I am currently doing the project "Automatic License Plate Detection and Recognition" and I am using AForge.Net, I used a simple algorithm; although it is not very effective, but it will meet my requirements. I calculated the ratio of height, width and height to width (height / width) greater than 40 number plate of the car; I looked for rectangles in the image that lie in the range of the ratio of height, width and height to width; now with this I can detect a license plate for most images, but the problem is that along with the license plate; I get some other rectangles; which is also in the height / width ratio range; now I want to distinguish between a rectangle containing a number plate and rectangles without a license plate;[I actually crop the rectangles and save them in bitmaps] now I have 2 solutions to solve this problem, but I don’t know how to implement them.
I am. I can make the assumption that the background of the number plate is white (this is not always, but it will work for most cases (I am ready for trafeoffs)); my question is that in Aforge.Net there is a built-in method that can look for an image on a white background; or how can I find that this photo has a white background? [if the image has a white background, then this number plate is not yet]
II. another option is to detect characters from the image; any image containing characters [we can assume that there should be at least 4 characters; therefore, we must look in the image that if it has 4 or more 4 characters, then this number plate is not there yet]; therefore, is there any method available in AForge that can be useful in this case ie, which can look for characters; if not, how can I search for characters through an image ...?
;
2 ; emguCV , ; emguCV.
PLz , ;
...
Usama