programmatically identify the second or third largest contour to include the Vehicle Number PLate

[RohitDhankar]

• This issue presents itself after we have reached the stage in code as mentioned below

No model has been trained yet to identify Number Plates on vehicles .
Current experimental set-up is just taking random Images from the wild , which have number plates in them , and trying to do OCR on them with PyTesseract.
Current effort is not about Object Detection - Number Plates on vehicles, its only about OpenCV transforms etc - after the Object has been identified.
Focus is on OCR of the Number PLates - that too only using OpenCV with PyTesseract.

Have been able to Sort and identify - the contour which had the number plate , for a small set of images that the Number plates lies within .
Now this approach needs to be further generalized - using OpenCV only .
Current efforts being tracked in this issue.