Table Structure Recognition Method Based on Lightweight Network and Channel Attention

Abstract

The table recognition model rows and columns aggregated network (RCANet) uses a semantic segmentation approach to recognize table structure, and achieves better performance in table row and column segmentation. However, this model uses ResNet18 as the backbone network, and the model has 11.35 million parameters and a volume of 45.5 M, which is inconvenient to deploy to lightweight servers or mobile terminals. Therefore, from the perspective of model compression, this paper proposes the lightweight rows and columns attention aggregated network (LRCAANet), which uses the lightweight network ShuffleNetv2 to replace the original RCANet backbone network ResNet18 to simplify the model size. Considering that the lightweight network reduces the number of feature channels, it has a certain impact on the performance of the model. In order to strengthen the learning between feature channels, the rows attention aggregated (RAA) module and the columns attention aggregated (CAA) module are proposed. The RAA module and the CAA module add the squeeze and excitation (SE) module to the original row and column aggregated modules, respectively. Adding the SE module means the model can learn the correlation between channels and improve the prediction effect of the lightweight model. The experimental results show that our method greatly reduces the model parameters and model volume while ensuring low-performance loss. In the end, the average F1 score of our model is only 1.77% lower than the original model, the parameters are only 0.17 million, and the volume is only 0.8 M. Compared with the original model, the parameter amount and volume are reduced by more than 95%.

A Lightweight Detection Method for Blueberry Fruit Maturity Based on an Improved YOLOv5 Algorithm

“…Each branch has an equal number of feature channels. The left branch performs identity mapping [27]. The right branch undergoes two 1 × 1 ordinary convolutions and one 3 × 3 depthwise convolution (DWConv) while maintaining an equal number of input and output channels.…”

Eye-Gaze Controlled Wheelchair Based on Deep Learning

“…The overall process of the CBAM module can be divided into two parts, as shown in Figure 10. Channel attention is mainly used to capture the correlation between different channels [49]. This module first obtained two channels through global average pooling and maximum pooling operations: the 1 × 1 × C channels.…”

“…In Equation ( 3), 𝜎 denotes the Sigmoid function, MLP denotes the multilayer perceptron, AvgPool denotes the average pooling layer, MaxPool denotes the maximum pooling layer, 𝑊 0 and 𝑊 1 denote the two weight matrices, 𝐹 𝑎𝑣𝑔 and 𝐹 𝑚𝑎𝑥 denote the results of the input data after the AvgPool and MaxPool operations, and the superscript 𝐶 Channel attention is mainly used to capture the correlation between different channels [49]. This module first obtained two channels through global average pooling and maximum pooling operations: the 1 × 1 × C channels.…”