百科.dev
全部条目在线工具AI 编程趋势榜开源项目技术资讯提交条目
登录
< 返回工具列表
T

text-detection

> 编程语言
开源

主要使用 MSER 和 SWT 进行文本检测

199 stars0 点赞0 次浏览
访问官网GitHub

工具介绍

主要使用 MSER 和 SWT 进行文本检测

text-detection

This project aims to detect text regions in images using only image processing techniques with MSER (Maximally Stable Extremal Regions) and SWT (Stroke Width Transform). And also Tesseract-OCR tool is used optionally, as assistance to the algorithm.

Please cite the paper:

Özgen, A.C., Fasounaki, M. and Ekenel, H.K., 2018, May. Text detection in natural and computer-generated images. In 2018 26th Signal Processing and Communications Applications Conference (SIU) (pp. 1-4). IEEE.

INSTALLING

Use Python>=3.11

Install requirements with pip

pip install -r requirements.txt

For OCR assistance, install Tesseract from package manager

sudo apt install tesseract-ocr

USAGE

Basic usage is

python detect.py -i

You can give output file

python detect.py -i assets/scenetext01.jpg -o

More options available

python detect.py -i assets/scenetext01.jpg -o -d -t

Option -i is image path, -o is output path, -d is SWT direction (default is both+), -t option chooses if Tesseract will be used. Normally Tesseract runs poorly if whole image given as input. But it is used as final decision of bounding boxes.

If you want to give whole image to Tesseract to see the impact of the algorithm, try this.

python detect.py -i assets/scenetext01.jpg -f

For more detail (seeing intermediate steps), the usage given below is also available.

python detect.py -i assets/scenetext01.jpg -d both+ -t --details

Sample Results

REFERENCES

B. Epshtein, E. Ofek, and Y. Wexler. Detecting text in natural scenes with stroke width transform. In 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pages 2963–2970, June 2010.

Á. González, L. M. Bergasa, J. J. Yebes, and S. Bronte. Text location in complex images. In Proceedings of the 21st International Conference on Pattern Recognition (ICPR2012), pages 617–620, Nov 2012.

Y. Li and H. Lu. Scene text detection via stroke width. In Proceedings of the 21st International Conference on Pattern Recognition (ICPR2012), pages 681–684, Nov 2012.

Issues· 0 开放

查看全部 Issues在 GitHub 打开

暂无开放 Issues,或尚未同步最近议题。

> 标签

Pythonmserocrswttesseract-ocr

暂无评论,来聊聊你的看法吧

> 工具信息

发布日期2026年8月1日
最后更新2026年9月18日
分类编程语言
定价开源

> 相关工具

T
TypeScript
JavaScript 的超集,为前端与全栈提供静态类型
P
Python
通用编程语言,广泛用于 Web、数据与 AI
G
Go
Google 推出的简洁高效系统语言