Optical Character Recognition
- 1. Optical Character Recognition on Vehicle Number Plate Avinash Singh Bagri 2009MT50541 EEL 709 Dr Sumeet Agarwal
- 2. Overview Takes image of the car and searches for the number plate in the image. Once the probable number plate area is located it is given to OCR. If OCR doesn’t recognize the characters from the image number plate area is searched again from the image. If characters are recognized then number plate search is terminated.
- 3. Steps Involved Image division into small images detecting probable number plate area Recognizing number plate area Parsing number plate to extract characters Apply OCR to the parsed characters
- 4. Example Original Image Binarised image Inverted binarised image
- 5. Extracting Number Plate One piece of image that will be tested for number plate
- 6. Recognizing Plate Search number plate in the broken pieces of vehicle image Apply peak to valley to the candidate image pieces to further break the image piece into possible character Image piece with maximum peaks in candidate character is selected as the number plate Column signature of the number plate image Column signature of the another image piece
- 7. Parsing plate Images of all characters
- 8. Recognition of Characters Method of recognition of characters from an image containing these characters is based on object recognition techniques used in Digital Image Processing. Two commonly used techniques Template Matching using Correlation Distance Measurement
- 9. Template Matching Template matching using correlation Based on performing correlation between segmented image A character is required to be recognized and character template image which is used for recognition
- 10. Correlation Modified form of convolution f(x,y): gray scale value at a specific element (x,y) in an image f(x,y): image g(x,y) character template h(x,y) image after correlation.
- 11. Result of Correlation Basic form of convolution Result is an image, convolution of two matrices The size of result matrix will be increased from input image matrices Due to which we have to apply some thresh holding on resultant image Normally value of thresh hold is little less than maximum value of resultant image.
- 12. Limitations Noise free image with uniform illumination required Numbers must be displayed in one line on the number plate Problem associated with template image is proper acquisition of template image is required
- 13. References http://www.sersc.org/journals/IJUNESST/vol6 no1/2.pdf http://www.ele.uri.edu/~hansenj/projects/ele585/OCR/OCR.pdf http://perun.pmf.uns.ac.rs/radovanovic/dmsem/completed/2006/OCR.pdf http://www.nicomsoft.com/optical-character-recognition-ocr-how-it-works/ http://www.ancient-asia-journal.com/article/view/aa.06113/25 License Plate Number Recognition - New Heuristics and a Comparative Study of Classifiers (2F79e4150656ca79aeb9.pdf) Kwaśnicka H. and WawrzyniakLicense B.; Plate Localization and Recognition in Camera Pictures
- 14. Thank you