I'm exploring whether it's feasible to run an overlay image recognition tool on random websites, focusing on detecting specific changes. I'm not concerned about speed—taking 5-10 seconds per site is fine. The idea is to mimic human interaction with the site, capturing certain elements (like a stock value displayed on the page) and saving that data to a file for further analysis. I'm wondering if there are any extensions or overlays that could facilitate this process, along with any tips to manage potential timeouts or errors during the data review phase.
1 Answer
It sounds like what you're trying to achieve is definitely possible! But rather than an overlay, I recommend checking the Document Object Model (DOM) of the website. DOM lets you see the structure of the webpage, so you could inspect elements directly and grab the data you need. It might be simpler and more effective than going for image recognition.
Thanks for the clarification! I wasn't familiar with DOM. So, if I load a webpage that shows stock values, I can pinpoint that number without directly using the website's API? And you're suggesting that I could even automate clicks on buttons based on their position in the DOM?