Updated mwo-screenshot-analyzer to V0.3 Contact: [email protected] Github: https://github.com/gnijieb/mwo-screenshot-analyzer

DISCLAIMER: Good results will only be achieved when requirements are met.

How do you use it? What does it do?

• put screenshots in input folder
• run script
• get raw data

You basically get the numbers/data from the end-of-round screen (as CSV) to process further (e.g. refer to the google spreadsheet below).

Info/Requirements

• Supported resolutions: 1920x1080, 1920x1200, 2560x1440
• Screenshots should be taken in a non-lossy format (e.g. png using FRAPS), lossy formats like jpg will produce significantly more errors.
• Only Quickplay (Solo or Group) supported

Installation on Windows

• Python Installer

  • Run the installer in resources folder ("python-2.7.11.msi", source: https://www.python.org/downloads/release/python-2711/)
  • IMPORTANT: Make sure to tick the option "Add python.exe to Path" in the installer

• tesseract-ocr

  • run the installer in resources folder ("tesseract-ocr-setup-3.05.00dev.exe", source: https://github.com/UB-Mannheim/tesseract/wiki
  • IMPORTANT: In installer tick registry settings for PATH

• Get Python modules:

  • Open a Command Prompt (cmd.exe)

  • Change directory to the location of mwo-screenshot-analyzer

    • e.g.: "cd C:\mwo-screenshot-analyzer"

  • Levenshtein:

    • Run: "pip install resources\python_Levenshtein-0.12.0-cp27-none-win32.whl" (source: http://www.lfd.uci.edu/~gohlke/pythonlibs/#python-levenshtein)

  • update pip (package manager)

    • run: "python -m pip install -U pip"

  • Python Imaging Library (PIL fork): Pillow

    • run: "pip install Pillow"

  • pytesseract

    • run: "pip install pytesseract"

• create directories in mwo-screenshot-analyzer folder (if not exist): input, output, intermediate, processed

Usage

• You need screenshots of both the player and team stats (end of round) in a specific order (don't mix):

  • first: teamstats, second: playerstats
  • first: playerstats, second: teamstats

• place your screenshots in input directory

• Open a Command Prompt (cmd.exe)

• Change directory to the location of mwo-screenshot-analyzer

  • e.g.: "cd C:\Users\USERNAME\Desktop\mwo-screenshot-analyzer"

• Example with screenshots created with FRAPS (player stats first, team stats second)

  • run: "python analyze.py -p "PLAYERNAME" -t fraps -m3"
  • run: "python analyze.py --help" to see available arguments/usage

• if something is not recognized correctly or as expected, you are asked to open the image in input folder and enter the desired value or select from a given list of best matches

• processed screenshots are moved to the processed folder

• the result is in form of CSV in: output/2016-05-01 20-56-49_data.csv

Optional: External Processing You can create your own external processing module to e.g. submit the data to a database or save it in a specific format. Refer to ext_example.py.

If anyone is interested, feel free to check it out at https://github.com/gnijieb/mwo-screenshot-analyzer

OR: Download Prepackaged ZIP including Python, Tesseract-OCR installers and prebuilt Python modules (Links to those downloads are also in the README).

Usage is explained in the README file.

EXAMPLE of evaluation: https://docs.google.com/spreadsheets/d/1L7uE4kvwKsK5Vhcz5K5A7WlRuirLKF70BmLtKdE4454/edit?usp=sharing (the data on the input tab is the output of the script)

SAMPLE CSV data: http://pastebin.com/r3rWVTTP

If you want to contribute/help out or if you encounter any issues, please PM me.