Vision System for Quality Assessment of Robotic Cleaning of Fish Processing Plants Using CNN

A vision system has been developed for automatic quality assessment of robotic cleaning of fish processing lines. The quality assessment is done by detecting residual fish blood on cleaned surfaces. The system is based on classification using convolutional neural networks (CNNs). The performance of...

Full description

Saved in:
Bibliographic Details
Published inIEEE access Vol. 7; pp. 71675 - 71685
Main Authors Bjorlykhaug, Emil, Egeland, Olav
Format Journal Article
LanguageEnglish
Published Piscataway IEEE 2019
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects
Aquaculture
Artificial neural networks
Blood
Cleaning
Computer architecture
computer vision
Fish
Image color analysis
machine learning
Machine vision
Neural networks
Production
Quality assessment
robot vision system
Training
Training data
Vision systems
Online AccessGet full text

Cover

More Information
Summary:A vision system has been developed for automatic quality assessment of robotic cleaning of fish processing lines. The quality assessment is done by detecting residual fish blood on cleaned surfaces. The system is based on classification using convolutional neural networks (CNNs). The performance of different convolutional neural network architectures and parameters is evaluated. The datasets that simulate various conditions in fish processing plants are generated using data augmentation techniques. Tests using further augmented training data to increase the performance of the neural network are performed, which results in a substantial increase in performance both compared to the color thresholding technique and the same neural network architecture without augmented training data. The performance of the system is validated in experiments in an industrial setting.
ISSN:2169-3536
2169-3536
DOI:10.1109/ACCESS.2019.2919656