How to train two objects of different sizes for recognition in OpenCv

Question

How to train two objects of different sizes for recognition in OpenCv

Navigation

#1 by (3 votes)

6

Hello, I'm using OpenCv version 2.4.11 to train the recognition of vehicle license plates for parking software. The problem is when training to identify the car plate, truck, truck where the boards used by them are of the same dimensions the cascade gets 100%.

When I train to recognize the motorcycle boards along with the car plates the answer is not effective when I try to capture the motorcycle boards because the dimensions are different.

The plates of cars and trucks etc. has the shape of a rectangle since the bikes is more for a square so if I train the motorcycle vectors and other vehicles together, it will only work the recognition either one or the other.

android c++ opencv

asked by anonymous 06.05.2015 / 17:26

1 answer

Select database record for variable Detect whether the event was triggered by the user or via script

score 3 · Answer 1

Of all the research I've done on this subject have not been satisfactory, but according to responses on the Open CV website we came to a conclusion that we should train the two types of separate board, ie we train the vectors for board search in a cascade for vehicle license plates and another cascade for search of license plates.

Within the software is that we must make the distinction, that at a moment before identification the user will inform if the license is motorcycle or vehicle (car, truck, pickup truck).