PF-ViT: Parallel and Fast Vision Transformer for Offline Handwritten Chinese Character Recognition

Abstract

Recently, Vision Transformer (ViT) has been widely used in the field of image recognition. Unfortunately, the ViT model repeatedly stacks 12-layer encoders, resulting in a large number of model computations, many parameters, and slow training speed, making it difficult to deploy on mobile devices. In order to reduce the computational complexity of the model and improve the training speed, a parallel and fast Vision Transformer method for offline handwritten Chinese character recognition is proposed. The method adds parallel branches of the encoder module to the structure of the Vision Transformer model. Parallel modes include two-way parallel, four-way parallel, and seven-way parallel. The original picture is fed to the encoder module after flattening and linear embedding processing operations. The core step in the encoder is the multihead attention mechanism. Multihead self-attention can learn the interdependence between image sequence blocks. In addition, the use of data expansion strategies increases the diversity of data. In the two-way parallel experiment, when the model is 98.1% accurate on the dataset, the number of parameters and the number of FLOPs are 43.11 million and 4.32 G, respectively. Compared with the ViT model, whose parameters and FLOPs are 86 million and 16.8 G, respectively, the two-way parallel model has a 50.1% decrease in parameters and a 34.6% decrease in FLOPs. This method has been demonstrated to effectively reduce the computational complexity of the model while indirectly improving image recognition speed.

Novel Heuristic Recurrent Neural Network Framework to Handle Automatic Telugu Text Categorization from Handwritten Text Image

“…The morphologically rich Dravidian classical language Telugu was the focus of a text categorization presentation by K. Rajan et al [8], which utilised a Vector Space Model and an Artificial Neural Network. From our experiments, we know that an A.N.N.…”

“…Besides, past examinations perform the research connected with picture text acknowledgment [4][5][6][7]. Likewise, optical person acknowledgement having the studies, which is going to be considered the agent method of today for recognizing the texts using picture information [8][9][10][11]. Text and penmanship acknowledgment began with the number acknowledgment in Latin text in the last part of the 1950s and is extending to incorporate different dialects, like Telugu, Japanese, Telugu, and Persian.…”

LW-ViT: The Lightweight Vision Transformer Model Applied in Offline Handwritten Chinese Character Recognition

“…The difficulty factor of offline HCCR is higher because of the difficulty of feature extraction. Due to the challenging research field, offline handwritten Chinese character recognition has attracted the attention of a large number of researchers and scholars [7][8][9][10][11][12][13].…”

Handwritten character recognition using optimization based skewed line segmentation method and multi-class support vector machine

“…*Author for correspondence However, the computer visualization, pattern identification, and document analysis research fields continue to face a significant problem in accurately detecting and recognizing text in photographs of natural scenes [9,10]. Even the traditional optical character recognition (OCR) techniques were not effective enough for recognizing text from scanned documents, even when the text is typically wellarranged and taken in a well-controlled setting [11,12]. Segmentation performs an important role in textual image recognition as well as in delivering efficient information about the visual framework of the system [13].…”