Projection-dependent input processing for 3D object recognition in human robot interaction systems

Abstract Human-Robot Interaction (HRI) provides assisted services in different real-time applications. The robotic systems identify objects through digital visualization wherein a three-dimensional (3D) image is converged to a plane-based projection. The projection is analyzed using the co-ordinates and identification points for recognizing the object. In such a converging process, the misidentification of projections in different planes results in recognition errors. This article proposes projection-dependent input processing (PDIP) method to reduce the misidentifications in object recognition. In this method, the input is the visualizing image projected in all the possible dimensions to identify the conjoining indices. The conjoined indices without intersection are segregated using labeled analysis. The non-correlating indices are identified in the possible dimension projections to prevent errors. The deviations in planes and indices matching are prevented by correlating the input with similar stored inputs with labels. The proposed method is verified using the metrics recognition ratio (96.4%), time (630.36 ms), complexity (5.93), and error (0.605).

[1]  Omar Y. Al-Jarrah,et al.  A Survey on 3D Object Detection Methods for Autonomous Driving Applications , 2019, IEEE Transactions on Intelligent Transportation Systems.

[2]  P. Mohamed Shakeel,et al.  Fuzzy rule-based environment-aware autonomous mobile robots for actuated touring , 2020, Intelligent Service Robotics.

[3]  Juan Zhang,et al.  3D Object Detection Incorporating Instance Segmentation and Image Restoration , 2019, Wuhan University Journal of Natural Sciences.

[4]  Yongbin Gao,et al.  3D Object Detection Based on Vanishing Point and Prior Orientation , 2019, Wuhan University Journal of Natural Sciences.

[5]  Abdullatif Baba.  A new design of a flying robot, with advanced computer vision techniques to perform self-maintenance of smart grids , 2020, ArXiv.

[6]  Thomas Probst,et al.  Automatic Tool Landmark Detection for Stereo Vision in Robot-Assisted Retinal Surgery , 2018, IEEE Robotics and Automation Letters.

[7]  Qingquan Li,et al.  3D map-guided single indoor image localization refinement , 2020 .

[8]  Guillermo L. Grinblat,et al.  Online Object Detection and Localization on Stereo Visual SLAM System , 2019, Journal of Intelligent & Robotic Systems.

[9]  Chi-Yi Tsai,et al.  Simultaneous 3D Object Recognition and Pose Estimation Based on RGB-D Images , 2018, IEEE Access.

[10]  Dianwei Wang,et al.  3D Object Detection Algorithm for Panoramic Images With Multi-Scale Convolutional Neural Network , 2019, IEEE Access.

[11]  Chen Wang,et al.  Research on the effect of image size on real-time performance of robot vision positioning , 2018, EURASIP J. Image Video Process..

[12]  Fabiano Hessel,et al.  Machine-learning-based system for multi-sensor 3D localisation of stationary objects , 2018, IET Cyper-Phys. Syst.: Theory & Appl..

[13]  Timothy Patten,et al.  Addressing the Sim2Real Gap in Robotic 3-D Object Classification , 2019, IEEE Robotics and Automation Letters.

[14]  C.-C. Jay Kuo,et al.  Context-Assisted 3D (C3D) Object Detection from RGB-D Images , 2018, J. Vis. Commun. Image Represent..

[15]  Erik B. Sudderth,et al.  Clouds of Oriented Gradients for 3D Detection of Objects, Surfaces, and Indoor Scene Layouts , 2019, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[16]  Hao Li,et al.  Large-scale 3D Semantic Mapping Using Stereo Vision , 2018, Int. J. Autom. Comput..

[17]  Xiaoliang Ma,et al.  3D object detection based on sparse convolution neural network and feature fusion for autonomous driving in smart cities , 2020 .

[18]  Xing Li,et al.  Human-robot interaction based on gesture and movement recognition , 2020, Signal Process. Image Commun..

[19]  Geng Yang,et al.  cGAN Based Facial Expression Recognition for Human-Robot Interaction , 2019, IEEE Access.

[20]  Wei Huang,et al.  Efficient Planar Surface-Based 3D Mapping Method for Mobile Robots Using Stereo Vision , 2019, IEEE Access.

[21]  Shengyong Chen,et al.  Recent Advances in 3D Data Acquisition and Processing by Time-of-Flight Camera , 2019, IEEE Access.

[22]  Xavier Savatier,et al.  A Vision-Based System for Robot Localization in Large Industrial Environments , 2019, Journal of Intelligent & Robotic Systems.

[23]  J. Gamez Garcia,et al.  3D object recognition for anthropomorphic robots performing tracking tasks , 2019 .

[24]  Yu Wang,et al.  3D Integral Imaging Display Processing Using the Similarity of Corresponding Points in Axially Recorded Images , 2020 .

[25]  Soon-Geul Lee,et al.  3D Localization of a Mobile Robot by Using Monte Carlo Algorithm and 2D Features of 3D Point Cloud , 2020, International Journal of Control, Automation and Systems.

[26]  Ilan Halachmi,et al.  3D Computer-vision system for automatically estimating heifer height and body mass , 2017, Biosystems Engineering.

[27]  Vincent Lepetit,et al.  Robust 3D Object Tracking from Monocular Images Using Stable Parts , 2018, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[28]  Miao Li,et al.  Robotic grasp detection based on image processing and random forest , 2019, Multimedia Tools and Applications.

[29]  Jie Zhang,et al.  Instance-based object recognition in 3D point clouds using discriminative shape primitives , 2017, Machine Vision and Applications.

[30]  Ling Shao,et al.  3D object detection: Learning 3D bounding boxes from scaled down 2D bounding boxes in RGB-D images , 2019, Inf. Sci..