Image Recognition Algorithm of Electrical Engineering Equipment Based on Machine Learning Method

Abstract

In order to quickly analyze and identify the massive images collected by the intelligent patrol system in substations, this paper proposes an image classification model based on the combination of deep learning and support vector machine (SVM). First of all, the methods of rotation and folding are used to expand the original data collected. Then, the extended image set is merged and randomly divided into training set and test set under the same type of conditions. Based on the actual image, the convolution neural network (CNN) is improved to extract the image features of the training set. Finally, the SVM classifier is trained by using the depth features of the training set images, and the classification test is realized on the test set images. We use 8000 pictures collected by the patrol robot to verify the accuracy of the model, and the results show that the model has strong classification performance.