Korean Journal of Agricultural Science (농업과학연구)

Institute of Agricultural Science, CNU (충남대학교 농업과학연구소)
권호 표지
DOI QR코드

DOI QR Code

Quality grading of Hanwoo (Korean native cattle breed) sub-images using convolutional neural network

  • Kwon, Kyung-Do (National Institute of Agricultural Sciences, RDA) ;
  • Lee, Ahyeong (National Institute of Agricultural Sciences, RDA) ;
  • Lim, Jongkuk (National Institute of Agricultural Sciences, RDA) ;
  • Cho, Soohyun (National Institute of Animal Sciences, RDA) ;
  • Lee, Wanghee (Department of Biosystems Machinery Engineering, Chungnam National University) ;
  • Cho, Byoung-Kwan (Department of Biosystems Machinery Engineering, Chungnam National University) ;
  • Seo, Youngwook (National Institute of Agricultural Sciences, RDA)
  • Received : 2020.08.25
  • Accepted : 2020.11.30
  • Published : 2020.12.01
Download PDF
⟨ Previous Next ⟩

Abstract

The aim of this study was to develop a marbling classification and prediction model using small parts of sirloin images based on a deep learning algorithm, namely, a convolutional neural network (CNN). Samples were purchased from a commercial slaughterhouse in Korea, images for each grade were acquired, and the total images (n = 500) were assigned according to their grade number: 1++, 1+, 1, and both 2 & 3. The image acquisition system consists of a DSLR camera with a polarization filter to remove diffusive reflectance and two light sources (55 W). To correct the distorted original images, a radial correction algorithm was implemented. Color images of sirloins of Hanwoo (mixed with feeder cattle, steer, and calf) were divided and sub-images with image sizes of 161 × 161 were made to train the marbling prediction model. In this study, the convolutional neural network (CNN) has four convolution layers and yields prediction results in accordance with marbling grades (1++, 1+, 1, and 2&3). Every single layer uses a rectified linear unit (ReLU) function as an activation function and max-pooling is used for extracting the edge between fat and muscle and reducing the variance of the data. Prediction accuracy was measured using an accuracy and kappa coefficient from a confusion matrix. We summed the prediction of sub-images and determined the total average prediction accuracy. Training accuracy was 100% and the test accuracy was 86%, indicating comparably good performance using the CNN. This study provides classification potential for predicting the marbling grade using color images and a convolutional neural network algorithm.

Keywords

Acknowledgement

This study was carried out with the support of "Research Program for Agricultural Science & Technology Development (PJ01181502)".

