Apricot Disease Detection with Convolutional Neural Network

Yıl 2026, Cilt: 15 Sayı: 1 , 31 - 42 , 30.03.2026

Zeynep Mine Alçin , Muzaffer Aslan

https://doi.org/10.46810/tdfd.1774549

https://izlik.org/JA25AE46UR

Öz

Apricot is a stone fruit grown in temperate climates and possesses high economic value globally. However, diseases and pests pose substantial threats to apricot production, undermining both crop quality and overall yield. As these pressures intensify, they further compromise fruit development and reduce harvest quantities, negatively affecting market value and productivity. In particular, canker, coryneum beijerinckii, drying symptom, and monilinia laxa stand out as the four main diseases that markedly reduce quality and yield worldwide. Therefore, early diagnosis and targeted management strategies for these diseases are critically important for preventing epidemic spread and ensuring efficient resource utilization. In this study, a novel deep learning-based convolutional neural network model is proposed for the detection of diseased apricot images. The proposed CNN model was tested on a publicly available dataset, meticulously compiled under real field conditions and encompassing the aforementioned four apricot diseases. The proposed model achieved a high accuracy rate of 97.74% in the detection and classification of diseases. It provided 8.1% to 21.16% higher accuracy than traditional image processing-based approaches in the literature. Furthermore, the final model achieved 0.44% to 23.87% higher performance compared to some CNN models. These results indicate that the proposed CNN model can provide rapid and reliable decision support in disease detection.

Anahtar Kelimeler

Detection of apricot diseases , Deep Learning , Convolutional neural networks

Evrişimli Sinir Ağı ile Kayısı Hastalıklarının Tespiti

https://izlik.org/JA25AE46UR

Öz

Kayısı, ılıman iklimlerde yetişen ve küresel ölçekte yüksek ekonomik katma değere sahip çekirdekli bir meyvedir. Ancak hastalık ve zararlılar kayısı üretimini önemli ölçüde tehdit ederek kalite ve verimi olumsuz etkilemektedir. Özellikle kanser, çil hastalığı, kuruma belirtileri ve monilya hastalığı, dünya genelinde kalite ve verimi belirgin biçimde düşüren dört ana hastalık olarak öne çıkmaktadır. Bu nedenle, bu hastalıkların erken teşhis ve hedefe yönelik yönetim stratejileri, epidemik yayılımın önlenmesi ve kaynakların verimli kullanımı açısından kritik öneme sahiptir. Çalışmada, hastalıklı kayısı görüntülerini tespiti için derin öğrenme temelli yeni bir evrişimli sinir ağ modeli önerilmiştir. Önerilen CNN modeli, gerçek saha koşulları altında titizlikle derlenmiş ve söz konusu dört hastalığı kapsayan halka açık bir veri seti üzerinde test edilmiştir. Önerilen model, hastalıkların tespiti ve sınıflandırmasında %97.74 gibi yüksek bir doğruluk oranına ulaşmıştır. Literatürdeki geleneksel görüntü işleme tabanlı yaklaşımlarda %8.1 ile 21.16 daha yüksek doğruluk sağlamıştır. Ayrıca son model bazı CNN Modelleri ile karşılaştırıldığında %0.44 ile %23.87 oranında daha yüksek başarım elde etmiştir. Bu sonuçlar, önerilen CNN modelinin hastalık tespitinde hızlı, güvenilir karar desteği sağlayabileceğini ortaya koymaktadır.

