Rose Leaf Diseases Detection Using Convolutional Neural Networks with Texture-Based Patch-Wise Selection

Abstract

In agricultural, early detection of leaf diseases enables control targeted at less chemical input, environmental impact, and expense. Towards this end, the see-and-spray strategy has been sought more intensely in recent decades in applications such as spot/precision spraying and UAV-based treatments. Because roses are converted into high-value medicinal and aromatic products, early identification of leaf diseases prevents quality losses and yield reductions, allowing selective pesticide application and timely intervention. In turn, this optimizes input use and lowers production costs. Convolutional neural networks (CNNs) are widely used for leaf disease detection; however, resizing images to meet fixed input dimensions can cause information loss and hinder the capture of subtle disease cues. To address this limitation, we propose a new approach that combines texture-based patch selection with CNN models for classifying rose leaf diseases. Texture-based patch selection provides a patch-wise pipeline that preserves fine lesion patterns otherwise lost when full images are resized for CNN input. A public rose leaf disease dataset was used. Each dataset image was initially divided into four patches of 224×224 pixels, and the patches with the highest information content were selected using Gray-Level Co-occurrence Matrix (GLCM), Local Binary Patterns (LBP), and Histogram of Oriented Gradients (HOG). The selected patches were classified using nine different CNN models. These CNN models were evaluated using transfer learning and five-fold cross-validation. Experimental results show that LBP-based patch-wise selection combined with DenseNet121 model achieved an accuracy of 95.73 ± 0.87%, while GLCM-based selection with DenseNet201 achieved 95.48 ± 0.70%. The findings indicate that texture-based patch selection can aid the detection of diseased regions on rose leaves and enable targeted pesticide application.

Keywords

• Rose leaf disease
• patch-wise selection
• texture analysis
• convolutional neural network
• detection
• precision spraying

Doku Tabanlı Yama-Bilgi Seçimi ile Evrişimsel Sinir Ağları Kullanılarak Gül Yaprağı Hastalıklarının Tespiti

Abstract

Tarımda, yaprak hastalıklarının erken tespiti, hedefe yönelik mücadeleyi mümkün kılmakla birlikte kimyasal kullanımını, çevresel etkiyi ve maliyetleri azaltabilir. Bu amaç doğrultusunda son zamanlarda noktasal veya drone ilaçlamaları gibi uygulamalarda algıla ve ilaçla mantığı popülerleşmiştir. Gül, tıbbi ve aromatik bir bitki olarak yüksek katma değerli ürünlere dönüştürüldüğünden, yaprak hastalıklarının erken tespiti; kalite kayıplarını ve verim düşüşünü önleyerek selektif ilaçlama ve zamanında müdahale olanağı sunar. Böylece girdi kullanımını optimize eder ve üretim maliyetlerini azaltır. Yaprak hastalıklarının tespitinde yaygın olarak evrişimsel sinir ağı (CNN) modelleri kullanılmaktadır. Ancak, modellerin giriş boyutuna uyum sağlamak amacıyla görüntü boyutlarının küçültülmesi, görüntülerde bilgi kayıplarının yaşanmasına ve bazı ince hastalık detaylarının belirlenememesine neden olmaktadır. Bu sorunu çözmek amacıyla çalışmada, gül yaprağı hastalıklarının sınıflandırılması için doku tabanlı yama seçim stratejisini CNN modelleriyle birleştiren yeni bir yaklaşım önerilmiştir. Doku tabanlı yama seçimi, bu çalışmadaki görüntülerin CNN için yeniden boyutlandırılmasında kaybolan ince lezyon desenlerini koruyan yama bazlı bir işlem hattı sunmaktadır. Çalışmada, gül yaprak hastalıklarını içeren açık erişimli bir veriseti kullanılmıştır. Öncelikle verisetindeki her bir görüntü 224 x 224 piksel boyutundaki dört yamaya bölünmüş, ardından Gri Düzeyi Eş Oluşum Matrisi (GLCM), Yerel İkili Desenler (LBP) ve Yönlendirilmiş Gradyanların Histogramı (HOG) teknikleri kullanılarak en yüksek bilgi yoğunluğuna sahip görüntü yamaları seçilmiştir. Seçilen bu yamalar, 9 farklı CNN modeliyle sınıflandırılmıştır. Bu CNN modelleri, transfer öğrenme ve 5-kat çapraz doğrulama yöntemi kullanılarak değerlendirilmiştir. Deneysel sonuçlara göre LBP-doku tabanlı yama seçimi ve DenseNet121 modelinde %95.73 ± 0.87 ve GLCM-doku tabanlı yama seçimi ve DenseNet201 modelinde ise %95.48 ± 0.70 doğruluk değerleri elde edilmiştir. Elde edilen bulgular, doku tabanlı yama seçiminin gül yaprağı üzerindeki hastalıklı bölgelerin tespiti ve hedefe yönelik ilaç uygulamasına katkı sağlayabileceğini ortaya koymaktadır.

