# Fine tuning image selctions

**URL:** <https://forum.opencv.org/t/fine-tuning-image-selctions/24569>\
**Category:** Python\
**Tags:** ocr\
**Created:** [June 9, 2026, 6:33am UTC](https://forum.opencv.org/t/fine-tuning-image-selctions/24569 "2026-06-09T06:33:26Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![Shiva\_Kumar](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.opencv.org/shiva_kumar/32/14302_2.png) [@Shiva\_Kumar](https://forum.opencv.org/u/Shiva_Kumar)\
**Post date:** [June 9, 2026, 6:33am UTC](https://forum.opencv.org/t/fine-tuning-image-selctions/24569/1 "2026-06-09T06:33:26Z")

</div>

Hi all, I am finetuning a OCR model for document analysis. so i have a bunch of images and i need to remove some of thema and only consider few of them using criteria’s like

-\> Having tables  
 → Having more than 60% text on page  
 → can have any special images like signs, stamps..etc

So any suggestion how to do?  
Thanks in advance!!!

---

<div class="post-metadata">

**Author:** ![crackwitz](https://sea2.discourse-cdn.com/flex020/user_avatar/forum.opencv.org/crackwitz/32/14_2.png) [@crackwitz](https://forum.opencv.org/u/crackwitz)\
**Post date:** [June 9, 2026, 10:09am UTC](https://forum.opencv.org/t/fine-tuning-image-selctions/24569/2 "2026-06-09T10:09:58Z")

</div>

curating the data set is a _human_ job. you can pay some people to help you do this. there are websites where you can offer such tasks.