Xinyi Wei, Benjamin M Bohrer, B. Uttaro, Manuel Juárez
{"title":"Centre pork chop colour classification using image analysis on the ventral surface of the loin","authors":"Xinyi Wei, Benjamin M Bohrer, B. Uttaro, Manuel Juárez","doi":"10.1139/cjas-2023-0011","DOIUrl":null,"url":null,"abstract":"Image analysis of ventral, anterior, and posterior ends of 550 pork loins was used to classify extreme-coloured centre chops. Only the ventral side was strongly correlated ( r = 0.79) to the centre chop. Thresholds at 2.5th, 5th, and 7.5th percentiles of ventral surface pale pork were selected to classify centre chop pale pork (5th percentile). Increasing stringency increased accuracy in excluding pale samples. At the 7.5th percentile, prediction accuracy of pale (85.7%) and non-pale samples (89.5%) was balanced. This method can be adopted by commercial pork processors as a classification or pre-sorting approach.","PeriodicalId":9512,"journal":{"name":"Canadian Journal of Animal Science","volume":"3 12","pages":"0"},"PeriodicalIF":1.2000,"publicationDate":"2023-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Canadian Journal of Animal Science","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1139/cjas-2023-0011","RegionNum":4,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"AGRICULTURE, DAIRY & ANIMAL SCIENCE","Score":null,"Total":0}
引用次数: 0
Abstract
Image analysis of ventral, anterior, and posterior ends of 550 pork loins was used to classify extreme-coloured centre chops. Only the ventral side was strongly correlated ( r = 0.79) to the centre chop. Thresholds at 2.5th, 5th, and 7.5th percentiles of ventral surface pale pork were selected to classify centre chop pale pork (5th percentile). Increasing stringency increased accuracy in excluding pale samples. At the 7.5th percentile, prediction accuracy of pale (85.7%) and non-pale samples (89.5%) was balanced. This method can be adopted by commercial pork processors as a classification or pre-sorting approach.
期刊介绍:
Published since 1957, this quarterly journal contains new research on all aspects of animal agriculture and animal products, including breeding and genetics; cellular and molecular biology; growth and development; meat science; modelling animal systems; physiology and endocrinology; ruminant nutrition; non-ruminant nutrition; and welfare, behaviour, and management. It also publishes reviews, letters to the editor, abstracts of technical papers presented at the annual meeting of the Canadian Society of Animal Science, and occasionally conference proceedings.