A front-end technique for visual gold detection and localization - Towards automation of the gold panning process.

Blessing Chipfurwe Makoni, Molaletsa Namoshe, Oduetse Matsebe

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

To do away with the time consuming, error prone and expert dependent gold detection and localization in the current gold panning procedure, automatic image-based techniques are considered in this paper. Identifying and locating gold particles during the gold panning process is a fundamental process in gold extraction. In the automation of the gold panning process and robotic handling it is important to identify and locate the gold particles in the images captured by the image sensor. Image segmentation is a vital step in image simplification, image understanding and object detection. Image segmentation is the process of identifying and extracting homogeneous regions (segments) in an image satisfying a homogeneity criterion based on features formulated from spectral components of the image. Three image thresholding techniques were tested and evaluated on sample gold panning images. Color image thresholding in the CIELAB color space performed better in detecting and locating gold particles in an image. The proposed method will serve as a front-end technique for an automated gold panning system as it will automate the visual feature identification of gold particles and aid in the control of the handling system of the gold particles during the panning process.

Original languageEnglish
Title of host publication2021 Southern African Universities Power Engineering Conference/Robotics and Mechatronics/Pattern Recognition Association of South Africa, SAUPEC/RobMech/PRASA 2021
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9780738112367
DOIs
Publication statusPublished - Jan 27 2021
Event2021 Southern African Universities Power Engineering Conference/Robotics and Mechatronics/Pattern Recognition Association of South Africa, SAUPEC/RobMech/PRASA 2021 - Virtual, Potchefstroom, South Africa
Duration: Jan 27 2021Jan 29 2021

Publication series

Name2021 Southern African Universities Power Engineering Conference/Robotics and Mechatronics/Pattern Recognition Association of South Africa, SAUPEC/RobMech/PRASA 2021

Conference

Conference2021 Southern African Universities Power Engineering Conference/Robotics and Mechatronics/Pattern Recognition Association of South Africa, SAUPEC/RobMech/PRASA 2021
Country/TerritorySouth Africa
CityVirtual, Potchefstroom
Period1/27/211/29/21

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence
  • Computer Vision and Pattern Recognition
  • Energy Engineering and Power Technology
  • Electrical and Electronic Engineering
  • Mechanical Engineering

Fingerprint

Dive into the research topics of 'A front-end technique for visual gold detection and localization - Towards automation of the gold panning process.'. Together they form a unique fingerprint.

Cite this