Shape-Based Alignment of the Scanned Objects Concerning Their Asymmetric Aspects

We introduce an integrated method for processing depth maps measured by a laser profile sensor. It serves for the recognition and alignment of an object given by a single example. Firstly, we look for potential object contours, mainly using the Retinex filter. Then, we select the actual object boundary via shape comparison based on Triangle Area Representation (TAR). We overcome the limitations of the TAR method by extension of its shape descriptor. That is helpful mainly for objects with symmetric shapes but other asymmetric aspects like squares with asymmetric holes. Finally, we use point-to-point pairing, provided by the extended TAR method, to calculate the 3D rigid affine transform that aligns the scanned object to the given example position. For the transform calculation, we design an algorithm that overcomes the Kabsch point-to-point algorithm’s accuracy and accommodates it for a precise contour-to-contour alignment. In this way, we have implemented a pipeline with features convenient for industrial use, namely production inspection.

[1]  Hui Wei,et al.  Scale-invariant contour segment context in object detection , 2014, Image Vis. Comput..

[2]  Rafal Doroz,et al.  The New Method of Signature Recognition Based on Least Squares Contour Alignment , 2009, 2009 International Conference on Biometrics and Kansei Engineering.

[3]  T. W. Ridler,et al.  Picture thresholding using an iterative selection method. , 1978 .

[4]  Qiu Yuanying,et al.  Precise on-line non-target pose measurement for cylindrical components based on laser scanning , 2019, Assembly Automation.

[5]  Ryan B. Wicker,et al.  Part re-registration during process interruption of electron beam melting additive manufacturing , 2018 .

[6]  Kun Liu,et al.  Image-Alignment Based Matching for Irregular Contour Defects Detection , 2018, IEEE Access.

[7]  Jianyu Yang,et al.  Contour Based Shape Matching for Object Recognition , 2016, ICIRA.

[8]  Georgios D. Evangelidis,et al.  Parametric Image Alignment Using Enhanced Correlation Coefficient Maximization , 2008, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[9]  Wanda Benesova,et al.  3D object recognition based on local descriptors , 2015, Electronic Imaging.

[10]  Naif Alajlan,et al.  Fast shape matching and retrieval based on approximate dynamic space warping , 2010, Artificial Life and Robotics.

[11]  Dongming Gan,et al.  Integrating laser profile sensor to an industrial robotic arm for improving quality inspection in manufacturing processes , 2020 .

[12]  Benjamin Drayer.  Object alignment and detection , 2017 .

[13]  R.M. Luque,et al.  Object recognition and inspection in difficult industrial environments , 2006, 2006 IEEE International Conference on Industrial Technology.

[14]  Jianyu Yang,et al.  A hybrid shape descriptor for object recognition , 2015, 2015 IEEE International Conference on Robotics and Biomimetics (ROBIO).

[15]  Jochen Deuse,et al.  Predictive model-based quality inspection using Machine Learning and Edge Cloud Computing , 2020, Adv. Eng. Informatics.

[16]  T. Gasser,et al.  Alignment of curves by dynamic time warping , 1997 .

[17]  Cordelia Schmid,et al.  3D Object Modeling and Recognition Using Local Affine-Invariant Image Descriptors and Multi-View Spatial Constraints , 2006, International Journal of Computer Vision.

[18]  Changyun Miao,et al.  On-line conveyor belts inspection based on machine vision , 2014 .

[19]  Naif Alajlan,et al.  Shape retrieval using triangle-area representation and dynamic space warping , 2007, Pattern Recognit..

[20]  Mathieu Aubry,et al.  Representing 3D models for alignment and recognition , 2015 .

[21]  Helin Dutagaci,et al.  A method for determining local coordinate frames attached to objects by a laser profile sensor in industrial robot workspaces , 2017, 2017 IEEE International Conference on Autonomous Robot Systems and Competitions (ICARSC).

[22]  B Karthikeyan,et al.  Shape, Color and Texture Based CBIR System Using Fuzzy Logic Classifier , 2019, Computers, Materials & Continua.

[23]  Ashraf A. Kassim,et al.  A comparative study of efficient generalised Hough transform techniques , 1999, Image Vis. Comput..

[24]  Bugao Xu,et al.  Shape recognition of profiled fibers using triangle-area representation , 2013 .

[25]  Young Soo Suh,et al.  Laser Sensors for Displacement, Distance and Position , 2019, Sensors.

[26]  Sasan Mahmoodi,et al.  Least-Squares Contour Alignment , 2009, IEEE Signal Processing Letters.

[27]  S. Roth,et al.  Laser , 2019, Optik.

[28]  Matti Pietikäinen,et al.  Adaptive document image binarization , 2000, Pattern Recognit..

[29]  Keiichi Abe,et al.  Topological structural analysis of digitized binary images by border following , 1985, Comput. Vis. Graph. Image Process..

[31]  C. D. Kuglin,et al.  The phase correlation image alignment method , 1975 .

[32]  Suprava Patnaik,et al.  Image Registration Using Log Polar Transform and Phase Correlation to Recover Higher Scale , 2012 .

[33]  Jean-Michel Morel,et al.  Multiscale Retinex , 2014, Image Process. Line.

[34]  Guoyou Wang,et al.  Contour descriptor based on space symmetry and its matching technique , 2013 .

[35]  唐莉萍,et al.  Triangle Area Representation for Profiled Fiber Shape , 2010 .

[36]  Li Shang,et al.  Shape matching and object recognition using common base triangle area , 2015, IET Comput. Vis..

[37]  W. Kabsch.  A solution for the best rotation to relate two sets of vectors , 1976 .

[38]  Gary R. Bradski,et al.  ORB: An efficient alternative to SIFT or SURF , 2011, 2011 International Conference on Computer Vision.

[39]  Rama Chellappa,et al.  An Efficient and Robust Algorithm for Shape Indexing and Retrieval , 2010, IEEE Transactions on Multimedia.

[40]  D. M. Potukuchi,et al.  Design of a Novel Shape Signature by Farthest Point Angle for Object Recognition , 2014 .