Acta Veterinaria et Zootechnica Sinica ›› 2024, Vol. 55 ›› Issue (9): 3843-3852.doi: 10.11843/j.issn.0366-6964.2024.09.010

• Animal Genetics and Breeding • Previous Articles     Next Articles

Development of a Pig Intramuscular Fat Content and Eye Muscle Area Measurement System Based on Computer Vision Technology

Dong CHEN1,2,3(), Wenxuan ZHOU1,2,3, Zhenjian ZHAO1,2,3, Qi SHEN1,2,3, Yang YU1,2,3, Shengdi CUI1,2,3, Junge WANG1,2,3, Ziyang CHEN1,2,3, Shixin YU1,2,3, Jiamiao CHEN1,2,3, Xiangfeng WANG1,2,3, Pingxian WU4, Zongyi GUO4, Jinyong WANG4, Guoqing TANG1,2,3,*()   

  1. 1. State Key Laboratory of Swine and Poultry Breeding Industry, College of Animal Science and Technology, Sichuan Agricultural University, Chengdu 611130, China
    2. Key Laboratory of Livestock and Poultry Multi-omics of Ministry of Agriculture and Rural Affairs, College of Animal Science and Technology, Sichuan Agricultural University, Chengdu 611130, China
    3. Farm Animal Genetic Resources Exploration and Innovation Key Laboratory of Sichuan Province, Sichuan Agricultural University, Chengdu 611130, China
    4. National Center of Technology Innovation for Pigs, Chongqing 402460, China
  • Received:2024-03-18 Online:2024-09-23 Published:2024-09-27
  • Contact: Guoqing TANG E-mail:1123278154@qq.com;tyq003@163.com

Abstract:

The aim of this study was to develop a rapid identification method of pork quality images based on computer vision technology, and to lay a technical foundation for solving the defects of traditional pork quality determination methods and constructing a rapid pork quality determination system. In this study, canny edge detection algorithm, normalization, gamma correction, adaptive histogram equalization and thresholding were used for image processing to simulate a three-dimensional model of porcine dorsal longest muscle using two-dimensional information, and an image-based computer vision method for determining porcine intramuscular fat content was developed. An image-processing-based eye muscle area recognition method was developed using manual depiction and computer vision contour search techniques. The underlying algorithmic programs of each method were implemented using Open CV and C++, and a system for the determination of porcine intramuscular fat content and eye muscle area was developed and tested using 250 test pigs of different breeds. The test results showed that the intramuscular fat content detection module could quickly realize the extraction and prediction of intramuscular fat graphical information, with an average detection time of no more than 2 minutes for a single sample, an accuracy of up to 0.54, and a discrete coefficient of 0.062, which was highly stable. The eye muscle area and backfat thickness detection module can quickly determine eye muscle area and backfat thickness by combining manual and computerized methods. The assay method and software system developed in this study enable rapid detection of intramuscular fat content, backfat thickness and eye muscle area in pigs.

Key words: computer vision, image processing, intramuscular fat content, pig

CLC Number: