Classification of C. annuum and C. frutescens Ripening Stages: How Well Does Deep Learning Perform?

Marsyita Hanafi; Siti Mariam Shafie; Zarina Ibrahim

doi:10.31436/iiumej.v25i2.2769

Authors

Marsyita Hanafi Universiti Putra Malaysia https://orcid.org/0000-0003-1012-6988
Siti Mariam Shafie Universiti Putra Malaysia
Zarina Ibrahim Universiti Putra Malaysia

DOI:

https://doi.org/10.31436/iiumej.v25i2.2769

Keywords:

Transfer Learning, Deep Learning, Fruit Classification, Chilli Fruit Dataset

Abstract

Chilli is one of the world's most widely grown crops. Among all of the chilli variants, C. annuum and C. frustescents are the most prevalent and consistently liked variants in Asia, where it is appreciated for its strong taste and pungency. Nevertheless, harvesting at the proper ripening stage according to their colour, size, and texture is essential to ensure the best quality, marketability, and shelf life. Currently, visual inspection is the primary method used by farmers, which is time-consuming and complicated. Even though automated chilli classification using computer vision and intelligent methods has received scholars' attention, the classification of C. annuum and C. frustescents ripening stages using deep learning models has not been extensively studied. Hence, this study aims to investigate the effectiveness of three deep learning models, namely EfficientNetB0, VGG16 and ResNet50, in classifying chilli ripening stages into unripe, ripe, and overripe classes. We also introduce a huge dataset comprising 9,022 images of C. annuum and C. frustescents chilli under various growth stages and imaging conditions which provides sufficient samples for the deep learning modelling. The experimental results show that the ResNet50 model outperforms other models with more than 95% accuracy for all classes.

ABSTRAK: Cili merupakan salah satu tanaman terbanyak ditanam di dunia. Antara semua varian cili, C. annuum dan C. frustescents adalah yang paling meluas ditanam dan merupakan varian paling pedas di Asia, kerana rasanya yang kuat. Namun begitu, penuaian pada peringkat cili matang mengikut warna, saiz dan teksturnya adalah penting bagi memastikan kualiti, kebolehpasaran dan jangka hayat terbaik. Pada masa ini, pemeriksaan visual adalah kaedah utama yang diguna pakai petani bagi memeriksa cili, tetapi ia memakan masa dan rumit. Walaupun pengelasan cili secara automatik menguna pakai kaedah komputer dan pintar mendapat perhatian sarjana, kajian tentang klasifikasi cili jenis C. annuum dan C. frustescent pada peringkat matang menggunakan model pembelajaran mendalam masih belum begitu meluas. Oleh itu, kajian ini bertujuan bagi mengkaji keberkesanan tiga model pembelajaran mendalam, iaitu EfficientNetB0, VGG16 dan ResNet50, dalam mengklasifikasi kematangan cili pada beberapa peringkat matang cili seperti belum masak, masak dan terlalu masak. Kami juga memperkenalkan set data yang besar terdiri daripada 9,022 imej cili C. annuum dan C. frustescents pada pelbagai peringkat pertumbuhan dan keadaan imej, bagi menyediakan sampel yang cukup untuk membina model pembelajaran mendalam. Hasil dapatan eksperimen mendapati model ResNet50 mengatasi model lain dengan peratusan 95% lebih tepat berbanding semua kelas.

Downloads

Download data is not yet available.

Metrics

Metrics Loading ...

References

Berke TG, Shieh SC. (2001) Capsicum , chillies , paprika , bird' s eye chilli Introduction?: classification and use. Handbook of Herbs and Spices, 1:111–122. http://dx.doi.org/10.1533/9781855736450.111. DOI: https://doi.org/10.1533/9781855736450.111

Chakrabarty S, Islam AKMA. (2017) Selection Criteria for Improving Yield in Chili (Capsicum annuum) . Advances in Agriculture, Volume 2017. https://doi.org/10.1155/2017/5437870. DOI: https://doi.org/10.1155/2017/5437870

Department of Statistics Malaysia Official Portal. (n.d.). Available: https://www.dosm.gov.my/v1/index.php?r=column/cthemeByCat&cat=164&bul_id=OTM1TD

MzS1IvYm5mU1JiU1Fwekt3UT09&menu_id=Z0VTZGU1UHBUT1VJMFlpaXRRR0xpdz09.

