Распознавание изображений элементов зерновых смесей методами глубокого обучения с использованием библиотек keras и tensorflow
Ключевые слова:
обработка изображений, анализ изображений, глубокое обучение, машинное обучение, нейросеть, классификация зерновых смесей
Аннотация
Методом вычислительного эксперимента исследована возможность использования нейросетей глубокого обучения для распознавания изображений элементов зерновых смесей. Проведено сравнение результатов для трех различных архитектур, реализованных в пакетах VGG16, VGG19 и MobileNet на наборе изображений зёрен различных растений. Выявлен относительный вклад морфологических, цветовых и текстурных признаков в решение задачи классификации изображений зерен с помощью нейросетей глубокого обучения.
Скачивания
Данные скачивания пока не доступны.
Литература
1. Shaw, W. E. Machine vision for detecting defects on fruit and vegetables / W. E. Shaw // In Food Processing Automation I – Proceedings of the 1990 conference 50-59. St. Joseph, MI: ASAE. –1990. – P.50–59.
2. Tillet, R. D. Image analysis for agricultural processes: A review of potential opportunities / R. D. Tillet // Journal of Agricultural Engineering Research. – 1991. – No 50. – P. 247–258.
3. Barker, D. A. The use of ray parameters for the discrimination of Australian wheat varieties / Barker D. A., Vouri T. A., Hegedus M. R. and My-ers D. G. // Plant Varieties and Seeds. – 1992. – No 5(1). – P. 35–45.
4. Barker, D. A. The use of slice and aspect ra-tio parameters for the discrimination of Australian wheat varieties / D. A. Barker, T. A. Vouri, and D.G. Myers // Plant Varieties and Seeds. – 1992. – No 5(l). – P. 47–52.
5. Barker, D. A. The use of Fourier descriptors for the discrimination of Australian wheat varieties / D. A. Barker, T. A. Vouri and D. G. Myers // Plant Varieties and Seeds. – 1992.– No 5(2). – P. 93–102.
6. Barker, D. A. The use of Chebychev coefficients for the discrimination of Australian wheat varieties / D. A. Barker, T. A. Vouri, M. R. Hegedus and D.G. Myers // Plant Varieties and Seeds. – 1992. – No 5(2). – P. 103–111.
7. Crowe, T. G. Color line-scan imaging of cereal grain kernels / T. G. Crowe, X. Y. Luo, D. S. Jayas and N. R. Bulley // Applied Engineer-ing in Agriculture. – 1997. – No 13(5). – P. 689–694.
8. Draper, S. R. Preliminary observations with a computer based system for analysis of the shape of seeds and vegetative structures / S. R. Draper and A. J. Travis // Journal of the National Insti-tute of Agricultural Botany. – 1984. – No16(3). – P. 387–395.
9. Keefe, P. D. An automated machine vision system for the morphometry of new cultivars and plant genebank accessions / P. D. Keefe and S. R. Draper // Plant Varieties and Seeds 1(1): 1–11.
10. Keefe, P. D. The measurement of new char-acters for cultivar identification in wheat using machine vision / P. D. Keefe and S. R. Draper // Seed Science and Technology. – 1986. – No14(3). – P. 715–724.
11. Lai, F. S. Application of pattern recogni-tion techniques in the analysis of cereal grains / F. S. Lai, I. Zayas, and Y. Pomeranz // Cereal chemistiy. 1986. – No63(2). – P. 168–172.
12. Majumdar, S. Textural features for au-tomated grain identification / S. Majumdar, D. S. Jayas, S. J. Symons and N.R. Bulley // CSAE Paper No. 96-602. Saskatoon, SK: CSAE, 1996.
13. Majumdar, S. Classification of various grains using optical properties / S. Majumdar, D. S. Jayas, J. L. Hehn and N. R. Bulley // Canadi-an Agricultural Engineering. – 1996. – No.38(2). – P. 139–144.
14. Majumdar, S. Single-kernel mass deter-mination for grain inspection using machine vision / S. Majumdar and D. S. Jayas // Applied Engineering in Agriculture. – 1999. – No15(4). – P. 357–362.
15. Majumdar, S. Classification of bulk samples of cereal grains using machine vision / S. Majum-dar and D. S. Jayas. // Journal of Agricultural En-gineering Research. – 1999. – No73(l). – P. 35–47.
16. Majumdar, S. Classification of cereal grains using machine vision. I. Morphology models / S. Majumdar and D. S. Jayas // Transactions of the ASAE. – 2000. – No 43(6). – P. 1669–1675.
17. Majumdar, S. Classification of cereal grains using machine vision. II. Color models / S. Majumdar and D. S. Jayas // Transactions of the ASAE. – 2000. – No43(6). – P.1677–1680.
18. Majumdar, S. Classification of cereal grains using machine vision. III. Texture models / S. Majumdar and D. S. Jayas // Transactions of the ASAE. – 2000. – No43(6). – P.1681–1687.
19. Majumdar, S. Classification of cereal grains using machine vision. IV. Combined morphology, color, and texture models / S. Majumdar and D. S. Jayas. // Transactions of the ASAE. – 1988. – No 43(6). – P. 1689–1694.
20. Myers, D. G. The application of image processing techniques to the identification of Australian wheat varieties / D. G. Myers and K. J. Edsall // Plant Varieties and Seeds. – 1989. – No 2(2). – P. 109–116.
21. Neuman, M. Discrimination of wheat class and variety by digital image analysis of whole grain samples / M. Neuman, H. D. Sapirstein, E. Shwedyk and W. Bushuk // Journal of Ce-real Science. – 1987. – No 6(2). – P. 125–132.
22. Paliwal J. Grain kernel identification using kernel signature / J. Paliwal, N. S. Shashidhar and D. S. Jayas // Transactions of the ASAE. – 1999. – No42(6). – P.1921–1924.
23. Sapirstein, H. D. Quantitative determination of foreign material and vitreosity in wheat by digital image analysis / H. D. Sapirstein and W. Bushuk // lnICC’89 Symposium: Wheat End-Use Properties. H. Salovaara (ed.). Lahiti, Finland, 1989.
24. Sapirstein, H. D. An instmmental system for cereal grain classification using digital image analysis / H. D. Sapirstein, M. Neuman, E. H. Wright, E. Shwedyk and Bushuk W. // Journal of Cereal Science. – 1987. – No 6(1). – P. 3–14.
25. Symons, S. J. Relationship between oat kernel weight and milling yield / S. J. Symons and R. G. Fulcher // Journal of Cereal Science. – 1988. – No7(3). – P. 215–217.
26. Symons, S. J. 1988b. Determination of variation in oat kernel morphology by digital image analysis / S. J. Symons and R. G. Fulcher // Journal of Cereal Science. – 1988. – No 7(3). – P. 219–228.
27. Zayas, I. Discrimination between wheat classes and varieties by image analysis / I. Zayas, F. S. Lai and Y. Pomeranz // Cereal Chemistry. 1986. – No63(l). – P. 52–56.
28. Zayas, I. Discrimination of whole from broken com kernels with image analysis / I. Zayas, H. Converse and J. Steele // Transactions of the ASAE. – 1990. – No35(5). – P. 1642–1646.
29. Zayas, I. Discrimination between Arthur and Arc an wheats by image analysis / I. Zayas, Y. Pomeranz and F. S. Lai // Cereal Chemistry. – 1985. – No 62(6). – P. 478–480.
30. Zayas, I. Discrimination of wheat and non-wheat components in grain samples by digital image analysis / I. Zayas, Y. Pomeranz and F. S. Lai // Cereal Chemistry. – 1989. – No66(3). – P. 233–237.
31. Wigger, W. D. Classification of fungal-damaged soybeans using color-image processing / W. D. Wigger, M. R. Paulsen, J. B. Litchfield and J. B. Sinclair // ASAE Paper No. 88-3053. – 1988. – St. Joseph, MI: ASAE.
