Script for OCR parsing of Ingress agent statistics screenshots
This project is in a early experimental status. Parsing of stats already looks fairly reliable, but needs more testing. See Testing on how to test and help.
This script is inspired by the fantastic agent-stats.com project. Automated parsing of Ingress agent stats screens would useful for many Ingress related community projects like local competitions. This project aims to get a reliable base for OCR on stats the agents provide to your project / database.
Be aware that this version is only tested on Ubuntu 14.04.
- Fork and send in pull requests!
- Bash 4
- ImageMagick
- tesseract-ocr
Basic test data is provided in the ingress-stats-parser-testdata repository. Clone a copy into your working copy of ingress-stats-parser. Use test.sh to validate the known results to your changes you applied to isp.sh.
Submitting more validated test files to ingress-stats-parser-testdata is highly appreciated!
- Cleanup of temp files
- Medal recognition
- dependency checks
- ... your ideas