alex-ong / NESTrisOCR

OCR for statistics in NESTris
24 stars 7 forks source link

Is this program support linux? #18

Open kennyfs opened 4 years ago

kennyfs commented 4 years ago

I tried it, but it seems to be not working.

timotheeg commented 4 years ago

@kennychenfs OCR detection should work fine on linux (it's python after all) BUT, there is no window capture method for linux's windowing systems implemented at this moment. Feel free to add one if you'd like.

If you are using an input device from a real NES, you might be able to use the OpenCV capture class.

alex-ong commented 4 years ago

Are you a programmer? If not then it's probably going to be hard to get this running...

kennyfs commented 4 years ago

Yes, but I'm not familiar with video process at all.

alex-ong commented 4 years ago

I'm glad you're a developer kenny! Use OpenCV capture class to get a video stream first.

I believe it's config.ini -> [calibration][capture-method] = OPENCV Then in calibrate.py, choose which device it is (0-> num_devices).

After that, calibrate as per normal; then close it and run main.py. You might want to uncomment some parts of main.py to see if the output is correct.

alex-ong commented 4 years ago

I've booted it in linux and the main problem is window-name results in 10+ duplicates. E.g. "OBS" will result in 10 windows, and the current implementation picks the first one. This is problematic of course...