32. Karunakaran, C. Machine Vision Systems for Agricultural Products / C. Karunakaran, N. S. Visen, J. Paliwal, G. Zhang, D. S. Jayas, N. D. G. White // CSAE Paper No. 01-305. – 2001. – Mansonville QC: CSAE/SCGR.
33. Paliwal, J. Digital image analysis of grain samples for potential use in grain cleaning (Ph. D. thesis) / J. Paliwal; Department of Biosystems Engineering, University of Manitoba, Winnipeg, MB, Canada, 2002. – 247 p.
34. Visen, N. S. Machine Vision Based Grain Handling System (Ph.D. thesis) / N. S. Visen; Department of Biosystems Engineering, University of Manitoba, Winnipeg, MB, Canada. – 175 p.
35. Computer Vision Technology for Food Quality Evaluation / Edited by DaWen Sun; Academic Press, Elsevier. – 2016. – 635 P.
36. Paliwal, J. Cereal grain and dockage identification using machine vision / J. Paliwal, N. S. Visen, D. S. Jayas, N. D. G. White // Biosys-tems Engineering. – 2003. – No85(1). – P. 51–57.
37. Haralick, R. M. Texture features for image classification. IEEE Transactions on Systems / R. M. Haralick, K. Shanmugam and I. Dinstein // Man and Cybernetics. – 1973. – No 3(6). – P. 610–621.
38. Galloway, M. M. Texture analysis using gray level run length // Galloway M.M. // Computer graphics and image processing. – 1975. – No 4. – P. 172–179.39.Гонсалес, Р. Цифровая обработка изображений / Р. Гонсалес, Р. Вудс. – М. : Техносфера, 2012. – 1104 с.
40. Simonyan, K. Very deep convolution-al networks for large-scale image recognition / K. Simonyan, A. Zisserman // CoRR. – 2014. – No 1409.1556. – 14 с.
41. Howard, A. G. MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications / A. G. Howard [и др.] // CoRR. – 2017. – No 1704.04861. – 9 с.
2. Tillet, R. D. Image analysis for agricultural processes: A review of potential opportunities / R. D. Tillet // Journal of Agricultural Engineering Research. – 1991. – No 50. – P. 247–258.
3. Barker, D. A. The use of ray parameters for the discrimination of Australian wheat varieties / Barker D. A., Vouri T. A., Hegedus M. R. and My-ers D. G. // Plant Varieties and Seeds. – 1992. – No 5(1). – P. 35–45.
4. Barker, D. A. The use of slice and aspect ra-tio parameters for the discrimination of Australian wheat varieties / D. A. Barker, T. A. Vouri, and D.G. Myers // Plant Varieties and Seeds. – 1992. – No 5(l). – P. 47–52.
5. Barker, D. A. The use of Fourier descriptors for the discrimination of Australian wheat varieties / D. A. Barker, T. A. Vouri and D. G. Myers // Plant Varieties and Seeds. – 1992.– No 5(2). – P. 93–102.
6. Barker, D. A. The use of Chebychev coefficients for the discrimination of Australian wheat varieties / D. A. Barker, T. A. Vouri, M. R. Hegedus and D.G. Myers // Plant Varieties and Seeds. – 1992. – No 5(2). – P. 103–111.
7. Crowe, T. G. Color line-scan imaging of cereal grain kernels / T. G. Crowe, X. Y. Luo, D. S. Jayas and N. R. Bulley // Applied Engineer-ing in Agriculture. – 1997. – No 13(5). – P. 689–694.
8. Draper, S. R. Preliminary observations with a computer based system for analysis of the shape of seeds and vegetative structures / S. R. Draper and A. J. Travis // Journal of the National Insti-tute of Agricultural Botany. – 1984. – No16(3). – P. 387–395.
9. Keefe, P. D. An automated machine vision system for the morphometry of new cultivars and plant genebank accessions / P. D. Keefe and S. R. Draper // Plant Varieties and Seeds 1(1): 1–11.
