OCR
光学字符识别（Optical Character Recognition，OCR）是一种将图像中的印刷体或手写文字转换为可编辑、可搜索的数字文本的技术。其核心原理是通过图像处理与模式识别算法，识别扫描文件、照片、PDF等图像中的字符信息。OCR广泛应用于文档数字化、票据识别、证件信息提取等场景，是计算机视觉与自然语言处理的重要交叉技术。
时间轴 (近 90 天)
AI prescription parsing typically combines OCR with LLM-based information extraction to convert prescription details into structured medication reminder plans
Prescription parsing typically follows a two-stage process: an OCR engine converts the prescription image into raw text, then a large language model performs semantic extraction
ALPR technology combines Optical Character Recognition (OCR) with infrared imaging to read license plate information under various lighting and weather conditions
CLIP等技术建立的文本-图像语义对齐使模型能将抽象风格描述映射到具体视觉特征空间
该项目使用OCR和图像识别替代传统的内存读取方式进行游戏状态感知，优势在于不需要逆向分析游戏内存结构，也不容易触发反作弊系统的检测
全部知识事实 (5)
ALPR technology combines Optical Character Recognition (OCR) with infrared imaging to read license plate information under various lighting and weather conditions
90%待验证该项目使用OCR和图像识别替代传统的内存读取方式进行游戏状态感知，优势在于不需要逆向分析游戏内存结构，也不容易触发反作弊系统的检测
80%待验证AI prescription parsing typically combines OCR with LLM-based information extraction to convert prescription details into structured medication reminder plans
80%待验证Prescription parsing typically follows a two-stage process: an OCR engine converts the prescription image into raw text, then a large language model performs semantic extraction
80%待验证CLIP等技术建立的文本-图像语义对齐使模型能将抽象风格描述映射到具体视觉特征空间
50%