Contour Identification method for human-robot communication

Published: 3 August 2019| Version 1 | DOI: 10.17632/dx6syd75z9.1
, Piotr Skrobek


The subject of research was task-oriented communication between humans and collaborative robots in industrial scenario Human-robot interface must be supported by image recognition system in order to identify objects referred to by operator A novel method for contour classification based on flexible editable contour templates (FECT) was developed Practical result was a tool facilitating customization of image recognition systems Verification of the method was done through the laboratory implementation


Steps to reproduce

Data presented in Word (Office) but any software for text editing can be used


Computer Vision, Speaker Recognition