References

  1. Barbon AP, Barbon Jr S, Campos GF, Seixas Jr JL, Peres LM, Mastelini SM, Andreo N, Ulrici A, Bridi AM. 2017. Development of a flexible computer vision system for marbling classification. Computers and Electronics in Agriculture 142:536-544. https://doi.org/10.1016/j.compag.2017.11.017
  2. Barbon DF, Sun DW, Su C. 2013. NIR hyperspectral imaging as non-destructive evaluation tool for the recognition of fresh and frozen-thawed porcine longissimus dorsi muscles. Innovative Food Science & Emerging Technologies 18:226-236. https://doi.org/10.1016/j.ifset.2012.12.011
  3. Dyrmann M, Karstoft H, Midtiby HS. 2016. Plant species classification using deep convolutional neural network. Biosystems Engineering 151:72-80. https://doi.org/10.1016/j.biosystemseng.2016.08.024
  4. Farrow RL, Loneragan GH, Pauli JW, Lawrence TE. 2009. An exploratory observational study to develop an improved method for quantifying beef carcass salable meat yield. Meat Science 82:143-150. https://doi.org/10.1016/j.meatsci.2008.12.014
  5. Glorot X, Bordes A, Bengio Y. 2011. Deep sparse rectifier neural networks. pp. 315-323. In Proceedings of the fourteenth international conference on artificial intelligence and statistics.
  6. Gotoh T, Joo ST. 2016. Characteristics and health benefit of highly marbled Wagyu and Hanwoo beef. Korean Journal Food Science of Animal Resources 36:709-718. https://doi.org/10.5851/kosfa.2016.36.6.709
  7. Greiner SP, Rouse GH, Wilson DE, Cundiff LV, Wheeler TL. 2003. Prediction of retail product weight and percentage using ultrasound and carcass measurements in beef cattle. Journal of Animal Science 81:1736-1742. https://doi.org/10.2527/2003.8171736x
  8. Ioffe S, Szegedy C. 2015. Batch normalization: Accelerating deep network training by reducing internal covariate shift. arXiv preprint arXiv:1502.03167.
  9. Kang TH, Cho SK, Seo J, Kim M, Kim BW. 2019. Contribution analysis of carcass traits and seasonal effect on auction price for Hanwoo steers. Korean Journal of Agricultural Science 46:461-469. [in Korean] https://doi.org/10.7744/KJOAS.20190030
  10. Krizhevsky A, Sutskever I, Hinton GE. 2012. Imagenet classification with deep convolutional neural networks. pp. 1097-1105. In Advances in Neural Information Processing Systems.
  11. Kvam J, Kongsro J. 2017. In vivo prediction of intramuscular fat using ultrasound and deep learning. Computers and Electronics in Agriculture 142:521-523. https://doi.org/10.1016/j.compag.2017.11.020
  12. Kwon KD, Seo Y, Kim GY, Yoo HC, Lee WH, Cho BK. 2019. Feasibility study of convolutional neural network for beef quality grade. p. 156. In Proceedings of the 2019 Spring Conference of Korean Society for Food Engineering.
  13. LeCun Y, Bengio Y, Hinton G. 2015. Deep learning. Nature 521:436-444. https://doi.org/10.1038/nature14539
  14. Lee HJ, Choe J, Kim KT, Oh J, Lee DG, Kwon KM, Choi YI, Jo C. 2017. Analysis of low-marbled Hanwoo cow meat aged with different dry-aging methods. Asian-Australasian Journal of Animal Sciences 30:1733. https://doi.org/10.5713/ajas.17.0318
  15. Lee SH, Park BH, Sharma A, Dang CG, Lee SS, Choi TJ, Choy YH, Kim HC, Jeon KJ, Kim SD, Yeon SH. 2014. Hanwoo cattle: Origin, domestication, breeding strategies and genomic selection. Journal of Animal Science and Technology 56:2. https://doi.org/10.1186/2055-0391-56-2
  16. Lin J, Sha D, Su D, Shen T. 2005. Distortion measurement of CCD imaging system with short focal length and largefield objective. Optics and Lasers in Engineering 43:1137-1144. https://doi.org/10.1016/j.optlaseng.2004.07.006
  17. Ma J, Pu H, Sun DW, Gao W, Qu JH, Ma KY. 2015. Application of Vis-NIR hyperspectral imaging in classification between fresh and frozen-thawed pork Longissimus dorsi muscles. International Journal of refrigeration 50:10-18. https://doi.org/10.1016/j.ijrefrig.2014.10.024
  18. Mohanty SP, Hughes DP, Salathe M. 2016. Using deep learning for image-based plant disease detection. Frontiers in Plant Science 7:1419. https://doi.org/10.3389/fpls.2016.01419
  19. Niu XX, Suen CY. 2012. A novel hybrid CNN-SVM classifier for recognizing handwritten digits. Pattern Recognition 45:1318-1325. https://doi.org/10.1016/j.patcog.2011.09.021
  20. Park J, Byun SC, Lee BU. 2009. Lens distortion correction using ideal image coordinates. IEEE Transactions on Consumer Electronics 55:987-991. https://doi.org/10.1109/TCE.2009.5278053
  21. Platter WJ, Tatum JD, Belk KE, Koontz SR, Chapman PL, Smith GC. 2005. Effects of marbling and shear force on consumers' willingness to pay for beef strip loin steaks. Journal of Animal Science 83:890-899. https://doi.org/10.2527/2005.834890x
  22. Pu H, Sun DW, Ma J, Cheng JH. 2015. Classification of fresh and frozen-thawed pork muscles using visible and near infrared hyperspectral imaging and textural analysis. Meat Science 99:81-88. https://doi.org/10.1016/j.meatsci.2014.09.001
  23. Russakovsky O, Deng J, Su H, Krause J, Satheesh S, Ma S, Huang Z, Karpathy A, Khosla A, Bernstein M, Berg AC. 2015. Imagenet large scale visual recognition challenge. International Journal of Computer Vision 115:211-252. https://doi.org/10.1007/s11263-015-0816-y
  24. Srivastava N, Hinton G, Krizhevsky A, Sutskever I, Salakhutdinov R. 2014. Dropout: A simple way to prevent neural networks from overfitting. The Journal of Machine Learning Research 15:1929-1958.
  25. Sun X, Young J, Liu JH, Newman D. 2018. Prediction of pork loin quality using online computer vision system and artificial intelligence model. Meat Science 140:72-77. https://doi.org/10.1016/j.meatsci.2018.03.005
  26. Weng J, Cohen P, Herniou M. 1992. Camera calibration with distortion models and accuracy evaluation. IEEE Transactions on Pattern Analysis and Machine Intelligence 14:965-980. https://doi.org/10.1109/34.159901
  27. Wu X, Song X, Qiu Z, He Y. 2016. Mapping of TBARS distribution in frozen-thawed port using NIR hyperspectral imaging. Meat Science 113:92-96. https://doi.org/10.1016/j.meatsci.2015.11.008
  28. Xie A, Sun DW, Xu Z, Zhu Z. 2015. Rapid detection of frozen pork quality without thawing by Vis-NIR hyperspectral imaging technique. Talanta 139:208-215. https://doi.org/10.1016/j.talanta.2015.02.027
  29. Yanai K, Kawano Y. 2015. Food image recognition using deep convolutional network with pre-training and fine-tuning. pp. 1-6. In 2015 IEEE International Conference on Multimedia & Expo Workshops (ICMEW).
  30. Zhang Z. 2000. A flexible new technique for camera calibration. IEEE Transactions on Pattern Analysis and Machine Intelligence 22:1330-1334. https://doi.org/10.1109/34.888718