Articles
COMPUTER VISION SYSTEM FOR ON-LINE SORTING OF POT PLANTS BASED ON LEARNING TECHNIQUES
Article number
421_8
Pages
91 – 98
Language
Abstract
ATO-DLO has built a flexible and universal sorting system for pot plants.
The system contains a colour camera, image processing hardware and specially developed software.
It can be applied for sorting of several types of pot plants because of the implementation of learning techniques using statistical discriminant analysis and a neural network classifier.
Two types of experiments are described for partially and full automated sorting of flowering pot plants.
With the basic configuration determination of flower colour and measurement of size and flowering stage is performed.
For complete automated sorting also recognition of bad quality plants is necessary.
For recognising B-quality plants a procedure of learning by example with an artificial neural network classifier gives good results.
The system contains a colour camera, image processing hardware and specially developed software.
It can be applied for sorting of several types of pot plants because of the implementation of learning techniques using statistical discriminant analysis and a neural network classifier.
Two types of experiments are described for partially and full automated sorting of flowering pot plants.
With the basic configuration determination of flower colour and measurement of size and flowering stage is performed.
For complete automated sorting also recognition of bad quality plants is necessary.
For recognising B-quality plants a procedure of learning by example with an artificial neural network classifier gives good results.
Authors
A.J.M. Timmermans
Keywords
Image processing, discriminant analysis, artificial neural networks, saintpaulia, rosa
Online Articles (31)