Abstract:Scene text detection is a fundamental research work in the field of image processing and has extensive application value. Segmentation-based methods have time-consuming feature processing, while post-processing algorithms are excellent. Real-time semantic segmentation methods use lightweight backbone networks for feature extraction and aggregation but lack effective post-processing methods. The pure convolutional network improves model performance by changing key components. Combining the advantages of three types of methods, we propose a Pure Convolutional Bilateral Segmentation Network (PCBSNet) for real-time natural scene text detection. First, we constructed a bilateral feature extraction backbone network to significantly improve detection speed. The low extraction detail branch captures spatial information, while the efficient semantic extraction branch accurately captures semantic features through a series of micro designs. Second, we built an efficient attention aggregation module to guide the efficient and adaptive aggregation of features from the two branches. The fused feature map undergoes feature enhancement to obtain more accurate and reliable feature representation. Finally, we used differentiable binarization post-processing to construct text instance boundaries. To evaluate the effectiveness of the proposed model, we compared it with mainstream lightweight models on three datasets: ICDAR2015, MSRA-TD500, and CTW1500. The F-measure scores were 82.9%, 82.8%, and 78.9%, respectively, and the FPS were 59.1, 94.3, and 75.5 frames per second. We also conducted extensive ablation experiments on the ICDAR2015 dataset to validate the rationality of the proposed improvements. The obtained results indicate that the proposed model significantly improves inference speed while enhancing accuracy and demonstrates good competitiveness compared to other advanced detection methods. However, when faced with curved text, the detection performance of PCBSNet needs to be improved.

Specific category region proposal network for text detection in natural scene

Natural Scene Text Detection with Multi-Layer Segmentation and Higher Order Conditional Random Field Based Analysis.

Natural Scene Text Detection Based on Multiscale Connectionist Text Proposal Network

CPN: Complementary Proposal Network for Unconstrained Text Detection

FTPN: Scene Text Detection with Feature Pyramid Based Text Proposal Network.

Character Proposal Network for Robust Text Extraction.

Text-Attentional Convolutional Neural Network for Scene Text Detection

Character Region Awareness Network for Scene Text Recognition

Text Proposals Based on Windowed Maximally Stable Extremal Region for Scene Text Detection.

Text-Attentional Convolutional Neural Networks for Scene Text Detection

Scene Text Detection and Segmentation Based on Cascaded Convolution Neural Networks.

Detecting Multi-Oriented Text with Corner-Based Region Proposals

Deep Strip-Based Network with Cascade Learning for Scene Text Localization

Natural scene text detection based on attention mechanism and deep multi-scale feature fusion

A Multi-Level Feature Fusion Network for Scene Text Detection with Text Attention Mechanism

DeepText: A New Approach for Text Proposal Generation and Text Detection in Natural Images.

Conceptual Text Region Network: Cognition-Inspired Accurate Scene Text Detection

CSFF-Net: Scene Text Detection Based on Cross-Scale Feature Fusion

Text Detection And Recognition Algorithm For Arbitrary Shapes In Natural Scenes

PCBSNet: A Pure Convolutional Bilateral Segmentation Network for Real-Time Natural Scene Text Detection

Deep Scene Text Detection with Connected Component Proposals.