Food and Fermentation Industries >

2024 , Vol. 50 >Issue 8: 262 - 270

DOI: https://doi.org/10.13995/j.cnki.11-1802/ts.037376

Multi-model quantitative analysis of beef freshness judgment by machine vision

  • LU Zhongchao ,
  • QIU Yue ,
  • ZHANG Anqiang ,
  • ZHANG Jianyou ,
  • CUI Pengbo ,
  • XIANG Yun ,
  • JIN Xia ,
  • LYU Fei
Expand
  • 1(College of Food Science and Technology, Zhejiang University of Technology, Hangzhou 310014, China)
    2(Institute of Cyberspace Security, Zhejiang University of Technology, Hangzhou 310014, China)
    3(Hangzhou Guanhuawang Food Company Limited, Hangzhou 310014, China)

Received date: 2023-09-14

  Revised date: 2023-10-16

  Online published: 2024-06-11

Fold

Abstract

Color is one of the most important indicators for identifying the freshness of fresh beef.With the development of artificial intelligence, machine vision provides a quantifiable solution for fresh meat freshness identification.This study acquired beef images by smartphone and used greyscale, binarization, and image segmentation and extraction to obtain Red-Green-Blue (RGB), L*a*b*, and Hue-Saturation-Intensity (HSI) color parameters of beef images with different freshness levels.The color parameters correlated with the conventional beef freshness evaluation indicators of total volatile salt base nitrogen (TVB-N), thiobarbituric acid (TBA), total viable count (TVC), and sensory index (SI) to establish multiple linear regression (MLR), back propagation neural network (BPNN), and support vector regression (SVR) models for quantitative beef freshness prediction.Results indicated that the MLR models achieved determination coefficients (R2) of 0.940 6, 0.931 6, 0.958 2, and 0.954 8 for predicting TVB-N, TBA, TVC, and SI values, respectively.The BPNN models exhibited R2 values of 0.962 7, 0.964 1, 0.992 0, and 0.986 4.Meanwhile, the SVR models demonstrated R2 values of 0.971 2, 0.967 9, 0.992 8, and 0.988 3.The SVR models demonstrated the best predictive performance for freshness quantification, with prediction relative errors consistently within 10%.Based on this, the study further developed an SVR model for predicting beef shelf life based on color parameters (R2=0.964 8).The model's predicted values had an average absolute error of <0.5 days compared to the actual values, which was superior to traditional shelf-life models.This study provides a new method for non-destructive and rapid identification of beef freshness and shelf-life.

Key words： machine vision; beef detection; freshness indicators; color parameters; support vector regression; shelf life

Cite this article

LU Zhongchao , QIU Yue , ZHANG Anqiang , ZHANG Jianyou , CUI Pengbo , XIANG Yun , JIN Xia , LYU Fei . Multi-model quantitative analysis of beef freshness judgment by machine vision[J]. Food and Fermentation Industries, 2024 , 50(8) : 262 -270 . DOI: 10.13995/j.cnki.11-1802/ts.037376

