I need to extract captcha, I’m researching opencv. My goal is to solve this form of captcha reliably.

I am writing a program to segment the different lines from a cropped image of paragraph with handwritten text. This is my current code which works in most cases but sometimes struggles and combines several lines into one.

I am trying to extract text from an image but seems however I do it tessaract gives me some random values even though I think I have processed the image to a very good format.

How to filter out unwanted contours during real-time OCR?
Additional unwanted contour
In the image above, with just a slight camera movement, additional contours appear.