Enhancing Kiwi Leaf Disease Detection Through Advanced Convolutional Neural Networks and Transfer Learning Techniques
Article Sidebar
Main Article Content
Abstract
To enhance the effectiveness of the kiwi leaf disease recognition model, a novel approach utilizing a convolutional neural network (CNN) has been developed. Initially, a comprehensive dataset comprising 7,310 images of kiwi leaves afflicted with four types of diseases—brown spot, ulcer, mosaic, and anthracnose—was compiled. Following the dataset creation, pre-processing steps were applied to optimize the data for model training. Subsequently, a nine-layer CNN model, inspired by AlexNet, was constructed. Additionally, transfer learning techniques were employed, leveraging pretrained models such as VGG16, VGG19, InceptionV3, and InceptionV4, which were initially trained on the ImageNet dataset, for the task of kiwi leaf disease classification and recognition. Experimental outcomes indicate that models trained with transfer learning exhibit superior classification performance and faster convergence rates. Data augmentation was also found to enhance data diversity and mitigate overfitting. Notably, utilizing the VGG19 pre-trained model, the combined application of transfer learning and data augmentation achieved a classification accuracy of 96.73%.
Article Details
This work is licensed under a Creative Commons Attribution 4.0 International License.
Mind forge Academia also operates under the Creative Commons Licence CC-BY 4.0. This allows for copy and redistribute the material in any medium or format for any purpose, even commercially. The premise is that you must provide appropriate citation information.
References
Special Investigation Report on Chinese Kiwi Planting and Deep Processing Market 2019
Lai Xinpu, Mao Jiamei, Guo Jun, et al. Comparative Analysis of Maximum Residue Limit Standards for Kiwi Pesticides at Home and Abroad. South China Fruit Tree, vol. 42 (2013), 79-84.
R. Pydipati, T.F. Burks, et al. Identification of citrus disease using color texture features and discriminant analysis. Computers and Electronics in Agriculture, vol. 52 (2006), 49-53.
Fernando López-García, Gabriela Andreu-García, et al. Automatic detection of skin defects in citrus fruits using a multivariate image analysis approach. Computers and Electronics in Agriculture, vol. 71 (2010), 38-44.
Bao Wenxia, Zhao Jian, et al. Wheat leaf disease recognition based on elliptic metric learning. Journal of Agricultural Machinery, vol. 49 (2018), 20-26.
Chai Ali, Liao Ningfang, Tian Lixun, et al. Cucumber disease recognition based on hyperspectral imaging and discriminant analysis. Spectroscopy and Spectral Analysis, vol. 30 (2010), 1357-1361.
Liu Lianzhong, Zhang Wu, et al. Research on wheat disease image recognition technology based on improved color characteristics. Anhui Agricultural Science, vol. 40 (2012), 12877-12879.
Lu Yang, Guo Dan, Shen He, et al. Research on Rice Leaf Disease Identification Method Based on Deep Learning. Information Recording Materials, vol. 19 (2018), 13-185.
Wang Yanxiang, Zhang Yan, et al. Advances in crop disease image recognition technology based on deep learning. Journal of Zhejiang Agricultural Sciences, vol. 31 (2019), 162-169.
An Qiangqiang, Zhang Feng, et al. Image recognition of plant diseases and insect pests based on deep learning. Agricultural Engineering, vol. 07 (2018), 48-50.
Ye Haijian, Lang Rui, et al. Identification method of cucumber downy mildew based on visual saliency map. Transactions of the Chinese Society of Agricultural Machinery, vol. 47 (2016), 270-274.
Liao Jingwei, Cai Ying, et al. Plant Disease Recognition Technology Based on Convolutional Neural Network. Modern Computer (Professional Edition). 2018 (19).
Simonyan K, Zisserman A. Very Deep Convolutional Networks for Large-Scale Image Recognition. Computer Science, vol. 37 (2014), 1-14.
Hermans J. Auto-line filter: self-cleaning, continuous filtration system. Filtration & Separation, vol. 39 (2002), 28-30.
Amburgey J E. Optimization of the extended terminal sub fluidization wash (ETSW) filter backwashing procedure. Water Research, vol. 39 (2005), 314-330.
Guo Xiaoqing, Fan Taojie, Shu Xin. Image recognition of tomato leaf diseases based on improved Multi-Scale AlexNet. Journal of Agricultural Engineering, vol. 13 (2019), 162-167.
Wei Liran, Yue Jun, et al. Multi-class detection method for plant leaf diseases based on kernel function support vector machine. Transactions of the Chinese Society of Agricultural Machinery, vol. 48 (2017), 166-171.
Qin Feng, Liu Dongxia, et al. Image recognition of 4 alfalfa leaf diseases based on deep learning and support vector machine. Journal of China Agricultural University, vol. 22 (2017), 123-126.
Xie Zeqi, Hu Xiaowei, et al. Crop leaf disease level recognition algorithm based on environmental information and color characteristics. Jiangsu Agricultural Sciences, vol. 44 (2016), 428-430.
Yu Xiuli, Xu Chao, Wang Dandan, et al. Study on wheat leaf disease recognition method based on SVM. Research of Agricultural Mechanization, vol. 52 (2014), 151-159.
Krizhevsky A, Sutskever I, Hinton G E. ImageNet Classification with Deep Convolutional Neural Networks. Advances in Neural Information Processing Systems, vol. 25 (2012), 158-163.