Article Details

Abstract

Aiming at the problems of low efficiency and complex operation of traditional disease identification methods, the pest and disease identification algorithm of Yingzui peach based on Deep Learning is proposed. Based on the ResNet50 network architecture, the algorithm uses the method of combining image recognition technology and Deep Learning to establish a high efficiency pest and disease monitoring system, which can quickly and accurately identify various types of pests and diseases. Firstly, a specialized dataset is established for 14 common pests and diseases of Yingzui peach in Lianping area of Heyuan. Secondly, the Deep Residual Network Model is used for training. Finally, the Graphical User Interface is developed using PyQt5 to realize the automatic diagnosis of pest and disease. This automatic diagnosis system is not only efficient and labor-saving but also environmentally friendly. It conforms to the development trend of smart agriculture and provides users with accurate control methods of pest and disease.

Keywords

image recognition technology; pest and disease monitoring; Deep Learning; pest and disease dataset; smart agriculture; Yingzui peach pest and disease

References

1. [1] Cai Guoqing, Wu Jianjun, Zhu Yuhua, etc Research progress on crop pest and disease detection based on convolutional neural networks [J]. Shandong Agricultural Science, 2024, 56 (11): 170-180
2. [2] Chen Hao, Tang Liuqing Image classification method for apple leaf diseases and pests based on transfer learning and VGG19 convolutional neural network [J]. Southern Agricultural Machinery, 2024, 55 (20): 6-9
3. [3] Liu Peng, Zhang Tianyi, Ran Xin, etc Rice pest and disease detection based on PBM-YOLOv8 [J]. Journal of Agricultural Engineering, 2024, 40 (20): 147-156
4. [4] Li Bing, Li Jiyun, Jia Meng Research on the accuracy of wheat pest and disease image recognition based on deep learning feature fusion technology [J]. Information Technology and Informatization, 2024 (8): 83-87
5. [5] Guan Bolun, Dong Wei, Zhang Liping, etc Research and development of a rice pest and disease monitoring platform based on computer vision [J]. Anhui Agricultural Science, 2024, 52 (16): 211-216
6. [6] Zheng Guo Research on Monitoring and Early Warning Technology of Rice Diseases and Pests Based on Machine Vision [D]. Chongqing: Chongqing Three Gorges University, 2023
7. [7] Yue Jian Video Coding Loop Filtering Method Based on Deep Learning [D]. Chengdu: University of Electronic Science and Technology of China, 2022
8. [8] Sun Rujie Short wave channel quality classification based on convolutional neural networks [J]. Modern Information Technology, 2021, 5 (6): 70-72+76
9. [9] Liu Yang Research and Implementation of a High Precision Aerial Image Object Detection Algorithm [J]. China Management Informatization, 2019, 22 (19): 149-151
10. [10] Qian Ruxue Flower image recognition and classification based on deep neural networks [J]. Information Technology and Informatization, 2022 (3): 210-213
11. [11] Cultivation Research and Implementation of Distributed Web Service Search Engine Based on Map/Reduce [D]. Xiangtan: Hunan University of Science and Technology, 2012
12. [12] Cao Huan, Fang Rui Classification and recognition of mango pests and diseases based on deep learning [J]. Computer Technology and Development, 2023, 33 (10): 115-119

Funding

Key Scientific Research Project of Guangdong Province (2024ZDZX4089); Guangdong Province College Youth Innovation Talent Project (2024KQNCX207); Guangdong Province University Youth Innovation Talent Project (2023KQNCX233)

About the authors

Gan Yuwan (1996-), female, Han ethnicity, from Mudanjiang, Heilongjiang Province, teaching assistant, master's student, research direction: image processing, machine vision.