Handwriting Detection and Recognition Improvements Based on Hidden Markov Model and Deep Learning

The online handwriting detection and recognition has become an important research in. area. An individual's writing can be easily forged and disguised in various ways including freehand simulation, tracing and image transfer, making genuine handwriting recognition a challenging task. With the advent of various online handwriting recognition systems developed, but for English characters recognition these still lack the simplicity and accuracy. While identification approaches were successfully reported, good forgeries are able to outsmart the existing tools. Existing flaws in recognition systems led to more research works in automatic detection and recognition works via computer techniques, feature extraction, classification accuracy comparison, performance evaluation and pattern recognition. To realize simpler and efficient English character recognition, we develop a handwriting detection and recognition system based on the Kohonen Network and deep learning. The system consists of interfaces for the online handwritten character was featured in matrix form of sizes 5x7 pixel and 35x33 pixels represented with binary values. Identifying all occupied character strokes in the series of binary string recognizes the full character. The recognition performance was compared between 35 pixels and 1155 pixels environment, evaluated in terms of accuracy, and consistency. An experiment was conducted with 25 online handwritten input data of straight stroke ('V', 'X', 'Y') and curve stroke ('C', 'O', 'S') characters collected from 25 participants. Findings show an overall improvement of 31% recognition accuracy of using 35x33 pixels against the 5x7 pixels. Handwriting characters featured in 35x33 pixels outperformed the 5x7 pixels accuracy by 37.49% on straight stroke characters and 24.52% on curve stroke.