Xidian Univerdity,2007.)
    [6]刘雪超.基于数字图像处理的自动调焦技术研究[D].长春:中国科学院研究生院:长春光学精密机械与物理研究所,2014.(Liu Xue-chao.Digital image processing based on auto-focus technology[D].Changchun:University of Chinese Academy of Science,2014.)
    [7]陶金有.基于图像处理的自动调焦算法研究及系统实现[D].西安:中国科学院研究生院:西安光学精密机械研究所,2014.(Tao Jin-you.Study on the auto-focus algorithms based on image processing and its system realization[D].Xi’an:University of Chinese Academy of Science,2014.)
    [8]Chih-Yung Chen,Rey-Chue Hwang,Yu-Ju Chen.A passive auto-focus camera control system[J].Applied Soft Computing Journal,2009(101).)
    [9]韩延祥,张志胜,戴敏.用于目标测距的单目视觉测量方法[J].光学精密工程,2011(5):1110-1117.(Han Yan-xiang,Zhang Zhi-sheng,Dai Min.Monocular vision system for distance measurement based on feature points[J].Optice and Precision Engineering,2011(5):1110-1117.)
    [10]高军强,刘桂华.多摄像机线结构光的钢轨表面三维缺陷检测[J].机械设计与制造,2017(3):170-172+176.(Gao Jun-qing,Liu Gui-hua.3D defect detection technology for rail surface withMulti-camera line structure light[J].Machinery Desion&Manufacture,2017(3):170-172+176.)
    [11]Kehtarnavaz N.Development and real-time implementation of a rulebased auto-focus algorithm[J].Real Time Imaging,2003:197-203.