A Machine Vision-Based Character Recognition System for Suspension Insulator Iron Caps
Changhua Zhang, Bin Liu, Zhongshu Chen, Jiafu Yan, Fenglian Liu, Wang Wang Yu, Qiming Zhang · IEEE Transactions on Instrumentation and Measurement · 2023
Character recognition on the surfaces of iron caps is pivotal in the automated production of suspension insulators. This paper proposes a machine vision-based character recognition system for the iron caps of suspension insulators in order to solve the problem of recognizing curved characters with three-dimensional (3D) structures and severe surface reflections. In this system, firstly, 3D character extraction system (3D-CES) converts 3D characters on curved surfaces to flat two-dimensional (2D) characters. Then, the 2D characters are segmented into individual characters using an improved Drop-Fall (IDF) segmentation algorithm. Finally, the individual characters are recognized using a pruned Visual Geometry Group 16 (VGG-16) network. This paper tested the proposed system on real iron cap data, and a recognition accuracy of 98.3% were achieved. The performance of the proposed system is higher than existing deep-learning-based text character recognition methods by 2.9%.