An Intelligent Automated System for Fruits [Apple] Disease Detection

Abstract

Production of crop is an important aspect in the field of agriculture. The production rate is directly dependent on the quality of the crop. Crop Infection factor becomes an important aspect of Quality. Most of the time traditional approaches such as Naked eye Survey, Survey through Expert etc. are used to detect the diseases during cultivation. This process requires huge processing time and cost. Automatic detection of severity is essential for high quality production. Apple fruit diseases significantly impact crop yield and quality, necessitating early and accurate detection. Traditional diagnostic methods are manual, time-consuming, and prone to human error. This study proposes a robust and automated disease detection system using machine learning and deep learning techniques, enhanced with transfer learning. Models including Support Vector Machine (SVM), Random Forest, Convolutional Neural Networks (CNN), and pre-trained architectures such as ResNet and MobileNet are analyzed.

Introduction

Agriculture is a cornerstone of India’s economy, with apple cultivation being especially critical in regions like Kashmir. Plant diseases—caused by bacteria, fungi, viruses, and environmental factors—pose a major threat to crop yield and farmer livelihoods. Traditional detection methods, which rely on visual inspection or expert guidance, are slow, subjective, and often inaccurate.

Advances in machine learning (ML) and deep learning (DL), particularly Convolutional Neural Networks (CNNs), have enabled automated, accurate, and scalable detection of fruit diseases. CNNs outperform traditional ML methods by automatically extracting features from images, achieving higher accuracy even under variable lighting and real-world conditions. Popular CNN architectures for apple disease detection include ResNet, VGG, DenseNet, MobileNet, EfficientNet, and hybrid models, with accuracies ranging from 92% to 99%.

The methodology involves data collection (e.g., PlantVillage datasets and real-field images), image preprocessing (resizing, normalization, augmentation), feature extraction, and model training/testing using metrics like accuracy, precision, recall, and F1-score. Challenges remain in dataset quality, symptom variability, environmental effects, computational costs, and real-world deployment. Among evaluated models, CNNs consistently achieve the best performance, making them the most effective approach for apple disease detection.

Future directions emphasize improving dataset diversity, real-time applicability, and robust model architectures to ensure scalable, accurate, and practical solutions for automated crop disease management.

Conclusion

This research presents a machine learning-based approach for detecting diseases in apple fruits. The study concludes that CNN models provide superior performance compared to traditional techniques such as SVM and Random Forest. The proposed system can help farmers in early disease detection, reducing crop loss and improving productivity. Research in fruit disease detection and classification has made significant progress through the use of machine learning, deep learning and image processing. The studies made have contributed unique insights to address key challenges in accurately identifying fruit diseases and enhancing sustainable agricultural productivity. Early approaches that combined traditional image processing with ML techniques especially for specific fruits like pomegranates demonstrated promising results under controlled conditions. However, these methods were limited by the manual nature of feature extraction, leading to poor adaptability across varying environments.

References

[1] Abbas, I., Liu, J., Amin, M., Tariq, A. and Tunio, M.H. (2021). Strawberry fungal leaf scorch disease identification in real-time strawberry field using deep learning architectures. Plants. 10(12): 2643. [2] Amara, J., Bouaziz, B. and Algergawy, A. (2017). A deep learning- based approach for banana leaf diseases classification. In Datenbanksystemefür Business, Technologie und Web (BTW 2017)-Workshopband. Gesellscha- ftfürInformatik eV. (pp. 79-88). [3] Awate, A., Deshmankar, D., Amrutkar, G., Bagul, U. and Sonavane, S. (2015). Fruit Disease Detection using Colour, Texture Analysis and ANN. In 2015 International Conference on Green Computing and Internet of Things (ICGCIoT) (pp. 970-975). IEEE. [4] Ayyub, S.R.N.M. and Manjramkar, A. (2019, March). Fruit Disease Classification and Identification using Image Processing. In 2019 3rd International Conference on Computing Methodo- logies and Communication (ICCMC) IEEE. (pp. 754-758). [5] Barbedo, J.G.A. (2016). A review on the main challenges in automatic plant disease identification based on visible range images. Biosystems Engineering. 144: 52-6 [6] Samajpati B.J., Degadwala S.D. A Survey on Apple Fruit Diseases Detection and Classification. Int. J. Comput. Appl. 2015;130:25–32. doi: 10.5120/ijca2015907153. [DOI] [Google Scholar] [7] Khirade S.D., Patil A. Plant disease detection using image processing; Proceedings of the 2015 International Conference on Computing Communication Control and Automation; Pune, India. 26–27 February 2015; pp. 768–771. [DOI] [Google Scholar] [8] Rao A., Kulkarni S. A Hybrid Approach for Plant Leaf Disease Detection and Classification Using Digital Image Processing Methods. Int. J. Electr. Eng. Educ. 2020:0020720920953126. doi: 10.1177/0020720920953126. [DOI] [Google Scholar] [9] Phadikar S., Sil J., Das A.K. Rice diseases classification using feature selection and rule generation techniques. Comput. Electron. Agric. 2013;90:76–85. doi: 10.1016/j.compag.2012.11.001. [DOI] [Google Scholar] [10] Rastogi A., Arora R., Sharma S. Leaf disease detection and grading using computer vision technology & fuzzy logic; Proceedings of the 2015 2nd International Conference on Signal Processing and Integrated Networks (SPIN); Noida, India. 19–20 February 2015; pp. 500–505. [DOI] [Google Scholar] [11] Hughes, D. and Salathé, M. (2015). An open access repository of images on plant health to enable thedevelopment of mobile disease diagnostics. arXiv preprint arXiv:1511.08060. [12] Dr M.S Chavan and Thorat Asmita “Diagnosis of Plant diseases using Neural Network”. [13] Pooja Kantale and Shubhada Thakare “Pomegranate Disease Classification using ADA Boost Ensemble Algorithm” [14] Raju Hosakoti, Soma Kumar, Padmaja Jain, in 2021 proposed “Disease detection in fruits using Deep Learning” [15] Poonam Dhiman, Vinay kukreja, proposed “A novel deep learning model for Detection of severity level of the disease in citrus fruits” [16] Xiaoting Liang, Xueying Jia, Wenqian Huang, Xin He, Lianjie Li proposed “Real –Time Grading of Defect Apples Using Semantic Segmentation Combination with a Pruned YOLO V4 Network” [17] Tomato Diseases and Pests Detection Based on Improved Yolo V3 Convolutional Neural Network Jun Liu and Xuewei Wang

Copyright

Copyright © 2026 Suneeta Devi, Dr. Rohait Kumar Singhal. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.