Application of a fuzzy classification technique in computer grading of fish products
IEEE Transactions on Fuzzy SystemsPublished 1 January 1998
Bao-Gang Hu, Raymond G. Gosine, Liang Cao, C.W. de Silva
Citations48
SJR quartileQ1
SJR score3.61
SNIP2.80
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
The enhancement and application of a fuzzy classification technique for automated grading of fish products and the method presented here provides a general capability for both grading and sorting-type problems in food processing.
Abstract
This work presents the enhancement and application of a fuzzy classification technique for automated grading of fish products. Common features inherent in grading-type data and their specific requirements in processing for classification are identified. A fuzzy classifier with a four-level hierarchy is developed based on the "generalized K-nearest neighbor rules". Both conventional and fuzzy classifiers are examined using a realistic set of herring roe data (collected from the fish processing industry) to compare the classification performance in terms of accuracy and computational cost. The classification results show that the generalized fuzzy classifier provides the best accuracy at 89%. The grading system can be tuned through two parameters-the threshold of fuzziness and the cost weighting of error types-to achieve higher classification accuracy. An optimization scheme is also incorporated into the system for automatic determination of these parameter values with respect to a specific optimization function that is based on process renditions, including the product price and labor cost. Since the primary common features are accommodated in the classification algorithm, the method presented here provides a general capability for both grading and sorting-type problems in food processing.
Keywords
Computer ScienceAgricultural and Biological SciencesEngineering
Elsevier eBooksIntroduction to Statistical Pattern Recognition
11,121 Citations1990Keinosuke Fukunaga
A system to determine the sex of fish from their length, as accurately as possible, using a set of labelled data, together with the class label (male or female), as a training set.
IEEE Transactions on Systems Man and CyberneticsA fuzzy K-nearest neighbor algorithm
2,614 Citations1985James M. Keller, Michael Gray +1 more
The theory of fuzzy sets is introduced into the K-nearest neighbor technique to develop a fuzzy version of the algorithm, and three methods of assigning fuzzy memberships to the labeled samples are proposed.
IEEE Transactions on Neural NetworksFuzzy min-max neural networks. I. Classification
783 Citations1992P.K. Simpson
The fuzzy min-max classifier neural network implementation is explained, the learning and recall algorithms are outlined, and several examples of operation demonstrate the strong qualities of this new neural network classifier.
Choice Reviews OnlineDecision, estimation, and classification: an introduction to pattern recognition and related topics
294 Citations1989
Probability theory for random vectors simple statistical decision procedures operations upon random vectors feature extraction and nonlinear mapping quadratic and linear classifiers parameter estimation nonparametric estimation and classification estimating and bounding the probability of error classification of stationary time series context-dependent methods other methods of classification.
Fuzzy Sets and SystemsGeneralized k-nearest neighbor rules
161 Citations1986James C. Bezdek, Siew K. Chuah +1 more
All four designs yield comparable (usually within 4%) error rates; the Fuzzy c-Means (FCM) based k-NNR is usually the best design; the FCM/1-NPR is the most efficient and perhaps most useful of the four designs; and finally, that generalized NNR's are an important and useful extension of the conventional ones.
Transactions of the ASAECOLOR AND DEFECT SORTING OF BELL PEPPERS USING MACHINE VISION
80 Citations1990S. A. Shearer, F. A. Payne
A machine vision algorithm for grading of fresh market produce according to color and damage was developed by treating the relative hue distribution of pixels in six orthogonal views as quantitative variables and discriminant analysis was used to classify observations.
Transactions of the ASAEApple Sorting with Machine Vision
73 Citations1986G. E. Rehkugler, J. A. Throop
Computers and Electronics in AgricultureLength measurement of fish by computer vision
63 Citations1993Norval J. C. Strachan
A machine for the length measurement of fish by computer vision that can cater for gutted and flexed fish, and has a throughput of greater than one fish per second.
Transactions of the ASAEPEACH DEFECT DETECTION WITH MACHINE VISION
60 Citations1991B. K. Miller, Michael J. Delwiche
A laboratory machine vision system was developed to detect and identify surface defects on fresh market peaches and classifier performance in identifying each segmented region as a member of one of eight classes.
Transactions of the ASAEComputer Vision Based System for Quality Separation of Fresh Market Tomatoes
56 Citations1985Nandan Sarkar, R. R. Wolfe
Anachine vision based inspection scheme for tomatoes which involves the analysis of stem and blossom end views for features indicative of fresh market quality and the flexibility of this package in implementing various grading schemes according to the needs of the marketplace is demonstrated.
Transactions of the ASAEMachine Vision for Crack Detection in Rotating Eggs
46 Citations1992John W. Goodrum, R. T. Elster
Continuously rotating eggs were inspected for cracks by a PC-based computer vision system and response surface analysis showed that inspection success rate was dependent on egg size and two software calibration constants.
Computers and Electronics in AgricultureClassification of apple surface features using machine vision and neural networks
45 Citations1993Qisen Yang
A neural network approach was proposed to perform the classification of apple surface features, namely non-defective area, patch-like blemish, and elongated blemishes, and an average classification accuracy of 96.6%.
Transactions of the ASAERaisin Grading by Machine Vision
36 Citations1993Naoyasu Okamura, Michael J. Delwiche +1 more
A machine vision system for grading raisins was developed, including an imaging test stand and image analysis algorithms, which identified wrinkle edge density, average gradient magnitude, angularity, and elongation as key features of raisin maturity.
Transactions of the ASAEFat Estimation in Beef Ultrasound Images Using Texture and Adaptive Logic Networks
24 Citations1994John McCauley, B. R. Thane +1 more
Journal of Agricultural Engineering ResearchEstimation of tip shape for carrot classification by machine vision
22 Citations1992M. S. Howarth, J.R. Brandon +2 more
Tip shape has been identified as an important carrot feature which is a major concern to both consumers and in post harvest operations and a classification method was developed which classified carrot tips into five classes (sharp tapered to extremely blunt tips).
Transactions of the ASAEMaturity Detection in Peanuts (Arachis Hypogaea L.) Using Machine Vision
21 Citations1993S. R. Ghate, Margaret D. M. Evans +2 more
A descriptor derived from the gradient image of a peanut was found to be most useful in identifying peanuts of the mid-mature and immature groups in three market size classifications when either white background or black background and red filter conditions were used.
Transactions of the ASAEDetermining Orientation and Shape of Bell Peppers by Machine Vision
19 Citations1987R. R. Wolfe, Madhavan Swaminathan
ABSTRACTAprocedure was developed for computing the orienta-tion of bell peppers using the Hough transform on six orthogonal views to locate stem and blossom end centers using image analysis algorithms.
Engineering Applications of Artificial IntelligenceFlexible automation of fish processing
17 Citations1993C.W. De Silva, Raymond G. Gosine +3 more
Robotics, machine vision, and intelligent control have been integrated in the technology development for flexible automation of fish processing, and an automated system for grading herring roe is being built.
AutomaticaClassification of fruits by a boltzmann perceptron neural network
10 Citations1992Uri Ben-Hanan, Kalman Peleg +1 more
Simulated sorting experiments of apples, into “red” and “green” categories showed that the system can emulate the decisions of a panel of human sorters quite well, indicating that the BPN can create intricate non-linear discriminant functions, even when the classes are noncontinguous and the training set is relatively small.
Artificial intelligence application in carcass beef grading automation
4 Citations2002Yumin Chen, Timothy McDonald
This paper presents the technical aspects of developing an automated carcass beef grading system at the US Meat Animal Research Center at Clay Center, NE.