Design of text position detection method for electrical equipment nameplate

Abstract

With the progress of science and technology and the deepening of the reform of electric power systems, robots gradually replace people in the power inspection task of a substation, which is also the inevitable trend of the development of the smart grid. Due to a large amount of equipment information on the device nameplate, the identification of the electrical name on the device nameplate has become the primary task of the inspection task. The traditional algorithm combining SWT and MSER and CTPN text detection algorithm based on deep learning are often used to realize the text position detection of the nameplate. In this paper, an improved CTPN algorithm electrical nameplate recognition method is proposed to solve the problem that the CTPN algorithm is not effective in the recognition of slanted text. The CTPN algorithm can improve the detection effect of slanted text through preprocessing and text fusion.