Title :
Identification method of waste based on gray level co-occurrence matrix and neural network
Author :
Kun, Wang ; Songtao, Kong
Author_Institution :
Sch. of Mech. & Power Eng., Chongqing Univ. of Sci. & Technol., Chongqing, China
Abstract :
According gray level co-occurrence matrix (glcm) has a good ability to express the characteristics of texture, a waste identification method based on glcm and probabilistic neural network was proposed. The method obtains waste images from the refuse conveyor belt by high-speed camera system, after image preprocessing, extracts the texture features -glcm, then trains the neural network with the glcm as samples, and waste intelligent identification was realized, lays the foundation for automatic classification of waste and provides a new harmless, reduction and resource way for the garbage disposal.
Keywords :
belts; cameras; conveyors; feature extraction; image texture; incineration; industrial waste; matrix algebra; neural nets; power engineering computing; refuse disposal; garbage disposal reduction; glcm; gray level cooccurrence matrix; high-speed camera system; image preprocessing; probabilistic neural network; refuse conveyor belt; texture feature extraction; waste identification method; waste images; waste intelligent identification; glcm; identification method; neural network;
Conference_Titel :
Materials for Renewable Energy & Environment (ICMREE), 2011 International Conference on
Conference_Location :
Shanghai
Print_ISBN :
978-1-61284-749-8
DOI :
10.1109/ICMREE.2011.5930954