So I'm having 3 days education seminar which isn't of much use for me, so I'm writing my program on a paper and I'm wondering something since I can't test it at-the-moment.

I'm basically scanning my screen for image, and screen is divided into 8 segments. I don't need to scan all 8 segments, but for instance now, let's say I only need to scan 5 out of those 8 where image can pop out.

Is it faster to just make pyautogui to scan whole screen and write down coordinates where image was seen, and I'll then know by coordinated on which screen it popped, or is it faster to make multiple pyautogui searches with defined regions? So is it faster if pyautogui searches whole screen once or if pyautogui makes multiple searches with defined regions, searching only 20-70% of screen?

If it depends on how much screen in the end I search, what would be aproximate percentage where one becomes faster than other? For instance: searching 4 times 50% of screen is faster than searching once 100% screen, but searching 6 times 75% of screen is slower than searching once 100%.

CodePudding user response：

I think that the faster and more efficient/accurate process here would be searching the whole screen through pyautogui and write down the coordinates where the image was seen, and you then knowing by the coordinated on which screen it popped... So I would reccomend using this method as the second methos would be more slower and not as efficient as this