Optical Character Recognition


Published on

How to perform OCR on Vehicular Images

1 Like
  • Be the first to comment

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

Optical Character Recognition

  1. 1. Optical Character Recognitionon Vehicle Number PlateAvinash Singh Bagri2009MT50541EEL 709Dr Sumeet Agarwal
  2. 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 areais searched again from the image.If characters are recognized then number plate search is terminated.
  3. 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. 4. ExampleOriginal ImageBinarised imageInverted binarised image
  5. 5. Extracting Number PlateOne piece of image that will be tested for number plate
  6. 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 pieceinto possible characterImage piece with maximum peaks in candidate character is selected as the numberplateColumn signature of the number plate image Column signature of the another image piece
  7. 7. Parsing plateImages of all characters
  8. 8. Recognition of CharactersMethod of recognition of characters from an image containing thesecharacters is based on object recognition techniques used in Digital ImageProcessing.Two commonly used techniquesTemplate Matching using CorrelationDistance Measurement
  9. 9. Template MatchingTemplate matching using correlationBased on performing correlation between segmented imageA character is required to be recognized and character template imagewhich is used for recognition
  10. 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. 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 resultantimage.
  12. 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 templateimage is required
  13. 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. 14. Thank you