https://github.com/z3t0/ugv-ocr
https://github.com/z3t0/ugv-ocr
Last synced: 8 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/z3t0/ugv-ocr
- Owner: z3t0
- License: mit
- Created: 2019-05-11T17:21:24.000Z (about 7 years ago)
- Default Branch: master
- Last Pushed: 2019-05-22T00:06:00.000Z (about 7 years ago)
- Last Synced: 2024-12-21T14:46:04.847Z (over 1 year ago)
- Language: Python
- Size: 90.2 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# ugv-ocr
This is an OCR module for the 2019 Canada UGV Competition.
The goal of this module is to extract text from images of pylons with objects with printed text.
## Requirements
- Docker
## Usage
``` shell
cd scripts
sh ./scripts/start.sh images/example_01.jpg
```
The directory ocr-module is the base path from which the images are being detected.
To test do the following
1. Take an image in jpg
2. Place it in the ocr-module/images directory (the top level project directory is mounted into the container)
3. Run the following command, where "test.jpg" is the name of the image.
``` shell
cd scripts
sh start.sh images/test.jpg
```
## Installation
``` shell
sh ./scripts/build.sh
```
## Attributions
This module uses many technologies and tools, some are listed below.
- Python
- OpenCV
- Tesseract
- iPython
## TODO
- Re-factor into a HTTP API
- Add RTSP stream support
- Use GPU
- Testing