System and Method for Processing Image for Identifying Alphanumeric Characters Present in a Series

• US 20130272607A1
• Filed: 03/25/2013
• Published: 10/17/2013
• Est. Priority Date: 03/26/2012
• Status: Active Grant

• Current Assignee: Indian Statistical Institute, TATA Consultancy Services Limited (Tata Sons Pvt Ltd.)
• Original Assignee: Indian Statistical Institute, TATA Consultancy Services Limited (Tata Sons Pvt Ltd.)
• Inventors: Chattopadhyay, Tanushyam, Bhattacharya, Ujjwal, Chaudhuri, Bidyut Baran
• Granted Patent: US 8,965,123 B2
• Time in Patent Office: Days
• US Class Current: 382/171
• CPC Class Codes: G06V 30/10 Character recognition
• CPC Class Codes: G06V 30/153 using recognition of characters or words
• CPC Class Codes: G06V 30/162 Quantising the image signal

Abstract

A system and a method for identification of alphanumeric characters present in a series in an image are disclosed. The system and method captures the image and further processes it for binarization by computing a pattern of the image. The generated binarized images are then filtered for removing unwanted components. Candidate images are identified out of the filtered binarized images. All the obtained candidate images are combined to generate a final candidate image which is further segmented in order to recognize a valid alphanumeric character present in the series.

Claims

1. A method for identification of alphanumeric characters present in a series in an image, the method comprising processor implemented steps of:

capturing the image comprising the series of alphanumeric characters;

processing the image for producing a set of identifiable characters out of the series of alphanumeric characters;

the processing further comprising;

computing a pattern for recognizing a pixel intensity distribution in the image for determining a background peak and a foreground peak;

generating a plurality of binarized images by selecting a plurality of dynamic threshold values between the background peak and the foreground peak;

filtering the generated binarized images by removing unwanted components from plurality of images to identify one or more valid characters;

identifying one or more candidate images by comparing the valid characters with respect to a known ground truth value;

generating a final candidate image by combining the candidate images such that the combination of the candidate images is dependent upon a predefined condition; and

splitting the final candidate image into a predefined segments and recognizing a valid alphanumeric character associated with each segment therein.

2. The method as claimed in claim 1, wherein the series of alphanumeric characters includes but is not limited to 17 alphanumeric characters.

3. The method as claimed in claim 1, wherein the pattern for recognizing the pixel intensity is in a form of histogram.

4. The method as claimed in claim 1, wherein the plurality of pixel values depends upon the number of alphanumeric characters present in the series.

5. The method as claimed in claim 1, wherein the unwanted components comprises of too big and too small components, such that a component is too small if the number of pixels with that particular level is less than 100 and a component is too big if the number of pixels with that particular level is width/4.

6. The method as claimed in claim 1, wherein the one or more candidate images are identified by disregarding over-segmented and under-segmented binarized images.

7. The method as claimed in claim 1, wherein the ground truth value is equal to the number of alphanumeric characters present in the series.

8. The method as claimed in claim 1, wherein the predefined condition for combining the candidate images refers to marking the pixels of all the candidate images as background if more than half of the candidate images are background.

9. The method as claimed in claim 1, wherein the candidate image is split into a number equal to the number of alphanumeric characters present in the series.

10. The method as claimed in claim 1, wherein the multiple component labels in the valid alphanumeric character are merged by applying morphological closing.

11. A system for identification of alphanumeric characters present in a series in an image, the system comprising:

an image capturing device for capturing the image comprising the alphanumeric characters present in the series;

a processor configured to produce a set of identifiable characters out of the series of alphanumeric characters;

the processor further comprising;

a computing module configured to compute a pattern for recognizing a pixel intensity distribution in the image for determining a background peak and a foreground peak;

a binarization module configured to generate a plurality of binarized images by selecting a plurality of dynamic threshold values between the background peak and the foreground peak;

a filter configured to remove unwanted components from the plurality of images to identify one or more valid characters;

a comparator configured to compare the valid characters with respect to a known ground truth value in order to identify one or more candidate images;

an image generator configured to generate a final candidate image by combining the candidate images such that the combination of the candidate images is dependent upon a predefined condition; and

an output generating module configured to split the final candidate image into a predefined segments and recognizing a valid alphanumeric character associated with each segment therein.

12. The system as claimed in claim 11, wherein the image capturing device includes a camera.

13. The system as claimed in claim 11, wherein the computing module computes a histogram pattern for recognizing the pixel intensity.

14. The system as claimed in claim 11, wherein the captured image includes 17 alphanumeric characters present in the series.