Papaya is a widespread fruit, but its production is heavily affected by a variety of diseases. Early detection of diseased papayas can help farmers take precautions to minimize harvest losses. This project presents a Papaya Fruit Disease Recognition System with machine learning technology (machine learning) to classify papayas as healthy or disease.
Models were meticulously trained on diverse datasets to ensure robust performance of the disease. To evaluate the model output, K-fold Cross validation was used to test each segmentation method for each classification. Under all combinations, segmentation of fuzzy C-means combined with random forest achieved the highest accuracy. Secondly, Streamlit was provided to this optimized model, so users uploaded Papaya images and received the disease of instant disease classification.
Introduction
This research focuses on developing an AI-based system for detecting diseases in papaya fruits using image segmentation and machine learning (ML) classification techniques. The goal is to automate the detection process to improve accuracy, reduce manual errors, and increase efficiency.
A dataset of 7,000 images (3,500 healthy and 3,500 diseased papayas) was used. After preprocessing (resizing and normalization), various image segmentation methods (threshold-based, edge-based, cluster-based, and fuzzy segmentation) and classifiers (Decision Tree, KNN, Naive Bayes, Logistic Regression, Random Forest) were tested. A voting mechanism was employed to combine classifier outputs.
Key Findings:
Fuzzy C-Means Segmentation + Random Forest yielded the highest accuracy of 97.14% and an F1-score of 97.08%, outperforming other segmentation-classifier combinations.
The system was deployed using Streamlit, enabling users to upload images and receive immediate classification (healthy or diseased).
The approach emphasizes real-time application, reliability, and practical usability in agricultural technology for enhancing papaya crop productivity and sustainability.
Technical Contributions:
Applied k-fold cross-validation to evaluate model robustness.
Used multiple segmentation and classification techniques to compare performance.
Developed a lightweight, web-accessible system for real-time disease detection.
Demonstrated that combining Fuzzy Segmentation with ensemble classifiers significantly improves disease classification accuracy.
Conclusion
The research paper on “IMAGE-BASED PAPAYA FRUIT HEALTH CLASSIFICATION USING MACHINE LEARNING” leveraged a machine learning-based papaya disease classification system using advanced image segmentation and classification techniques.
Out of all the classifiers evaluated, the combination of Random Forest and Fuzzy C-Means Clustering reached the highest accuracy rate of 97.14%, establishing it as the most efficient model for disease detection. This system was implemented using Streamlit, offering an interactive and user-friendly platform for classification in real-time.
The results indicate that machine learning can greatly assist in the automated detection of diseases, allowing farmers to recognize impacted fruits early and implement preventive actions. Upcoming efforts might include deep learning techniques, mobile apps for real-time use, and the incorporation of IoT-based monitoring systems to improve both accuracy and usability.
References
[1] Ahad, M. T., Mustofa, S., Sarker, A., & Emon, Y. R. (2024). Bdpapayaleaf: A Dataset of Papaya Leaf for Disease Detection, Classification, and Analysis. Classification, and Analysis..
[2] De Moraes, J. L., de Oliveira Neto, J., Badue, C., Oliveira-Santos, T., & de Souza, A. F. (2023). Yolo-Papaya: A papaya fruit disease detector and classifier using CNNs and convolutional block attention modules. Electronics, 12(10), 2202.
[3] Priya Mehta, Rohit S. Kumar, Snehal Reddy (2023). “Comparative Analysis of CNN and SVM for Papaya Disease Classification”, Springer Journal of Computational Intelligence..
[4] Islam, Md Ashiqul, et al. \"Machine learning based image classification of papaya disease recognition.\" 2020 4th International Conference on Electronics, Communication and Aerospace Technology (ICECA). IEEE, 2020.
[5] Behera, S. K., Rath, A. K., & Sethy, P. K. (2021). Maturity status classification of papaya fruits based on machine learning and transfer learning approach. Information Processing in Agriculture, 8(2), 244-250.
[6] Praveena, S., Pavithra, S. M., Kumar, A. D. V., & Veeresha, P. (2024). CNN-based Indian medicinal leaf type identification and medical use recommendation. Neural Computing and Applications, 36(10), 5399-5412.
[7] Nair, B. B., Rushab, D., & Kruthik, S. (2024, November). A Comparison of Deep Learning Based Methods for Grading Papaya Fruit by Maturity Level. In 2024 3rd Edition of IEEE Delhi Section Flagship Conference (DELCON) (pp. 1-5). IEEE.
[8] Gani, R., Rashid, M. R. A., Ahmed, J., Isty, M. N., Islam, M., Hasan, M., ... & Ali, M. S. (2024). Smartphone image dataset to distinguish healthy and unhealthy leaves in papaya orchards in Bangladesh. Data in Brief, 55, 110599.
[9] Ratha, A. K., Barpanda, N. K., Sethy, P. K., & Behera, S. K. (2023). Papaya fruit maturity estimation using wavelet and ConvNET. Ingenierie des Systemes d\'Information, 28(1), 175.
[10] Apostolopoulos, I. D., Tzani, M., & Aznaouridis, S. I. (2023). A general machine learning model for assessing fruit quality using deep image features. AI, 4(4), 812-830.
[11] Bucao, R. D. C., Tapado, E. H., & Hortinela, C. C. (2024, June). Grading of Carica Papaya Using Image Processing and Support Vector Machine on Raspberry PI. In 2024 IEEE International Conference on Automatic Control and Intelligent Systems (I2CACIS) (pp. 175-180). IEEE.
[12] Al-Masawabe, M. M., Samhan, L. F., AlFarra, A. H., Aslem, Y. E., & Abu-Naser, S. S. (2021). Papaya Maturity Classifications using Deep Convolutional Neural Networks.
[13] Kumar, R. R., Jain, A. K., Sharma, V., Das, P., & Sahni, P. (2024, June). Enhancing Precision in Papaya Crop Health Management: A CNN-Based Approach for Early Disease Detection and Classification. In 2024 IEEE International Conference on Information Technology, Electronics and Intelligent Communication Systems (ICITEICS) (pp. 1-6). IEEE.
[14] Salim, N. O., & Mohammed, A. K. (2024). Comparative Analysis of Classical Machine Learning and Deep Learning Methods for Fruit Image Recognition and Classification. Traitement du Signal, 41(3).
[15] Kumaran, P., Tirkey, A., Kumar, U., & Kumar, K. V. (2024, March). Anthracnose Disease Detection in Papaya Fruit Using Machine Learning Techniques. In 2024 5th International Conference on Innovative Trends in Information Technology (ICITIIT) (pp. 1-6). IEEE.
[16] Kumar, R. R., Jain, A. K., Sharma, V., Das, P., & Sahni, P. (2024, June). Enhancing Precision in Papaya Crop Health Management: A CNN-Based Approach for Early Disease Detection and Classification. In 2024 IEEE International Conference on Information Technology, Electronics and Intelligent Communication Systems (ICITEICS) (pp. 1-6). IEEE.