Research on Image Recognition of Power Equipment Using Random Forest Algorithms
- Resource Type
- Conference
- Authors
- Zhang, Yuan; Hu, Jiapeng; Han, Ruiqi; Wang, Zhanyu; Tian, Kanghui
- Source
- 2022 IEEE 2nd International Conference on Electronic Technology, Communication and Information (ICETCI) Electronic Technology, Communication and Information (ICETCI), 2022 IEEE 2nd International Conference on. :348-352 May, 2022
- Subject
- Communication, Networking and Broadcast Technologies
Fields, Waves and Electromagnetics
Power, Energy and Industry Applications
Robotics and Control Systems
Image recognition
Computational modeling
Neural networks
Computer architecture
Transforms
Hybrid power systems
Convolutional neural networks
Random forest algorithm
Deep learning
Recognition of power equipment images
- Language
The primary function of image recognition is to convert an image style from one region to another. This task identifies new requirements for the traditional convolutional neural network architecture. Therefore, the random forest algorithm model is typically employed in image recognition tasks' current processing power equipment. Power equipment image recognition aims to convert the image of the input power equipment into a brand new image that can be used for computer program input or system monitoring. This paper proposes an image recognition model of power equipment based on a random forest algorithm in light of the above background. This technique improves the quality of images through the above processing. The random forest algorithm model relies primarily on adversarial network generation and autoencoder. This paper presents unsupervised and supervised image recognition methods by analyzing different basic image recognition tasks. On this basis, we propose the unsupervised image recognition task of adversarial neural network based on maximum entropy method and the supervised image recognition task of adversarial network based on random forest generation. Therefore, this paper introduces unsupervised and supervised standard datasets for image recognition to improve the quality of generated images.