Intelligent recognition method of indication of substation pointer instrument based on deformable convolution neural network

Intelligent identification method of indicator number of substation instrument based on deformable convolutional neural network there are many equipment in substation, and the value and scale of oil thermometer, oil level gauge, pressure gauge and other instrument equipment reflect the operation state of most instruments and meters, which is particularly important. Therefore, the research on value reading of instrument equipment in substation is particularly key. At present, for the instrument recognition of substation, most studies use traditional image processing and machine learning methods. However, in the recognition process, due to the influence of uneven illumination, complex background, rotation angle, image blur, shooting angle, proportion change and other factors, the recognition accuracy of pointer instrument is low and its usability is poor. In order to solve the above problems, this paper combines the traditional image processing technology with the deep learning method, and proposes an automatic recognition method of substation pointer instrument based on deformable convolutional neural network. The idea of deformable volume is introduced to enhance the modeling ability of convolutional neural network, so as to improve the accuracy of instrument recognition. The main idea is that firstly, the deformable convolution neural network method is used to detect the instrument image in the image, then the residual neural network is used to extract the key points of the instrument dial and pointer, then the detected key points are used to fit the dial circle and pointer, and finally the readout value is calculated according to the deflection angle of the pointer relative to the scale. The experimental results show that this method is very effective for the identification of pointer instruments, and has high accuracy and practicability, which is conducive to promoting the realization of intelligent operation and maintenance of substation.