با همکاری انجمن علمی گیاهان دارویی ایران

نوع مقاله : مقاله علمی پژوهشی

نویسندگان

1 مربی، گروه مهندسی کامپیوتر، دانشگاه تربت‌حیدریه ، تربت حیدریه، ایران

2 مربی، گروه کامپیوتر دانشگاه آزاد اسلامی واحد تربت‌حیدریه ، تربت حیدریه، ایران

3 دانشجو، کارشناسی کامپیوتر دانشگاه تربت‌حیدریه، تربت حیدریه، ایران

4 استادیار، گروه مهندسی کامپیوتر، دانشگاه تربت‌حیدریه، تربت حیدریه، ایران

5 مربی، گروه علوم و صنایع غذایی دانشگاه آزاد اسلامی واحد تربت‌حیدریه و مدیر واحد R&D شرکت زعفران کیان توس، تربت حیدریه، ایران

چکیده

زعفران به‌عنوان یک کالای تجاری مهم در کشور به­شمار می‌آید و توجه به مکانیزه کردن آن از مرحله تولید تا بسته‌بندی اهمیت زیادی دارد. در بدو ورود زعفران به فرایند کیفی سنجی در آزمایشگاه ، ارزیابی اولیه بر اساس مشخصات ظاهری زعفران توسط شخص خبره انجام می‌شود. لیکن بروز خطای انسانی در تشخیص کیفیت زعفران بر مبنای ویژگی‌های ظاهری آن امری اجتناب‌‌ناپذیر‌ است؛ استفاده از تکنیک‌های مبتنی بر هوش مصنوعی می‌تواند ضمن مکانیزه کردن سیستم، در کاهش خطاهای انسانی نیز تأثیرگذار باشد. این مطالعه از نوع تشخیصی بوده و پایگاه داده آن مشتمل بر 113 نمونه زعفران با 7 ویژگی می‌باشد که توسط محققین این پژوهش، در مهر‌ماه 1396 از آزمایشگاه‌ معتبر زعفران و تحت نظارت شخص خبره جمع‌آوری‌ شده است. کیفی سنجی نمونه‌ها به کمک ویژگی‌ها در 4 کلاس مختلف زعفران پوشال درجه‌یک (نگین)، پوشال درجه دو (خوب)، پوشال درجه سه (معمولی) و پوشال درجه چهار (معمولی درجه‌دو) انجام ‌شده است. به‌منظور درجه­بندی زعفران، از روش‌های مبتنی بر شبکه‌های عصبی مصنوعی استفاده‌شده ‌است. پس از تحلیل و مقایسه مدل‏های تولیدشده با استفاده از دو نوع شبکه‌ عصبی پرسپترون چندلایه و شبکه عصبی بردار یادگیر، بالاترین دقت کلاس‌بندی روی نمونه‌های آموزش و آزمون به ترتیب با 75/93 و 75/75 درصد حاصل شد. دقت به‌دست‌آمده نشان‌دهنده آن است که مدل شبکه عصبی پرسپترون چندلایه می‌تواند به‌عنوان یک تصمیم گیر در کنار شخص خبره و یا به‌صورت مستقل در مراکز آزمایشگاهی زعفران مورد استفاده قرار ‌گیرد.

کلیدواژه‌ها

عنوان مقاله [English]

Classification of saffron based on its apparent characteristics using artificial neural networks

نویسندگان [English]

  • Seyaed Ehsan Yasrebi 1
  • Iman Zabbah 2
  • Behnaz Behzadiyan 3
  • Ali Maroosi 4
  • Roya Rezaie 5

1 Lecturer, Computer Department, Torbat Heydarieh University, Torbat Heydarieh,Iran

2 Lecturer, Computer Department, Torbat Heydarieh Islamic Azad University, Torbat Heydarieh,Iran

3 Bachelor of Computer University of Torbat Heydarieh, Torbat Heydarieh,Iran

4 Assistant professor, Computer Department, Torbat Heydarieh University, Torbat Heydarieh,Iran

5 Lecturer , Department of Food Science and Technology, Islamic Azad University of Torbat Heydarieh, and Director of R & D Department, Saffron Kian Toos Co , Torbat Heydarieh,Iran

چکیده [English]

Saffron is an important commercial good in Iran and it is important to pay attention to its mechanization from production to packaging. Upon arrival of the saffron to the laboratory's qualitative process, an initial assessment is carried out by an expert on the basis of the apparent features. However, human error in determining the quality of saffron based on its apparent features is inevitable; use of artificial intelligence techniques can be effective in reducing human errors while mechanizing the system. It was a diagnostic study and its database consisted of 113 samples of saffron with 7 features, which were collected by the researchers on October 2016 from the valid laboratory of Saffron and under the supervision of an expert. Sample qualitative analysis was performed with the help of features in 4 different classes including excellent, good, average and second grade average. Artificial neural networks have been used to classify saffron. After analyzing and comparing the generated models using multilayer perceptron neural networks and learning vector neural network, the highest accuracy of classification on the training and testing samples was obtained with 75.93 and 75.75%, respectively. The accuracy obtained indicated that the multi-layer perceptron neural network model can be used as a decision maker by an expert or independently in saffron lab centers.

