FGO drop screen parser

Settings and ref files layout

├── input
    ├── quest_1
        ├── files
            ├── material_ref_1
            ├── material_ref_2
            ├── currency_ref_1
            ├── ...
        ├── settings.json
        ├── quest_1_drop_screen_1.png
        ├── quest_1_drop_screen_2.jpg
        ├── ...
    ├── quest_2
        ├── ...
    ├── ...

Manually parse files

> python fgo_mat_counter.py -h
usage: fgo_mat_counter.py [-h] [-v] [-d] [-nc] [-i IMAGE]

Helper Script that uses basic image recognition via opencv to count mat drops in FGO Screenshots

optional arguments:
  -h, --help            show this help message and exit
  -v, --verbose         Enables printing info level messages
  -d, --debug           Enables printing debug level messages and creation of temporary images useful for debug
  -nc, --nocharSearch   Disable search and labeling for characters in images (improves performance outside events)
  -i IMAGE, --image IMAGE
                        Path to image to process

If you have the layout above, the script will use the settings.json and files ref folder. If the script can not find the files ref folder at the same level as the --image, the script will use the default refs and settings at ref/ and ref/settings.json.

Example:

> python fgo_mat_counter.py -v -i input/quest_1/quest_1_drop_screen_1.png

Continously monitor the input folder

> python frontend.py -h
usage: frontend.py [-h] [-j NUM_PROCESSES] [-p POLLING_FREQUENCY]

optional arguments:
  -h, --help            show this help message and exit
  -j NUM_PROCESSES, --num_processes NUM_PROCESSES
                        Number of processes to allocate in the process pool
  -p POLLING_FREQUENCY, --polling_frequency POLLING_FREQUENCY
                        how often to check for new images in seconds

You can either run this script manually or you can run the Dockerfile. The script will monitor the /input folder, parse, move the parsed images and write the output jsons to /output. /input must have the layout above.

Development

• Use black to format Python code