Three-dimensional measurement and defect detection based on single image

被引:0
|
作者
Song, L [1 ]
Qu, X [1 ]
Xu, K [1 ]
Lv, L [1 ]
机构
[1] Tianjin Univ, State Key Lab Precis Measuring Technol & Instrume, Tianjin 300072, Peoples R China
来源
关键词
defect; detection of defects; tonal value; slant; tilt;
D O I
暂无
中图分类号
T [工业技术];
学科分类号
08 ;
摘要
Three-dimensional measurement and defect detection is an effective guarantee for improving the product quality. In this paper, we present a new 3-D overall measurement method using only one image, which can be effectively used in the defect detection. Compared with the former defect technology, such as the eddy current technology, the leakage of magnetic technology, the infrared technology, and the radial technology, our defect detection method is low cost, easy to operate, and can be expanded widely. The main device in the detection system is a CCD. When fixed the CCD in the workstation, CCD can capture the image of each workpiece. From the single image, we can get the three-dimensional data of the workpiece's shape ranged from 0 degrees to 180 degrees. Moreover, it can give an overall recognition of plane dimensions and depth information of the defect. The main working theory is, according to the 3D cues, "shading value" information left by single measurement image. According to shading, we can acquire the workpiece slant and tilt of every point's normal in the light source system of coordinates, and then transferring it to the image system of coordinates, calculating the depth information by slant and tilt. Compared with the former three-dimensional shape recovery method based on single image, this method has great improvement on the design of optical system, such as improving three-dimensional recovery accuracy and speed, with the simplification of the arithmetic, and first apply it to the industrial environment detection of defect. The three-dimensional detection method of defects, mentioned in this paper, has advantages of simple working pattern and workpiece configuration that just needs light source, CCD, image-collection card and computer, which greatly reduces the cost, and is good for use in industrial application.
引用
收藏
页码:1029 / 1038
页数:10
相关论文
共 50 条
  • [41] Examination of facial measurement on the image using a noncontact three-dimensional measurement device
    Watarai, Yuko
    Koide, Kaoru
    Ota, Shin
    Mizuhashi, Fumi
    JOURNAL OF FAMILY MEDICINE AND PRIMARY CARE, 2022, 11 (12) : 7776 - 7782
  • [42] Three-dimensional particle image velocimetry measurement through three-dimensional U-Net neural network
    Cao, Lixia
    Hossain, Md. Moinul
    Li, Jian
    Xu, Chuanlong
    PHYSICS OF FLUIDS, 2024, 36 (04)
  • [43] Three-Dimensional Defect Characterization of Ultrasonic Detection Based on GCNet Improved Contrast Learning Optimization
    Wang, Xinghao
    Wang, Qiang
    Zhang, Lei
    Yu, Jiayang
    Liu, Qiuhan
    ELECTRONICS, 2023, 12 (18)
  • [44] Wind turbine blade defect depth detection based on three-dimensional heat conduction model
    Zhang X.
    Zhou B.
    Li H.
    Yi Qi Yi Biao Xue Bao/Chinese Journal of Scientific Instrument, 2021, 42 (01): : 174 - 182
  • [45] Single Image Based Three-Dimensional Scene Reconstruction Using Semantic and Geometric Priors
    Yoon, Gang-Joon
    Song, Jinjoo
    Hong, Yu-Jin
    Yoon, Sang Min
    NEURAL PROCESSING LETTERS, 2022, 54 (05) : 3679 - 3694
  • [46] Automated Three-Dimensional Image Based Localization of Blastomeres for Single Cell Surgical Applications
    Abu Zanouneh, Basil
    Mills, James K.
    PROCEEDINGS OF 2022 IEEE INTERNATIONAL CONFERENCE ON MECHATRONICS AND AUTOMATION (IEEE ICMA 2022), 2022, : 1197 - 1203
  • [47] Single Image Based Three-Dimensional Scene Reconstruction Using Semantic and Geometric Priors
    Gang-Joon Yoon
    Jinjoo Song
    Yu-Jin Hong
    Sang Min Yoon
    Neural Processing Letters, 2022, 54 : 3679 - 3694
  • [48] Selection of three-dimensional garment image style based on three-dimensional body shape image classification of young women
    Masuda, Tomoe
    Murakami, Kaori
    Okabe, Hidehiko
    Nishi, Mikako
    SEN-I GAKKAISHI, 2007, 63 (12) : 277 - 286
  • [49] A fast three-dimensional reconstruction method applied for the fabric defect detection
    Song, Limei
    Zhang, Chunbo
    Xiong, Hui
    Wei, Yiying
    Chen, Huawei
    OPTICAL METROLOGY AND INSPECTION FOR INDUSTRIAL APPLICATIONS, 2010, 7855
  • [50] Three-Dimensional Residual Neural Architecture Search for Ultrasonic Defect Detection
    McKnight, Shaun
    MacKinnon, Christopher
    Pierce, S. Gareth
    Mohseni, Ehsan
    Tunukovic, Vedran
    MacLeod, Charles N.
    Vithanage, Randika K. W.
    O'Hare, Tom
    IEEE TRANSACTIONS ON ULTRASONICS FERROELECTRICS AND FREQUENCY CONTROL, 2024, 71 (03) : 423 - 436