ABSTRACT
Fruit grading is a significant problem in the fruit industry because each maturity stage of the fruit represents a distinct economic worth. Banana is one of the most mass-produced fruits that must be visually classified. However, because human eye perception varies, precise classification using a machine is necessary to standardise the grading system. This research develops a four-layered CNN deep-learning model to classify bananas into seven ripening stages. To train the model, we employed Mazen and Nashat dataset and expanded it using data augmentation techniques. As a baseline, we trained a basic four-layer CNN model and achieved 88.2% of accuracy in fine-grained categorisation due to the similarity of the adjacent ripening class. To enhance the accuracy of our basic model, we applied a parameter optimisation approach to get the best hyper-parameters for the profound banana ripeness indicator. As a result, the time-constrained parameter optimisation method that we utilised successfully increased the model accuracy up to 91.2% and the F1 score at 90.5%, which is satisfactory for fine-grained banana classification compared to the previous research.
- James Bergstra, James Bergstra@umontreal Ca, and Yoshua Bengio@umontreal Ca. 2012. Random Search for Hyper-Parameter Optimization Yoshua Bengio., 281-305 pages. http://scikit-learn.sourceforge.net.Google Scholar
- C. Dewi, E. Arisoesilaningsih, W. F. Mahmudy, and Solimun. 2022. Identifying of unripe Ambon and Hijau banana fruits using computer vision and extreme learning machine classifier. IOP Conference Series: Earth and Environmental Science 951, 1 (jan 2022), 012031. https://doi.org/10.1088/1755-1315/951/1/012031Google ScholarCross Ref
- FAO. 2022. International Trade Banana: Market Review Preliminary Results 2021. Technical Report. United Nations, Rome.Google Scholar
- Indika Fernando, Jiangang Fei, Roger Stanley, Hossein Enshaei, and Alieta Eyles. 2019. Quality deterioration of bananas in the post-harvest supply chain- an empirical study. Modern Supply Chain Research and Applications 1, 2(2019), 135–154. https://doi.org/10.1108/mscra-05-2019-0012Google ScholarCross Ref
- M Hailu, Ts Workneh, and D Belew. 2013. Review on postharvest technology of banana fruit. African Journal of... 12, 7 (2013), 635–647. https://doi.org/10.5897/AJBX12.020Google Scholar
- Jun Cai Hou, Yao Hua Hu, Li Xia Hou, Kang Quan Guo, and Takaaki Satake. 2015. Classification of ripening stages of bananas based on support vector machine. International Journal of Agricultural and Biological Engineering 8, 6(2015), 99–103. https://doi.org/10.3965/j.ijabe.20150806.1275Google Scholar
- Fatma M. A. Mazen and Ahmed A. Nashat. 2019. Ripeness Classification of Bananas Using an Artificial Neural Network. Arabian Journal for Science and Engineering 44, 8 (aug 2019), 6901–6910. https://doi.org/10.1007/s13369-018-03695-5Google ScholarCross Ref
- Parkash Meghwar and Syeda Mahvish Zahra. 2021. An Overview on Vital role of Banana and its Valorization. International Journal of Food Chemistry and Human NutritionDecember (2021).Google Scholar
- F. Mendoza and J.M. Aguilera. 2006. Application of Image Analysis for Classification of Ripening Bananas. Journal of Food Science 69, 9 (may 2006), E471–E477. https://doi.org/10.1111/j.1365-2621.2004.tb09932.xGoogle ScholarCross Ref
- Jiangong Ni, Jiyue Gao, Limiao Deng, and Zhongzhi Han. 2020. Monitoring the Change Process of Banana Freshness by GoogLeNet. IEEE Access 8(2020), 228369–228376. https://doi.org/10.1109/ACCESS.2020.3045394Google ScholarCross Ref
- Ebenezer Obaloluwa Olaniyi, Oyebade Kayode Oyedotun, and Khashman Adnan. 2017. Intelligent Grading System for Banana Fruit Using Neural Network Arbitration. Journal of Food Process Engineering 40, 1 (2017), e12335. https://doi.org/10.1111/jfpe.12335 arXiv:https://onlinelibrary.wiley.com/doi/pdf/10.1111/jfpe.12335Google ScholarCross Ref
- Chayan Saha, Md Ahamed, Md Hosen, Rajesh Nandi, and Mahjabin Kabir. 2021. Post-harvest losses of banana in fresh produce marketing chain in Tangail District of Bangladesh. Journal of Bangladesh Agricultural University 19, 3(2021), 389. https://doi.org/10.5455/JBAU.74902Google ScholarCross Ref
- Mauro Santoyo-Mora, Agustin Sancen-Plaza, Alejandro Espinosa-Calderon, Alejandro Israel Barranco-Gutierrez, and Juan Prado-Olivarez. 2019. Nondestructive Quantification of the Ripening Process in Banana (Musa AAB Simmonds) Using Multispectral Imaging. Journal of Sensors 2019(2019), 6742896. https://doi.org/10.1155/2019/6742896Google ScholarCross Ref
- Raymond Erz Saragih and Andi W. R. Emanuel. 2021. Banana Ripeness Classification Based on Deep Learning using Convolutional Neural Network. In 2021 3rd East Indonesia Conference on Computer and Information Technology (EIConCIT). IEEE, 85–89. https://doi.org/10.1109/EIConCIT50028.2021.9431928Google ScholarCross Ref
- N. Saranya, K. Srinivasan, and S. K. Pravin Kumar. 2022. Banana ripeness stage identification: a deep learning approach. Journal of Ambient Intelligence and Humanized Computing 13, 8 (aug 2022), 4033–4039. https://doi.org/10.1007/s12652-021-03267-wGoogle ScholarCross Ref
- W M C B Wasala, C a K Dissanayake, D a N Dharmasena, C R Gunawardane, and T M R Dissanayake. 2014. Postharvest losses, current issues and demand for postharvest technologies for loss management in the main banana supply chains in Sri Lanka. Journal of Postharvest Technology 02, 01 (2014), 80–87.Google Scholar
- Yan Zhang, Jian Lian, Mingqu Fan, and Yuanjie Zheng. 2018. Deep indicator for fine-grained classification of banana’s ripening stages. EURASIP Journal on Image and Video Processing 2018, 1 (dec 2018), 46. https://doi.org/10.1186/s13640-018-0284-8Google ScholarCross Ref
Index Terms
- CNN Model with Parameter Optimisation for Fine-Grained Banana Ripening Stage Classification
Recommendations
Selection of genetic algorithm operators for urban drainage model parameter optimisation
Recently, genetic algorithms (GAs) have proven to be successful and efficient in identifying the optimal parameters for water resource modelling applications. However, in order to produce efficient and robust solutions, proper selection of GA operators ...
Improved XGBoost model based on genetic algorithm
An optimised XGBoost model based on genetic algorithm to search for optimal parameter combinations is proposed in this paper. It was proved that the improved algorithm has better classification effect than existing approaches through the liver disease ...
Differential evolution-based parameters optimisation and feature selection for support vector machine
This paper addresses the problem of SVM classification optimisation. For this purpose, the authors propose an SVM classification system based on differential evolution DE to improve the generalisation performance of the SVM classifier. In the ...
Comments