Tess4.0 windows编译与使用

tesseract-OCR4.00 dev VS2015 安装编译

编译教程参考：https://github.com/tesseract-ocr/tesseract/wiki/Compiling#windows

1. 下载最新的CPPAN，将cppan.exe所在的文件路径作为环境变量的值。
2. 下载最新的cmake，解压文件压缩包，将解压后目录下的bin文件夹的目录地址加载至系统环境变量PATH中。
3. 下载Tesseract最新版的源码：git clone https://github.com/tesseract-ocr/tesseract tesseract
4. 执行以下命令，编译：

• cd tesseract
• mkdir win64 && cd win64
• cppan ..
• cmake .. -G "Visual Studio 14 2015 Win64"
• cppan --generate .

5. 在win64目录下找到tesseract.sln文件，用VS2015打开该文件，点击【生成】按钮，根据相应错误进行改错。

Tess4.0 常用命令

tesseract::PageSegMode

• PSM_OSD_ONLY, ///< Orientation and script detection only. 定位和脚本检测
• PSM_AUTO_OSD, ///< Automatic page segmentation with orientation and < script detection. (OSD) 自动定位、检测、分割
• PSM_AUTO_ONLY, ///< Automatic page segmentation, but no OSD, or OCR. 自动(仅)分割
• PSM_AUTO, ///< Fully automatic page segmentation, but no OSD. 完全自动分割
• PSM_SINGLE_COLUMN, ///< Assume a single column of text of variable sizes. 仅有一列大小变化的文本
• PSM_SINGLE_BLOCK_VERT_TEXT, ///< Assume a single uniform block of vertically aligned text. 假设一个统一的垂直对齐的文本块
• PSM_SINGLE_BLOCK, ///< Assume a single uniform block of text. (Default.) 假设一个统一的文本块
• PSM_SINGLE_LINE, ///< Treat the image as a single text line. 将图像视为一条单独的文本行
• PSM_SINGLE_WORD, ///< Treat the image as a single word. 单独的词
• PSM_CIRCLE_WORD, ///< Treat the image as a single word in a circle. 将图像看作一个圆圈中的单个单词
• PSM_SINGLE_CHAR, ///< Treat the image as a single character. 一个字符
• PSM_SPARSE_TEXT, ///< Find as much text as possible in no particular order. 找到尽可能多的文本，没有特别的顺序
• PSM_SPARSE_TEXT_OSD, ///< Sparse text with orientation and script det. 带定向和脚本依据的稀疏文本
• PSM_RAW_LINE, ///< Treat the image as a single text line, bypassing hacks that are Tesseract-specific. 将图像看作一个单独的文本行，绕过Tesseract-specific的检测
• PSM_COUNT ///< Number of enum entries. 枚举的条目数

page segmentation modes (psm命令)

• 0 Orientation and script detection (OSD) only 定位和脚本检测
• 1 Automatic page segmentation with OSD
• 2 Automatic page segmentation, but no OSD, or OCR
• 3 Fully automatic page segmentation, but no OSD. (Default)
• 4 Assume a single column of text of variable sizes
• 5 Assume a single uniform block of vertically aligned text
• 6 Assume a single uniform block of text
• 7 Treat the image as a single text line
• 8 Treat the image as a single word
• 9 Treat the image as a single word in a circle
• 10 Treat the image as a single character
• 11 Sparse text. Find as much text as possible in no particular order
• 12 Sparse text with OSD
• 13 Raw line. Treat the image as a single text line, bypassing hacks that are Tesseract-specific.

OcrEngineMode (OEM)

• OEM_TESSERACT_ONLY, // Run Tesseract only - fastest
• OEM_LSTM_ONLY, // Run just the LSTM line recognizer.
• OEM_TESSERACT_LSTM_COMBINED, // Run the LSTM recognizer, but allow fallback to Tesseract when things get difficult.
• OEM_DEFAULT, // Specify this mode when calling init_*(), to indicate that any of the above modes should be automatically inferred from the variables in the language-specific config, command-line configs, or if not specified in any of the above should be set to the default OEM_TESSERACT_ONLY.
• OEM_CUBE_ONLY, // Run Cube only - better accuracy, but slower
• OEM_TESSERACT_CUBE_COMBINED, // Run both and combine results - best accuracy