Vázquez-Espinosa M, Fayos O, González-De-Peredo AV, Espada-Bellido E, Ferreiro-González M, Palma M, Garcés-Claver A, Barbero GF. (2020) Changes in capsiate content in four chili pepper genotypes (capsicum spp.) at different ripening stages. Agronomy 10(9):1337. https://doi.org/10.3390/agronomy10091337. DOI: https://doi.org/10.3390/agronomy10091337

Martínez-López LA, Ochoa-Alejo N, Martínez O. (2014) Dynamics of the chili pepper transcriptome during fruit development. BMC Genomics 15:143. https://doi.org/10.1186/1471-2164-15-143. DOI: https://doi.org/10.1186/1471-2164-15-143

Olatunji TL, Afolayan AJ. (2019) Contributions to the classification of Capsicum annuum L. and Capsicum frutescens L. in West Africa using morphological traits. Notulae Botanicae Horti Agrobotanici Cluj-Napoca, 47(1):135–142. https://doi.org/10.15835/nbha47111204. DOI: https://doi.org/10.15835/nbha47111204

Sudianto YH, Haristu A, Hardhienata M. (2020) Chilli quality classification using deep learning. International Conference on Computer Science and Its Application in Agriculture, 1-5. https://doi.org/10.1109/ICOSICA49951.2020.9243176. DOI: https://doi.org/10.1109/ICOSICA49951.2020.9243176

Abdul Manan AA, Mohd Razman MA, Mohd Khairuddin I, Shapiee MNA. (2020) Chili Plant Classification using Transfer Learning models through Object Detection. Mekatronika, 2(2): 23–27. https://doi.org/10.15282/mekatronika.v2i2.6743. DOI: https://doi.org/10.15282/mekatronika.v2i2.6743

Khuriyati N, Pamungkas AP, Pambudi AA. (2019) The Sorting and Grading of Red Chilli Peppers (Capsicum annuum L.) Using Digital Image Processing. International Journal of Agriculture and Environmental Science, 6(4):17–23. https://doi.org/10.14445/23942568/ijaes-v6i4p104. DOI: https://doi.org/10.14445/23942568/IJAES-V6I4P104

Purwaningsih T, Anjani IA, Utami PB. (2018) Convolutional Neural Networks Implementation for Chili Classification. 2018 International Symposium on Advanced Intelligent Informatics (SAIN), 190–194. https://doi.org/10.1109/SAIN.2018.8673373. DOI: https://doi.org/10.1109/SAIN.2018.8673373

Cruz-Domínguez O, Carrera-Escobedo JL, Guzmán-Valdivia CH, Ortiz-Rivera A, García-Ruiz M, Durán-Muñoz HA, Vidales-Basurto CA, Castaño VM. (2021) A novel method for dried chili pepper classification using artificial intelligence. Journal of Agriculture and Food Research 3:100099. https://doi.org/10.1016/j.jafr.2021.100099. DOI: https://doi.org/10.1016/j.jafr.2021.100099

Miraei Ashtiani SH, Javanmardi S, Jahanbanifard M, Martynenko A, Verbeek FJ. (2021) Detection of mulberry ripeness stages using deep learning models. IEEE Access, 9:100380–100394. https://doi.org/10.1109/ACCESS.2021.3096550. DOI: https://doi.org/10.1109/ACCESS.2021.3096550