Keywords

• Gül yaprak hastalığı
• yama-bilgi seçimi
• doku analizi
• evrişimli sinir ağı
• tespit
• hassas püskürtme

References

1. [1] A. Rajbongshi, T. Sarker, M. M. Ahamad, and M. M. Rahman, "Rose diseases recognition using MobileNet," in 2020 4th International Symposium on Multidisciplinary Studies and Innovative Technologies (ISMSIT), Turkey, 2020, IEEE, pp. 1-7.
2. [2] R. Sobolu, L. Stanca, D. Pusta, I. Pop, and M. Cordea, "Image Processing Technique Applying To Detect Black Spot And Rust Diseases At Roses," Managerial Challenges of the Contemporary Society. Proceedings, vol. 12, no. 1, pp. 68-73, 2019.
3. [3] S. Sazzad, A. Rajbongshi, R. Shakil, B. Akter, and M. S. Kaiser, "RoseNet: Rose leave dataset for the development of an automation system to recognize the diseases of rose," Data in Brief, vol. 44, p. 108497, 2022.
4. [4] J. Sharma, "Enhanced Rose Leaf Disease Classification Using Vision Transformer (ViT-B/16) Detecting Black Spot, Downy Mildew, and Healthy Leaves for Improved Plant Health Management," in 2024 5th International Conference on Data Intelligence and Cognitive Informatics (ICDICI), India, 2024, IEEE, pp. 52-56.
5. [5] E. R. Reddy, S. D. Satturi, M. Harshini, and S. Shaik, "Rose Plant Leaf Disease Recognition Using Machine Learning Methodologies," Asian Journal of Research in Computer Science, vol. 17, no. 11, pp. 65-72, 2024.
6. [6] J. Ma, L. Pang, L. Yan, and J. Xiao, "Detection of black spot of rose based on hyperspectral imaging and convolutional neural network," AgriEngineering, vol. 2, no. 4, pp. 556-567, 2020.
7. [7] R. Kaur et al., "Pesticides: An alarming detrimental to health and environment," Science of the Total Environment, vol. 915, p. 170113, 2024.
8. [8] P. Pokharel, A. Sharda, D. Flippo, and K. Ladino, "Design and systematic evaluation of an Under-Canopy robotic spray system for row crops," Smart Agricultural Technology, vol. 8, p. 100510, 2024.

