针对阿拉伯手写文档，提出一套高效识别系统，准确率超93%。
Arabic Handwritten Document OCR Solution with Binarization and Adaptive Scale Fusion Detection
- 先用可微二值化和自适应多尺度融合分割文本行
- 单字识别率达99.20%，句子识别率达83.76%
- 适合阿拉伯手写文字识别任务，尤其适用于数据少场景
将图像中的文字转化为纯文本是学术界和工业界广泛研究的问题。由于书写风格多样且标注数据有限，阿拉伯手写文字识别（AHTR）面临额外挑战。本文提出一套完整的OCR流程：首先利用可微二值化与自适应多尺度融合技术进行文本行分割，确保检测精度；随后采用CNN-BiLSTM-CTC架构进行字符识别。系统在阿拉伯多字体数据集（AMFDS）上训练，对含7至10个字符的单字样本实现99.20%的字符识别率（CRR）和93.75%的单词识别率（WRR），句子识别率达83.76%。结果表明该系统在处理阿拉伯文字方面表现优异，为AHTR系统树立了新基准。
原文摘要 · Abstract (English)
The problem of converting images of text into plain text is a widely researched topic in both academia and industry. Arabic handwritten Text Recognation (AHTR) poses additional challenges due to diverse handwriting styles and limited labeled data. In this paper we present a complete OCR pipeline that starts with line segmentation using Differentiable Binarization and Adaptive Scale Fusion techniques to ensure accurate detection of text lines. Following segmentation, a CNN-BiLSTM-CTC architecture is applied to recognize characters. Our system, trained on the Arabic Multi-Fonts Dataset (AMFDS), achieves a Character Recognition Rate (CRR) of 99.20% and a Word Recognition Rate (WRR) of 93.75% on single-word samples containing 7 to 10 characters, along with a CRR of 83.76% for sentences. These results demonstrate the system's strong performance in handling Arabic scripts, establishing a new benchmark for AHTR systems.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。