Saragih RE, Emanuel AWR. (2021) Banana Ripeness Classification Based on Deep Learning using Convolutional Neural Network. 3rd 2021 East Indonesia Conference on Computer and Information Technology, EIConCIT, 85–89. https://doi.org/10.1109/EIConCIT50028.2021.9431928. DOI: https://doi.org/10.1109/EIConCIT50028.2021.9431928

Zhang Y, Lian J, Fan M, Zheng Y. (2018) Deep indicator for fine-grained classification of banana's ripening stages. Eurasip Journal on Image and Video Processing, 46(2018). https://doi.org/10.1186/s13640-018-0284-8. DOI: https://doi.org/10.1186/s13640-018-0284-8

Foong CC, Meng GK, Tze LL. (2021) Convolutional Neural Network based Rotten Fruit Detection using ResNet50. 2021 IEEE 12th Control and System Graduate Research Colloquium, ICSGRC, 75–80. https://doi.org/10.1109/ICSGRC53186.2021.9515280. DOI: https://doi.org/10.1109/ICSGRC53186.2021.9515280

Pardede J, Sitohang B, Akbar S, Khodra ML. (2021) Implementation of Transfer Learning Using VGG16 on Fruit Ripeness Detection. International Journal of Intelligent Systems and Applications, 13(2):52–61. https://doi.org/10.5815/ijisa.2021.02.04. DOI: https://doi.org/10.5815/ijisa.2021.02.04

Duong LT, Nguyen PT, Di Sipio C, Di Ruscio D. (2020). Automated fruit recognition using EfficientNet and MixNet. Computers and Electronics in Agriculture, 171:105326. https://doi.org/10.1016/J.COMPAG.2020.105326. DOI: https://doi.org/10.1016/j.compag.2020.105326

Suharjito, Elwirehardja GN, Prayoga JS. (2021) Oil palm fresh fruit bunch ripeness classification on mobile devices using deep learning approaches. Computers and Electronics in Agriculture, 188: 106359. https://doi.org/10.1016/j.compag.2021.106359. DOI: https://doi.org/10.1016/j.compag.2021.106359

Szyc K. (2018). Comparison of Different Deep-Learning Methods for Image Classification. INES 2018 - IEEE 22nd International Conference on Intelligent Engineering Systems, Proceedings, 000341–000346. https://doi.org/10.1109/INES.2018.8523931. DOI: https://doi.org/10.1109/INES.2018.8523931

Zhu W, Braun B, Chiang LH, Romagnoli JA. (2021) Investigation of transfer learning for image classification and impact on training sample size. Chemometrics and Intelligent Laboratory Systems, 211:104269. https://doi.org/10.1016/j.chemolab.2021.104269. DOI: https://doi.org/10.1016/j.chemolab.2021.104269

Tan M, Le QV. (2019) EfficientNet: Rethinking model scaling for convolutional neural networks. 36th International Conference on Machine Learning, ICML, 10691–10700.

Atila Ü, Uçar M, Akyol K, Uçar E. (2021) Plant leaf disease classification using EfficientNet deep learning model. Ecological Informatics, 61:101182. https://doi.org/10.1016/j.ecoinf.2020.101182. DOI: https://doi.org/10.1016/j.ecoinf.2020.101182

Kumar AC, Mubarak DMN. (2021) Classification of Early Stages of Esophageal Cancer Using Transfer Learning. IRBM, 43(4):251–258. https://doi.org/10.1016/j.irbm.2021.10.003. DOI: https://doi.org/10.1016/j.irbm.2021.10.003

Huang ML, Chuang TC, Liao YC. (2022). Application of transfer learning and image augmentation technology for tomato pest identification. Sustainable Computing: Informatics and Systems, 33: 100646. https://doi.org/10.1016/j.suscom.2021.100646. DOI: https://doi.org/10.1016/j.suscom.2021.100646

Masters D, Luschi C. (2018) Revisiting Small Batch Training for Deep Neural Networks. http://arxiv.org/abs/1804.07612.