Thanks for Andre, Jimmy CA license plate Font is available. I tired to find the sample file to train my ocr, but haven't find anything yet. You are right, I may need to use alot of photoshop, but again, not sure how many LP will give me the whole set of numbers and characters. I didn't train the tesseract, becausei thought OCR will be able to figure out, since the provided images have no noise. I will email you the final images that I am providing to OCR. Most of the CA license plate are black on white, but there are color and other different type of LP there, but I am ignoring those and assuming that most of the LP characters are black on light background.
Just for curiosity, when you take the image, do you only focus on LP area or the whole car? In some of my images, there was a reflection in the image and I need to get rid of reflection some how, but haven't figured out. I used the suggested site that was supposed to give the name of the font or other information, but when i provided the image, it was not able to correctly identify the character and it didn't work. I think Jimmy had the link. I think, I need to capture enough images and then use photoshop, and then i need to read on, how to train my data. Quite of work ahead. Anyway, any of you have any idea, about scanning image and getting the LP (image was filtered using edge filter, i can see the rectangle box of LP, just need to figure out, how to scan and how to extract. The ratio of CA LP is 1 to 2, or 6 to 12 inches (height=6, width=12) thanks On Jul 30, 1:36 pm, Andres <[email protected]> wrote: > 2010/7/30 Jimmy O'Regan <[email protected]> > > > > > On 30 July 2010 20:45, Andres <[email protected]> wrote: > > > By the way, the fonts used in the licence plates in Argentina are not > > > commercial. So I had to build my training image with pictures that I took > > > with my own camera on the street. If that's your case, prepare yourself > > for > > > a lot of photoshop work, to make the size of the characters uniform > > (tips: > > > (paste) -> Ctrl+T (transform) -> drag the edges holding shift to keep > > > proportions ---->when you finish with all fonts, merge visible layers > > > (Shift+Ctrl+E) to avoid having a multilayer TIFF file------use the rulers > > to > > > guide you vertically-----finally you might dicide if you want to > > threshold) > > > > Question to the list: > > > The images that I use have black background and the letters are white. I > > > trained Tesseract for that. Does that make any difference, should I get > > > better results by inverting the image (in the training image and captured > > > image) ? > > > Tesseract is supposed to handle that gracefully, though for training > > it would be better to use black on white. > > > You mean that I can train on black on white and then read white on black > > with no difference ? > > > -- > > <Leftmost> jimregan, that's because deep inside you, you are evil. > > <Leftmost> Also not-so-deep inside you. > > > -- > > You received this message because you are subscribed to the Google Groups > > "tesseract-ocr" group. > > To post to this group, send email to [email protected]. > > To unsubscribe from this group, send email to > > [email protected]<tesseract-ocr%[email protected]> > > . > > For more options, visit this group at > >http://groups.google.com/group/tesseract-ocr?hl=en. -- You received this message because you are subscribed to the Google Groups "tesseract-ocr" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/tesseract-ocr?hl=en.
