Design of a Portable Handwriting Recognition System Based on FPGA

In this paper, we propose a method for extracting the characteristic value of an image after the structural analysis of the word group. The Euclidean distance difference and the angle difference of each handwriting angle were calculated by using the median filtering and Sobel edge detection algorithm. Experimental results indicate that the Euclidean distance difference of the handwriting of the same writer is generally below 0.2, and the angle difference of the handwriting angle is below 0.05. The proposed system is simple in operation, short in time consumption, and high in accuracy. It provides a powerful auxiliary tool for handwriting recognition. Furthermore, the algorithm design provides a new idea for intelligent offline handwriting recognition.