OCRExtract dynamic image

I am trying to extract information from this section. (this is from Linkedin Recruiter). The information inside, the order of the elements and the size changes for each record. I want to find a specific skills and the number behind it. I thought I could use OCRExtract but for the life of me cannot figure out how to use it with a dynamic image. The documentation on this command is not so good. I cannot find a SINGLE example of using OCRExtract. Every example encourages to use OCRExtractRelative but how can I define green and pink sections on a dynamic image?

I have never used a tool that got me more excited and at the same time frustrated as this one. One moment I want to buy it and the next I want to delete it from my system forever.


What is dynamic in this image? Maybe a larger screenshot or video would help.

Thank you Ulrich. I think you are doing a tremendous job keeping track and answering the posts on the forum. And apologize for not being clear. Everything in this box can change. The number of elements (skills) can change, which skills are listed and of course if there are less than 15 skills there will not be a show more button.

However the problems solved itself. My challenge is that I am so fascinated by the OCR feature that I forget to try the more conventional methods. In this case I was able to use storetext and the whole darned thing was in my variable. Amazing. Then I wrote a little sub in excel to parse the string and just like that the data was in my db.

I must have tried EVERY scraping tool out there in order to grab information from LinkedIn. You can’t really use cloud scraping because of the blocks that LinkedIn has put in their system. An interactive tool with a live web page is the only way (I found) and once you get to know UI.Vision it is extremely powerful.

1 Like