10. Keefe, P. D. The measurement of new char-acters for cultivar identification in wheat using machine vision / P. D. Keefe and S. R. Draper // Seed Science and Technology. – 1986. – No14(3). – P. 715–724.
11. Lai, F. S. Application of pattern recogni-tion techniques in the analysis of cereal grains / F. S. Lai, I. Zayas, and Y. Pomeranz // Cereal chemistiy. 1986. – No63(2). – P. 168–172.
12. Majumdar, S. Textural features for au-tomated grain identification / S. Majumdar, D. S. Jayas, S. J. Symons and N.R. Bulley // CSAE Paper No. 96-602. Saskatoon, SK: CSAE, 1996.
13. Majumdar, S. Classification of various grains using optical properties / S. Majumdar, D. S. Jayas, J. L. Hehn and N. R. Bulley // Canadi-an Agricultural Engineering. – 1996. – No.38(2). – P. 139–144.
14. Majumdar, S. Single-kernel mass deter-mination for grain inspection using machine vision / S. Majumdar and D. S. Jayas // Applied Engineering in Agriculture. – 1999. – No15(4). – P. 357–362.
15. Majumdar, S. Classification of bulk samples of cereal grains using machine vision / S. Majum-dar and D. S. Jayas. // Journal of Agricultural En-gineering Research. – 1999. – No73(l). – P. 35–47.
16. Majumdar, S. Classification of cereal grains using machine vision. I. Morphology models / S. Majumdar and D. S. Jayas // Transactions of the ASAE. – 2000. – No 43(6). – P. 1669–1675.
17. Majumdar, S. Classification of cereal grains using machine vision. II. Color models / S. Majumdar and D. S. Jayas // Transactions of the ASAE. – 2000. – No43(6). – P.1677–1680.
18. Majumdar, S. Classification of cereal grains using machine vision. III. Texture models / S. Majumdar and D. S. Jayas // Transactions of the ASAE. – 2000. – No43(6). – P.1681–1687.
19. Majumdar, S. Classification of cereal grains using machine vision. IV. Combined morphology, color, and texture models / S. Majumdar and D. S. Jayas. // Transactions of the ASAE. – 1988. – No 43(6). – P. 1689–1694.
20. Myers, D. G. The application of image processing techniques to the identification of Australian wheat varieties / D. G. Myers and K. J. Edsall // Plant Varieties and Seeds. – 1989. – No 2(2). – P. 109–116.
21. Neuman, M. Discrimination of wheat class and variety by digital image analysis of whole grain samples / M. Neuman, H. D. Sapirstein, E. Shwedyk and W. Bushuk // Journal of Ce-real Science. – 1987. – No 6(2). – P. 125–132.
22. Paliwal J. Grain kernel identification using kernel signature / J. Paliwal, N. S. Shashidhar and D. S. Jayas // Transactions of the ASAE. – 1999. – No42(6). – P.1921–1924.
23. Sapirstein, H. D. Quantitative determination of foreign material and vitreosity in wheat by digital image analysis / H. D. Sapirstein and W. Bushuk // lnICC’89 Symposium: Wheat End-Use Properties. H. Salovaara (ed.). Lahiti, Finland, 1989.
24. Sapirstein, H. D. An instmmental system for cereal grain classification using digital image analysis / H. D. Sapirstein, M. Neuman, E. H. Wright, E. Shwedyk and Bushuk W. // Journal of Cereal Science. – 1987. – No 6(1). – P. 3–14.
25. Symons, S. J. Relationship between oat kernel weight and milling yield / S. J. Symons and R. G. Fulcher // Journal of Cereal Science. – 1988. – No7(3). – P. 215–217.
26. Symons, S. J. 1988b. Determination of variation in oat kernel morphology by digital image analysis / S. J. Symons and R. G. Fulcher // Journal of Cereal Science. – 1988. – No 7(3). – P. 219–228.
27. Zayas, I. Discrimination between wheat classes and varieties by image analysis / I. Zayas, F. S. Lai and Y. Pomeranz // Cereal Chemistry. 1986. – No63(l). – P. 52–56.
