buttons not found

[DrEffect]

the bot doesnt find the right buttons, oftern it didnt found any button and if it find a button on the completly wron position, maybe some settings error on my site?

[Krazalm]

I just fixed mine by replacing new images. U can also put any img and it will click them

[DrEffect]

maybe you can upload Your Fixed images anywhere?:-)

[Krazalm]

I dont think it will work due to my computer color tone I recommend to do it in ur pc manually

[DrEffect]

just take a screenshot of it? Any special recommendations for the Screenshot? hopefully i will get it work.... somethin else where i have a look on it? cause its detected, completely the whole buttons

Am Mo., 7. Aug. 2023 um 13:36 Uhr schrieb Krazalm @.***

:

I dont think it will work due to my computer color tone I recommend to do it in ur pc manually

— Reply to this email directly, view it on GitHub https://github.com/lewisgibson/monopoly-go-bot/issues/7#issuecomment-1667693200, or unsubscribe https://github.com/notifications/unsubscribe-auth/AIUEUNCRMGTAOSP4R3UMUC3XUDHNHANCNFSM6AAAAAA3FINDNU . You are receiving this because you authored the thread.Message ID: @.***>

[Krazalm]

these are my images(16)

[lewisgibson]

Although we do image detection in black and white (xor comparison) there are still some issues with slight variations of grey. The solution is to probably make the searching a little more aggressive.

[lewisgibson]

You can change the confidence parameter here if you wish to do that: https://github.com/lewisgibson/monopoly-go-bot/blob/main/main.py#L56

[DrEffect]

higher is more aggresiv i think?

[lewisgibson]

higher is more aggresiv i think?

It's the other way around. A value of 1.00 (100%) means the screen capture and the target image need to be 100% identical. A value of 0.75 (75%) means they need 75% similarity.

The comparison is done using OpenCV's matchTemplate call in TM_CCOEFF_NORMED mode.

https://stackoverflow.com/a/55469775 https://docs.opencv.org/4.0.1/df/dfb/group__imgproc__object.html#ga3a7850640f1fe1f58fe91a2d7583695d

[Krazalm]

Hey edited ur code

Optimized and Enhanced Code:

The code has been enhanced for improved readability and performance. Configurable settings have been organized for easier management. The code now relies on the pyautogui library(it works) for faster mouse input and has been optimized for efficiency. Logging has been introduced to track the code's execution.

Notable Changes:

• Replaced the usage of 'pyautogui.locateOnScreen' with 'pyautogui.locateCenterOnScreen' for precise mouse positioning.
• Removed dependencies on 'pydirectinput' and 'sys' for a cleaner codebase. #Resolution affects image recognition so make sure to not resize the app or u will need new image for that resolution

main.py.txt

higher is more aggresiv i think?

It's the other way around. A value of 1.00 (100%) means the screen capture and the target image need to be 100% identical. A value of 0.75 (75%) means they need 75% similarity.

The comparison is done using OpenCV's matchTemplate call in TM_CCOEFF_NORMED mode.

https://stackoverflow.com/a/55469775 https://docs.opencv.org/4.0.1/df/dfb/group__imgproc__object.html#ga3a7850640f1fe1f58fe91a2d7583695d

[Krazalm]

Will also do a priority list function although its 2 am for me im a do it tommorow i got whole day 😊

[DrEffect]

ok slowly it will work, but now when i build and doent have money, it always try to build, is there anywhere a routine to stop doing the Same?

[lewisgibson]

@DrEffect Building is experimental. Delete the images to remove build support.