S-Swin Transformer: simplified Swin Transformer model for offline handwritten Chinese character recognition

Abstract

The Transformer shows good prospects in computer vision. However, the Swin Transformer model has the disadvantage of a large number of parameters and high computational effort. To effectively solve these problems of the model, a simplified Swin Transformer (S-Swin Transformer) model was proposed in this article for handwritten Chinese character recognition. The model simplifies the initial four hierarchical stages into three hierarchical stages. In addition, the new model increases the size of the window in the window attention; the number of patches in the window is larger; and the perceptual field of the window is increased. As the network model deepens, the size of patches becomes larger, and the perceived range of each patch increases. Meanwhile, the purpose of shifting the window’s attention is to enhance the information interaction between the window and the window. Experimental results show that the verification accuracy improves slightly as the window becomes larger. The best validation accuracy of the simplified Swin Transformer model on the dataset reached 95.70%. The number of parameters is only 8.69 million, and FLOPs are 2.90G, which greatly reduces the number of parameters and computation of the model and proves the correctness and validity of the proposed model.

A fault diagnosis method for motor vibration signals incorporating Swin transformer with locally sensitive hash attention

“…Next, the results obtained after the first stage processing are input to the second stage, and feature extraction and exchange are continued, with the number of each token reduced to H/8 × W/8 and the output dimension expanded to 2 C [36]. Similarly, the above operation is repeated next, with the number of tokens in the third and fourth stages being H/16 × W/16 and H/32 × W/32, respectively.…”

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

“…Before, the man-made reasoning field has achieved a quantum jump. Presently, studies using man-made brainpower are directed and different fields are applied [1][2][3]. Particularly, attributable to the improvement is done in figuring power, in the field of PC vision the areas of use continually become more extensive.…”

“…The information is as bitmap picture. Disconnected written by hand acknowledgment [1] is similarly troublesome, as various individuals have different penmanship styles, and at times, the composing style of a similar individual fluctuates at various examples. On the web penmanship acknowledgment framework is the one in which they perceive the text of transcribed, while composing through the touch cushion utilizing pointer pen.…”