TY - JOUR ID - 2431 TI - AgriNet: a New Classifying Convolutional Neural Network for Detecting Agricultural Products’ Diseases JO - Journal of AI and Data Mining JA - JADM LA - en SN - 2322-5211 AU - Salimian Najafabadi, F. AU - Sadeghi, M. T. AD - Azadi Campus, Yazd University, Yazd, Iran. AD - Department of Electrical Engineering, Yazd University, Yazd, Iran. Y1 - 2022 PY - 2022 VL - 10 IS - 2 SP - 285 EP - 302 KW - Plant Diseases Recognition KW - Convolutional neural network KW - ShuffleNet KW - SENet DO - 10.22044/jadm.2022.11360.2295 N2 - An important sector that has a significant impact on the economies of countries is the agricultural sector. Researchers are trying to improve this sector by using the latest technologies. One of the problems facing farmers in the agricultural activities is plant diseases. If a plant problem is diagnosed soon, the farmer can treat the disease more effectively. This study introduces a new deep artificial neural network called AgriNet which is suitable for recognizing some types of agricultural diseases in a plant using images from the plant leaves. The proposed network makes use of the channel shuffling technique of ShuffleNet and the channel dependencies modeling technique of SENet. One of the factors influencing the effectiveness of the proposed network architecture is how to increase the flow of information in the channels after explicitly modelling interdependencies between channels. This is in fact, an important novelty of this research work. The dataset used in this study is PlantVillage, which contains 14 types of plants in 24 groups of healthy and diseased. Our experimental results show that the proposed method outperforms the other methods in this area. AgriNet leads to accuracy and loss of 98% and 7%, respectively on the experimental data. This method increases the recognition accuracy by about 2% and reduces the loss by 8% compared to the ShuffleNetV2 method. UR - https://jad.shahroodut.ac.ir/article_2431.html L1 - https://jad.shahroodut.ac.ir/article_2431_10ea85357f4b93525c9f43aa020d424a.pdf ER -