Optical Character Recognition of Electrical Equipment Nameplate with Contrast Enhancement

Automatic identification of the electrical equipment via optical character recognition (OCR) of the electrical equipment nameplate images has been widely used as a primary inspection task of a substation. However, most existing OCR methods assume that the input image is obtained under ideal ambient lighting conditions and only focus on text detection and recognition. This commonly results in poor OCR performance when dealing with images with, e.g., reflection effects that are captured under complex ambient light conditions. In this paper, we propose an OCR system that can reliably handle the common reflection effects shown in the electrical equipment nameplate images captured under challenging ambient light conditions. Specifically, we did this by applying a robust contrast enhancement method that is based on a logarithmic mapping function (LMF) to the electrical equipment nameplate images prior to OCR with one of the state-of-the-art OCR networks, PPOCRv3. Experimental results with 24 test electrical equipment nameplate images show that the LMF contrast enhancement not only improves the image quality in terms of the lighting condition and overall scene statistics but also improves the overall OCR sentence recognition accuracy.