28. Zayas, I. Discrimination of whole from broken com kernels with image analysis / I. Zayas, H. Converse and J. Steele // Transactions of the ASAE. – 1990. – No35(5). – P. 1642–1646.
29. Zayas, I. Discrimination between Arthur and Arc an wheats by image analysis / I. Zayas, Y. Pomeranz and F. S. Lai // Cereal Chemistry. – 1985. – No 62(6). – P. 478–480.
30. Zayas, I. Discrimination of wheat and non-wheat components in grain samples by digital image analysis / I. Zayas, Y. Pomeranz and F. S. Lai // Cereal Chemistry. – 1989. – No66(3). – P. 233–237.
31. Wigger, W. D. Classification of fungal-damaged soybeans using color-image processing / W. D. Wigger, M. R. Paulsen, J. B. Litchfield and J. B. Sinclair // ASAE Paper No. 88-3053. – 1988. – St. Joseph, MI: ASAE.
32. Karunakaran, C. Machine Vision Systems for Agricultural Products / C. Karunakaran, N. S. Visen, J. Paliwal, G. Zhang, D. S. Jayas, N. D. G. White // CSAE Paper No. 01-305. – 2001. – Mansonville QC: CSAE/SCGR.
33. Paliwal, J. Digital image analysis of grain samples for potential use in grain cleaning (Ph. D. thesis) / J. Paliwal; Department of Biosystems Engineering, University of Manitoba, Winnipeg, MB, Canada, 2002. – 247 p.
34. Visen, N. S. Machine Vision Based Grain Handling System (Ph.D. thesis) / N. S. Visen; Department of Biosystems Engineering, University of Manitoba, Winnipeg, MB, Canada. – 175 p.
35. Computer Vision Technology for Food Quality Evaluation / Edited by DaWen Sun; Academic Press, Elsevier. – 2016. – 635 P.
36. Paliwal, J. Cereal grain and dockage identification using machine vision / J. Paliwal, N. S. Visen, D. S. Jayas, N. D. G. White // Biosys-tems Engineering. – 2003. – No85(1). – P. 51–57.
37. Haralick, R. M. Texture features for image classification. IEEE Transactions on Systems / R. M. Haralick, K. Shanmugam and I. Dinstein // Man and Cybernetics. – 1973. – No 3(6). – P. 610–621.
38. Galloway, M. M. Texture analysis using gray level run length // Galloway M.M. // Computer graphics and image processing. – 1975. – No 4. – P. 172–179.39.Гонсалес, Р. Цифровая обработка изображений / Р. Гонсалес, Р. Вудс. – М. : Техносфера, 2012. – 1104 с.
40. Simonyan, K. Very deep convolution-al networks for large-scale image recognition / K. Simonyan, A. Zisserman // CoRR. – 2014. – No 1409.1556. – 14 с.
41. Howard, A. G. MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications / A. G. Howard [и др.] // CoRR. – 2017. – No 1704.04861. – 9 с.
Опубликован
2018-06-21
Как цитировать
Крыловецкий, А. А., & Суходолов, Д. М. (2018). Распознавание изображений элементов зерновых смесей методами глубокого обучения с использованием библиотек keras и tensorflow. Вестник ВГУ. Серия: Системный анализ и информационные технологии, (2), 139-148. https://doi.org/10.17308/sait.2018.2/1222
Выпуск
Раздел
Интеллектуальные информационные системы
- Авторы сохраняют за собой авторские права и предоставляют журналу право первой публикации работы, которая по истечении 6 месяцев после публикации автоматически лицензируется на условиях Creative Commons Attribution License , которая позволяет другим распространять данную работу с обязательным сохранением ссылок на авторов оригинальной работы и оригинальную публикацию в этом журнале.
- Авторы имеют право размещать их работу в сети Интернет (например в институтском хранилище или персональном сайте) до и во время процесса рассмотрения ее данным журналом, так как это может привести к продуктивному обсуждению и большему количеству ссылок на данную работу (См. The Effect of Open Access).