References

[1] 封晴霞, 王利强.4 ℃冷藏牛肉的新鲜度检测及指标分析[J].包装与食品机械, 2020, 38(6):1-4;29.
FENG Q X, WANG L Q.Freshness test and index analysis of 4 ℃ chilled beef[J].Packaging and Food Machinery, 2020, 38(6):1-4;29.
[2] 付丽, 高雪琴, 胡晓波, 等.保鲜盒包装冷鲜牛肉贮藏期间品质及菌相变化[J].食品研究与开发, 2019, 40(14):35-41.
FU L, GAO X Q, HU X B, et al.Change in quality and bacteria phase of chilled beef in fresh box packaging during storage[J].Food Research and Development, 2019, 40(14):35-41.
[3] HOSSEINPOUR S, ILKHCHI A H, AGHBASHLO M.An intelligent machine vision-based smartphone app for beef quality evaluation[J].Journal of Food Engineering, 2019, 248:9-22.
[4] TAHERI-GARAVAND A, FATAHI S, OMID M, et al.Meat quality evaluation based on computer vision technique:A review[J].Meat Science, 2019, 156:183-195.
[5] LUO X Z, XIONG L J, GAO X, et al.Determination of beef tenderness based on airflow pressure combined with structural light three-dimensional (3D) vision technology[J].Meat Science, 2023, 202:109206.
[6] 陈士进, 丁冬, 李泊, 等.基于机器视觉的牛肉结缔组织特征和嫩度关系研究[J].南京农业大学学报, 2016, 39(5):865-871.
CHEN S J, DING D, LI B, et al.Research on relationship between beef connective tissue features and tenderness by computer vision technology[J].Journal of Nanjing Agricultural University, 2016, 39(5):865-871.
[7] 李岩, 刘波, 陈怡帆.机器视觉技术在牛肉生理成熟度检测中的应用[J].河北农业大学学报, 2017, 40(6):119-124.
LI Y, LIU B, CHEN Y F.Application of machine vision technology in the detection of beef physiological maturity[J].Journal of Agricultural University of Hebei, 2017, 40(6):119-124.
[8] 姜沛宏, 张玉华, 钱乃余, 等.基于机器视觉技术的肉新鲜度分级方法研究[J].食品科技, 2015, 40(3):296-300.
JIANG P H, ZHANG Y H, QIAN N Y, et al.Research on method to freshness grading of meat based on machine vision technology[J].Food Science and Technology, 2015, 40(3):296-300.
[9] 张茹. 基于机器视觉与质构检测的冷鲜牛肉新鲜度判别[D].无锡:江南大学, 2022.
ZHANG R.Freshness identification of chilled beef based on machine vision and texture analysis[D].Wuxi:Jiangnan University, 2022.
[10] 黄幸幸. 南美白对虾货架期预测指标优化及模型研究[D].上海:上海海洋大学, 2018.
HUANG X X.The prediction index and model of the shelf-life of Penaeus vannamei[D].Shanghai:Shanghai Ocean University, 2018.
[11] 陆逸峰, 何子豪, 曾宪明, 等.基于BP神经网络的鸡蛋货架期和贮藏时间预测模型研究[J].食品科学, 2023,44(21):44-53.
LU Y F, HE Z H, ZENG X M, et al.Investigation on egg shelf life and storage time prediction model based on BP neural network[J].Food Science, 2023,44(21):44-53.
[12] 袁彦彦, 王兴芬.基于SVM的生鲜食品货架期预测[J].物流技术, 2015, 34(19):64-67.
YUAN Y Y, WANG X F.Study on fresh food shelf life forecasting based on SVM[J].Logistics Technology, 2015, 34(19):64-67.
[13] 罗礼纯. 基于残差神经网络模型的冷鲜大黄鱼新鲜度识别研究及应用[D].杭州:浙江工商大学, 2022.
LUO L C.Research and application of freshness recognition of cold and fresh yellow croaker based on the residual neural network model[D].Hangzhou:Zhejiang Gongshang University, 2022.
[14] 张茹, 张奋楠, 周星宇, 等.基于机器视觉的生鲜牛肉冷藏时间识别研究[J].食品与发酵工业, 2022, 48(18):75-80.
ZHANG R, ZHANG F N, ZHOU X Y, et al.Identification of the refrigerated time for chilled beef based on machine vision[J].Food and Fermentation Industries, 2022, 48(18):75-80.
[15] CHENG J H, SUN D W, QU J H, et al.Developing a multispectral imaging for simultaneous prediction of freshness indicators during chemical spoilage of grass carp fish fillet[J].Journal of Food Engineering, 2016, 182:9-17.
[16] 凌双明. 智能视觉机器人图像灰度化处理方法研究[J].山西电子技术, 2023(2):88-91.
LING, S M.Research on image gray processing method of intelligent vision parallel sorting robot[J].Shanxi Electronic Technology, 2023(2):88-91.
[17] 何晓军, 徐爱功, 李玉.利用HSI空间相似性的彩色形态学图像处理方法[J].计算机科学, 2019, 46(4):285-292.
HE X J, XU A G, LI Y.Color morphology image processing method using similarity in HSI space[J].Computer Science, 2019, 46(4):285-292.
[18] 付丽, 马丽卿, 马微, 等.葡萄皮提取物协同冰鲜对牛肉的抗氧化效果[J].食品研究与开发, 2020, 41(6):13-18.
FU L, MA L Q, MA W, et al.Antioxidant effect of grap skin extract and controlled freezing point technique on beef[J].Food Research and Development, 2020, 41(6):13-18.
[19] 郝教敏, 杨文平, 靳明凯, 等.荞麦多酚在卤猪肉制品防腐保鲜中的应用[J].中国调味品, 2019, 44(12):22-25;30.
HAO J M, YANG W P, JIN M K, et al.Application of buckwheat polyphenols in preservation and fresh keeping of marinated pork products[J].China Condiment, 2019, 44(12):22-25;30.
[20] 磨佳琳, 刘程惠, 戴祥英, 等.Nisin抑制冷鲜牛肉氧化的研究[J].肉类工业, 2019(12):38-42.
MO J L, LIU C H, DAI X Y, et al.Study on inhibiting oxidation of chilled beef by nisin[J].Meat Industry, 2019(12):38-42.
[21] PEREIRA P F M, DE SOUSA PICCIANI P H, CALADO V, et al.Electrical gas sensors for meat freshness assessment and quality monitoring:A review[J].Trends in Food Science & Technology, 2021, 118:36-44.
[22] SANAEIFAR A, BAKHSHIPOUR A, DE LA GUARDIA M.Prediction of banana quality indices from color features using support vector regression[J].Talanta, 2016, 148:54-61.
[23] KAYMAK-ERTEKIN F, GEDIK A.Kinetic modelling of quality deterioration in onions during drying and storage [J].Journal of Food Engineering, 2005, 68(4):443-453.
[24] 范中建. 基于机器视觉的羊肉新鲜度无损检测研究[D].新疆:石河子大学, 2018.
FAN Z J.Study on nondestructive testing of mutton freshness based on machine vision[D].Xinjiang:Shihezi University, 2018.
[25] 陈晓宇, 朱志强, 张小栓, 等.食品货架期预测研究进展与趋势[J].农业机械学报, 2015, 46(8):192-199.
CHEN X Y, ZHU Z Q, ZHANG X S, et al.Research progress and trend of food shelf life prediction[J].Transactions of the Chinese Society for Agricultural Machinery, 2015, 46(8):192-199.
Options
Outlines

/

〈 〉

Powered by Beijing Magtech Co. Ltd,.