9. [9] H. Asaei, A. Jafari, and M. Loghavi, "Site-specific orchard sprayer equipped with machine vision for chemical usage management," Computers and Electronics in Agriculture, vol. 162, pp. 431-439, 2019.
10. [10] J. Llorens, E. Gil, J. Llop, and A. Escolà, "Variable rate dosing in precision viticulture: Use of electronic devices to improve application efficiency," Crop protection, vol. 29, no. 3, pp. 239-248, 2010.
11. [11] S. Sladojevic, M. Arsenovic, A. Anderla, D. Culibrk, and D. Stefanovic, "Deep neural networks based recognition of plant diseases by leaf image classification," Computational intelligence and neuroscience, vol. 2016, no. 1, p. 3289801, 2016.
12. [12] V. Solanki, R. Ahuja, V. Khullar, and S. Thapliyal, "Optimizing Rose Leaf Disease Detection Performance via Explainable Transfer Learning: A Comparative Analysis," in 2024 International Conference on Electrical Electronics and Computing Technologies (ICEECT), India, 2024, IEEE, pp. 1-6.
13. [13] T. Ashine, "Image-Based Rose Leaf Diseases Detection Using Deep Learning," St. Mary's University, 2024.
14. [14] N. Akter, M. N. Ullah, M. S. Rifat, and A. S. Sikder, "Optimized Convolutional Neural Network Architecture for High-Accuracy Classification of Rose Leaf Diseases: Neural Network on Rose Leaf Diseases," International Journal of Imminent Science & Technology., vol. 2, no. 2, 2024.
15. [15] S. Poornima, R. Divya, R. S. Krishnan, S. Jegadeesan, G. Yamini, and G. V. Rajkumar, "Advancing Rose Disease Diagnosis: A Deep Learning Framework with EfficientNet-B7," in 2024 8th International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud)(I-SMAC), Nepal, 2024, IEEE, pp. 1747-1754.
16. [16] S. I. Nayan, S. M. Rahman, and N. I. Mahbub, "RoseVision: An Android Application To Detect Rose Leaf Diseases Using Modified Convolutional Neural Network," in 2023 26th International Conference on Computer and Information Technology (ICCIT), Bangladesh, 2023, IEEE, pp. 1-6.
17. [17] S. Nuanmeesri, "A hybrid deep learning and optimized machine learning approach for rose leaf disease classification," Engineering, Technology & Applied Science Research, vol. 11, no. 5, pp. 7678-7683, 2021.
18. [18] D. Banerjee, N. Sharma, R. Chauhan, M. Singh, and B. V. Kumar, "Improving Precision in Rose Leaf Disease Recognition with Integrated CNN and SVM Models," in 2024 4th International Conference on Innovative Practices in Technology and Management (ICIPTM), India, 2024, IEEE, pp. 1-6.
19. [19] S. Sabour, N. Frosst, and G. Hinton, "Matrix capsules with EM routing," in 6th international conference on learning representations, ICLR, Canada, 2018, pp. 1-15.
20. [20] G. Altan, "Performance evaluation of capsule networks for classification of plant leaf diseases," International Journal of Applied Mathematics Electronics and Computers, vol. 8, no. 3, pp. 57-63, 2020.
21. [21] L. Hou, D. Samaras, T. M. Kurc, Y. Gao, J. E. Davis, and J. H. Saltz, "Patch-based convolutional neural network for whole slide tissue image classification," in Proceedings of the IEEE conference on computer vision and pattern recognition, 2016, pp. 2424-2433.
22. [22] P. Poudel, A. Illanes, M. Sadeghi, and M. Friebe, "Patch based texture classification of thyroid ultrasound images using convolutional neural network," in 2019 41st Annual international conference of the IEEE engineering in medicine and biology society (EMBC), Germany, 2019, IEEE, pp. 5828-5831.
23. [23] G. Lachaud, P. Conde-Cespedes, and M. Trocan, "Patch selection for melanoma classification," in International Conference on Computational Collective Intelligence, 2022: Springer, pp. 148-159.
24. [24] Jarin Tasmim Jinia, Sakibur Rahman, and M. U. Mojumdar. A Comprehensive dataset on diseases affecting rose leaves: Identification, Symptoms, and Control Strategies. [Online]. Available: https://data.mendeley.com/datasets/hmjtdzkfh3/2
25. [25] J. Sklansky, "Image segmentation and feature extraction," IEEE Transactions on Systems, Man, and Cybernetics, vol. 8, no. 4, pp. 237-247, 1978.
26. [26] G. S. Raghtate and S. S. Salankar, "Comparison of classification methods with second order statistical analysis and wavelet transform for texture image classification," in 2015 International Conference on Computational Intelligence and Communication Networks (CICN), India, 2015, IEEE, pp. 312-317.
27. [27] R. C. Gonzalez and R. E. Woods, Digital image processing (third edition). Pearson, 2007, p. 976.
28. [28] R. M. Haralick, K. Shanmugam, and I. H. Dinstein, "Textural features for image classification," IEEE Transactions on systems, man, and cybernetics, no. 6, pp. 610-621, 1973.
29. [29] F. Albregtsen, B. Nielsen, and H. E. Danielsen, "Adaptive gray level run length features from class distance matrices," in Proceedings 15th International Conference on Pattern Recognition. ICPR-2000, 2000, vol. 3: IEEE, pp. 738-741.
30. [30] T. Ojala, M. Pietikäinen, and D. Harwood, "A comparative study of texture measures with classification based on featured distributions," Pattern recognition, vol. 29, no. 1, pp. 51-59, 1996.
31. [31] M. Heikkilä, M. Pietikäinen, and C. Schmid, "Description of interest regions with local binary patterns," Pattern recognition, vol. 42, no. 3, pp. 425-436, 2009.
32. [32] N. Dalal and B. Triggs, "Histograms of oriented gradients for human detection," in 2005 IEEE computer society conference on computer vision and pattern recognition (CVPR'05), 2005, vol. 1: IEEE, pp. 886-893.
33. [33] G. Huang, Z. Liu, L. Van Der Maaten, and K. Q. Weinberger, "Densely connected convolutional networks," in Proceedings of the IEEE conference on computer vision and pattern recognition, Hawaii, 2017, pp. 4700-4708.
34. [34] K. He, X. Zhang, S. Ren, and J. Sun, "Identity mappings in deep residual networks," in Computer Vision–ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11–14, 2016, Proceedings, Part IV 14, 2016: Springer, pp. 630-645.
35. [35] A. G. Howard et al., "Mobilenets: Efficient convolutional neural networks for mobile vision applications," arXiv preprint arXiv:1704.04861, 2017.
36. [36] M. Sandler, A. Howard, M. Zhu, A. Zhmoginov, and L.-C. Chen, "Mobilenetv2: Inverted residuals and linear bottlenecks," in Proceedings of the IEEE conference on computer vision and pattern recognition, USA, 2018, pp. 4510-4520.
37. [37] B. Zoph, V. Vasudevan, J. Shlens, and Q. V. Le, "Learning transferable architectures for scalable image recognition," in Proceedings of the IEEE conference on computer vision and pattern recognition, USA, 2018, pp. 8697-8710.