I have thousands of files (200,000~) with just numbers... So I need a Python script with Tesseract that does OCR on png and jpg images and renames file with like first 100-150 characters (win file name limit?) found on that image. Most ideally would be if entire text in image is embedded in EXIF or IPTC metadata.
Anyone has ever created something like this? Would be nice if you could share.
The problem isn't the software, it's the training. Here's a list of some potential helpful repos. Certainly the nsfw one might be a good place to start. github.com/topics/image-classifier?l=python
Liam Roberts
>Most ideally would be if entire text in image is embedded in EXIF or IPTC metadata No, you can't get away with it.
He made something similar, but his output is in a text files... I need text embedded..
i already looked on github, couldn't find anything useful. tags would the most ideal, if it scanned through all of text and outputted certain tags related to topic, then filename would be multiple tags...
I would all 4 of them, but the second is the best.
First will have gravity problems soon enough. Third is hiding her teeth and fourth will get fat.
Landon Butler
help me organize my george floyd meme folder pls i have over 2TB of fentanyl floyd memes
Leo Cox
>I need text embedded.. So? Run that first then write something to apply the text files to the images - or make the changes inline. stackoverflow.com/questions/3081444/changing-a-files-metadata-in-python Doesn't sound like you know how to code or just can't be bothered to put in the work, desu.
Angel Collins
>Doesn't sound like you know how to code or just can't be bothered to put in the work, desu.
It's all too time consuming (would take me like 1 month to make it), I'm not a professional programmer, yes, I'm a beginner.