کلیدواژه‌ها [English]

  • Saffron classification
  • Artificial Neural Network
  • Artificial intelligence
Akbarpour, A., Khorashadizadeh, O., Shahidi, A., and Ghochanian, E. 2013. Performance evaluation of artificial neural network models in estimate production of yield saffron based on climate parameters. Journal of Saffron Research 1 (1): 27-35. (In Persian with English Summary).
Aliabadi, R., and Mohammadi, M. 2011. Qualitative study of the saffron flower using smart techniques. In National Conference on Computer and Information Technology. Kerman. Iran. pp. 1-5. (In Persian).
Aliabadi, R., and Mohammadi, M. 2012. Presentation of a new method for saffron flower cutting automation using intelligent techniques. In 2th National Conference on Computer Engineering, Electrical and Information Technology. Khomein Islamic Azad University. pp. 1-5. (In Persian).
Asgharpour, M. 2013. Multi-Criteria Decision Making. Tehran University Publication, Tehran, Iran. pp. 412. Available online at http://press.ut.ac.ir/. (In Persian).
Coulibaly, P., Anctil, F., and Bobee, B. 1999. Prévision hydrologique par réseaux de neurones artificiels: état de l’art. Journal of Canadian, Journal of Civil Engineering 26 (3): 293-304. Gracia, L., PerezVidal, C., and GracialÓpez, C. 2009. An automated cutting system to obtain the stigmas of the saffron flower. Journal of Biosystems Engineering 104 (1): 8-17.
Dehghan, P., Mogharabi, M., Zabbah, I., Layeghi, K., and Maroosi, A. 2018. Modeling breast cancer using data mining methods. Journal of Health and Biomedical Informatics. 4 (4): 266-278.
Gardner, M.W., and Dorling, S.R. 1998. Artificial neural networks (the multilayer perceptron)—a review of applications in the atmospheric sciences. Atmospheric Environment 32 (14-15): 2627-2636.
Hosseini, M., HemmatiKakhaki, A., and Karbasi, A.R. 2003. Study and evaluation of social and economic effects of saffron ten years research. In 3rd Iranian National Conference on Saffron, Institute of Food Science and Technology, Mashhad, Iran. (In Persian).
Hosseini, M.T., SioseMarde, A., Fathi, P., and SioseMarde. M. 2007. Application of artificial neural network (ANN) and multiple regressions for estimating assessing the performance of dry farming wheat yield in Ghorveh region Kurdistan province. Journal of Agricultural Research: Water, Soil and Plant in Agriculture 7 (1): 41-54. (In Persian with English Summary).
Khalili, K., and Serajpor, M. 2006. Saffron cutting automation using image processing. In the 4th Conference on Visual Machines and Image Processing. Mashhad Ferdowsi University. pp. 1-7. (In Persian).
Kiani, S., and Minaei, S. 2015. Development and evaluation of an intelligent system based on machine vision and machine olfaction to determine the compounds and quality assessment of herbal medicinal products (Case study of saffron). In 1st National Conference on Medicinal Herbs and Herbal Medicines. Shahid Beheshti University, Tehran, Iran. pp. 1-12. (In Persian).
Leffingwell, J. 2002. Saffron, this a part of our series on aroma materials produced by carotenoid degradation. Leffingwell Reports 2 (5): 1-7. Available online at http://www.leffingwell.com/saffron.htm. (verified October 2002).
Mahdavi, M. 2007. Comparison of quantitative and qualitative sampling of saffron samples in different regions of Iran. Ph.D. Pharmacy dissertation, Mashhad University of Medical Sciences, Mashhad, Iran. (In Persian).
Moghaddasi, M.S. 2010. Saffron chemicals and medicine usage. Journal of Medicinal Plants Research 4 (6): 427-430. Available online at http://www.academicjournals.org/jmpr.
Mollafilabi, A. 2009. The new methods of saffron production. In 4th National Festival of Saffron. Khorasan Razavi, Iran, 27- 28 October 2009. (In Persian).
Omkarprasad, V., and Sushil, K. 2006. Analytic hierarchy process: an overview of applications. Journal of European Operational Research 169 (1): 1-29.
Rahmani, E., Khalili, A., and Liaghat, A. 2008. Quantitative survey of drought effects on barley yield in East Azerbaijan by classical statistical methods. Journal of the JWSS 12 (44): 25-36. Publisher:  Isfahan University of Technology. (In Persian).
RashidSorkhabadi, M., Shahidi, A., and KhasheiSiuki, A. 2014. Determination of suitable region for saffron cultivation based on water and soil characteristics using hierarchical analysis process method (Case study: Torbat e Hydariyeh city). Journal of Saffron Research 2 (1): 58-72. (In Persian with English Summary).
Remesan, R., Shamim, M.A., and Han, D. 2008. Model data selection using gamma test for daily solar radiation estimation. Journal of Hydrological Processes 22 (1): 4301-4039. Publisher: Water and Environmental Management Research Centre, Department of Civil Engineering, University of Bristol, Lunsford House, Cantocks Close, Clifton, Bristol, BS8 1UP, UK. Retrieved (www.interscience.wiley.com).
Salary, M., Najafi, R., and Karaghian, H. 2010. Evaluation of physicochemical changes of saffron during the one-year preservation period. Journal of Food Science and Technology 2 (1): 35-43. (In Persian).
Sato, A., and Yamada, K. 1996. Generalized learning vector quantization. In Advances in neural information processing systems pp. 423-429.
Shyam, N.J.Ha. 2010. Nondestructive Evaluation of Food Quality, Theory, and Practice. pp. 298. Available online at http://www.springer.com/gp/book/9783642157950#aboutbook.
Valluru, R., and Hayagriva, R. 1995. C++ Neural Networks and Fuzzy Logic. MIS: Press Publication. New Delhi, India. pp. 380-381.
Zarghani, F., Karimi, A., Khorasani, R., and Lakzian, A. 2016. To evaluation the effect of soil physical and chemical characteristics on the growth characteristics of saffron (Crocus sativus L.) corms in Torbat-e Heydariyeh area. Journal of Agroecology 8 (1): 120-133. (In Persian with English Summary).