Fruit Image Segmentation Based on a Colour Perception Neural Network Inspired by the Retina Structure

DOI

10.2991/aiie-15.2015.9How to use a DOI?

Keywords

fruit images segmentation; retina structure; multiple colour space; neural network

Abstract

Fruit images segmentation is one of the important issues for fruit recognition. Inspired by the retina structure, a colour perception neural network is proposed to extract multiple colour features, and is applied to tomato fruits segmentation under natural complex background. The network is composed three layers. Input layer is colour perception layer inspired by the retina. The multiple colour space features based on retina structure and gray information are extracted. The middle layer is a back propagation neural network. The output layer is the segmented image. The experimental results show that the network can be used to well segment mature or immature tomato fruits from colour images with complex background, and eliminate the holes of segmentation image due to fruit surface refection. The proposed network is promising to be applied to fruit recognition of robots.

Copyright

© 2015, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Q.Y. Sun
AU  - Q.X. Wu
AU  - X. Wang
AU  - L. Hou
PY  - 2015/07
DA  - 2015/07
TI  - Fruit Image Segmentation Based on a Colour Perception Neural Network Inspired by the Retina Structure
BT  - Proceedings of the 2015 International Conference on Artificial Intelligence and Industrial Engineering
PB  - Atlantis Press
SP  - 29
EP  - 32
SN  - 1951-6851
UR  - https://doi.org/10.2991/aiie-15.2015.9
DO  - 10.2991/aiie-15.2015.9
ID  - Sun2015/07
ER  -