Natural Object Recognition Using the Combination of Gaussian Model and Region Growing

2009 ◽  
Author(s):  
Yongmei Cheng ◽  
Yanru Wu ◽  
Lihua Yang ◽  
Chunhui Zhao ◽  
Shaowu Zhang
Keyword(s):  
Object Recognition ◽  
Gaussian Model ◽  
Region Growing ◽  
Natural Object

Point Cloud Based Bin Picking: Object Recognition and Pose Estimation Using Region Growing Segmentation Algorithm

2015 ◽  
Vol 791 ◽  
pp. 189-194
Author(s):  
Frantisek Durovsky
Keyword(s):  
Object Recognition ◽  
Pose Estimation ◽  
Point Cloud ◽  
Region Growing ◽  
Segmentation Algorithm ◽  
Suction Cup ◽  
Bin Picking ◽  
Region Growing Segmentation ◽  
Dual Arm ◽  
Sensor Region

Presented paper describes experimental bin picking using Kinect sensor, region-growing algorithm, latest ROS-Industrial drivers and dual arm manipulator Motoman SDA10f.As well known if manipulation with objects of regular shapes by suction cup is required, it is sufficient to estimate only 5DoF for successful pick. In such a case simpler region growing algorithm may be used instead of complicated 3D object recognition and pose estimation techniques, resulting in shorter processing time and decrease of computational load. Experimental setup for such a scenario, manipulated objects and results using region growing segmentation algorithm are explained in detail. Video and link to open-source code of described application are provided as well.

Fast Algorithms of Multi-Object Recognition and High Precision Localization for Pose Estimation

2013 ◽  
Vol 333-335 ◽  
pp. 1192-1197
Author(s):  
Ying Jin Zhang ◽  
Shi Yin Qin ◽  
Xiao Hui Hu
Keyword(s):  
Object Recognition ◽  
Pose Estimation ◽  
Median Filter ◽  
Region Growing ◽  
Random Errors ◽  
Bilinear Interpolation ◽  
Visual Measurement ◽  
Accurate Localization ◽  
Autonomous Rendezvous ◽  
Rendezvous And Docking

The recognition and localization of cooperative objects are very important for spacecraft pose estimation towards autonomous rendezvous and docking (RVD). In this paper, an adaptive threshold segmentation algorithm is proposed base on weighted maximum gray value for multi-object detection, and eight-neighborhood region growing is employed for multi-object recognition. In order to achieve high-accurate localization, a sub-pixel object positioning approach is put forward by combination bilinear interpolation with median filtering, which employs bilinear interpolation to calculate sub-pixel centroid for reducing algorithm systematic errors, and applies median filter to reduce random errors produced by image noises. The experimental results show that the proposed algorithms are feasible and effective with high positioning accuracy of 0.01 pixels, and have outstanding advantages of anti-disturbance and real-time performance, thus can satisfy the practical requirements in the visual measurement and pose estimation of cooperative objects for the RVD in space exploration.

Based on the Human Eye Fatigue Detection System of Recognition and FPGA Technology

2014 ◽  
Vol 672-674 ◽  
pp. 1995-1998 ◽  
Author(s):  
Jun Wang ◽  
Qiang Liu ◽  
Hang Zhao
Keyword(s):  
High Speed ◽  
Detection System ◽  
A Priori ◽  
Color Space ◽  
Gaussian Model ◽  
Region Growing ◽  
Detection Algorithm ◽  
Detection Accuracy ◽  
Threshold Detection ◽  
Human Eye

This paper introduces the field programmable gate array (fpga) application in high-speed visual inspection system.ALTERA EP1K30QC208-2 are used in the system for system calculation and control of the core, to perform high-speed real-time visual detection algorithm, this paper adopts a yawning based on eye closure and to detect driver fatigue, the method of in YCr, Cb in the color space using gaussian model skin detection of human face area, in the face of a gray binarization figure using a priori knowledge of facial features geometry in rough positioning the human eye, eye contour are obtained by region growing and morphological operation and calculation of the closure of the eye;Best threshold detection lips when using lip color roughly locate the lips, on the basis of accurate positioning lips by face grey value characteristics, and then through the mouth level to determine whether a driver yawn;Based on the two characteristics of driving fatigue, experiments show that this system detection speed, excellent versatility, and can improve the detection accuracy.

Sign in / Sign up

Export Citation Format

Share Document