Anahtar Kelimeler

Kayısı hastalıkların tespiti , Derin öğrenme , Evrişimli sinir ağları

Kaynakça

• Durmaz S, Ağır HB. Assessing the Effect of El Niño–Southern Oscillation on Apricot Yield in Malatya Province, Türkiye. Appl Fruit Sci 2024;66:2231–8.
• Li M, Tao Z, Yan W, Lin S, Feng K, Zhang Z, et al. Apnet: Lightweight network for apricot tree disease and pest detection in real-world complex backgrounds. Plant Methods 2025;21:4.
• Uzundumlu AS, Karabacak T, Ali A. Apricot Production Forecast of the Leading Countries in The Period of 2018-2025. Emirates J Food Agric 2021:682.
• Amari K, Ruiz D, Gómez G, Sánchez-Pina MA, Pallás V, Egea J. An important new apricot disease in Spain is associated with Hop stunt viroid infection. Eur J Plant Pathol 2007;118:173–81.
• Han B, Duan P, Zhou C, Su X, Yang Z, Zhou S, et al. Implementation and Evaluation of Spatial Attention Mechanism in Apricot Disease Detection Using Adaptive Sampling Latent Variable Network. Plants 2024;13:1681.
• Arı B, Arı A, Şengür A, Tuncer SA. Classification of Apricot Leaves with Extreme Learning Machines Using Deep Features. 2019 1st Int. Informatics Softw. Eng. Conf., IEEE; 2019, p. 1–5.
• TURKOGLU M, HANBAY D. Apricot Disease Identification based on Attributes Obtained from Deep Learning Algorithms. 2018 Int. Conf. Artif. Intell. Data Process., IEEE; 2018, p. 1–4.
• Turkoglu M, Yanikoğlu B, Hanbay D. PlantDiseaseNet: convolutional neural network ensemble for plant disease and pest detection. Signal, Image Video Process 2022;16:301–9.
• Ahmad R, Ali A, Abbas A. Distribution and Etiology of Gummosis Syndrome Associated with Apricot Fruit Trees of Gilgit-Baltistan, Pakistan. Phytopathogenomics Dis Control 2024;3:77–86.
• Muhammad M, Hussain A, Ali S, Akram W, Roomi I, Faiz F, et al. Geostatistical Analysis of Apricot Shot Hole Disease and Influence Factors in District Nagar, Gilgit-Baltistan, Pakistan. Int J Phytopathol 2022;11:227–38.
• Kim GH, Jo KY, Shin JS, Shin GH, Koh YJ. Epidemiological Characteristics of Scab of Japanese Apricot in Korea. Plant Pathol J 2017;33:450–7..
• Mohanty SP, Hughes DP, Salathé M. Using Deep Learning for Image-Based Plant Disease Detection. Front Plant Sci 2016;7.
• Sladojevic S, Arsenovic M, Anderla A, Culibrk D, Stefanovic D. Deep Neural Networks Based Recognition of Plant Diseases by Leaf Image Classification. Comput Intell Neurosci 2016;2016:1–11. TÜRKOĞLU M, HANBAY K, SARAÇ SİVRİKAYA I, HANBAY D. Derin Evrişimsel Sinir Ağı Kullanılarak Kayısı Hastalıklarının Sınıflandırılması. Bitlis Eren Üniversitesi Fen Bilim Derg 2020;9:334–45.
• Lecun Y, Bengio Y, Hinton G. Deep learning. Nature 2015;521:436–44.
• Rawat W, Wang Z. Deep Convolutional Neural Networks for Image Classification: A Comprehensive Review. Neural Comput 2017;29:2352–449.
• Saxena A. An Introduction to Convolutional Neural Networks. Int J Res Appl Sci Eng Technol 2022;10:943–7.
• Krois J, Ekert T, Meinhold L, Golla T, Kharbot B, Wittemeier A, et al. Deep Learning for the Radiographic Detection of Periodontal Bone Loss. Sci Rep 2019;9:8495.
• Zhou C, Zhang F, Nacpil EJC, Wang Z, Xu F-X. Driver Steering Intention Prediction for Human-Machine Shared Systems of Intelligent Vehicles Based on CNN-GRU Network. Sensors 2025;25:3224.
• Soylu E, Gül S, Koca KA, Türkoğlu M, Terzi M, Şengür A. Speech signal-based accurate neurological disorders detection using convolutional neural network and recurrent neural network based deep network. Eng Appl Artif Intell 2025;149:110558.
• Abdulsattar NS, Hussain MN. Facial expression recognition using HOG and LBP features with convolutional neural network. Bull Electr Eng Informatics 2022;11:1350–7.
• Bayar B, Stamm MC. Design Principles of Convolutional Neural Networks for Multimedia Forensics. Electron Imaging 2017;29:77–86.
• Hu W, Huang Y, Wei L, Zhang F, Li H. Deep Convolutional Neural Networks for Hyperspectral Image Classification. J Sensors 2015;2015:1–12..
• Zhang Z, Duan F, Sole-Casals J, Dinares-Ferran J, Cichocki A, Yang Z, et al. A Novel Deep Learning Approach With Data Augmentation to Classify Motor Imagery Signals. IEEE Access 2019;7:15945–54.
• Purwono P, Ma’arif A, Rahmaniar W, Fathurrahman HIK, Frisky AZK, Haq QM ul. Understanding of Convolutional Neural Network (CNN): A Review. Int J Robot Control Syst 2023;2:739–48.