METHOD AND APPARATUS FOR IMAGE MATCHING

• US 20160125253A1
• Filed: 04/25/2014
• Published: 05/05/2016
• Est. Priority Date: 05/09/2013
• Status: Active Grant

• Current Assignee: TATA Consultancy Services Limited (Tata Sons Pvt Ltd.)
• Original Assignee: TATA Consultancy Services Limited (Tata Sons Pvt Ltd.)
• Inventors: CHATTOPADHYAY, Tanushyam, VISVANATHAN, Aishwarya
• Granted Patent: US 9,679,218 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06V 10/751 Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
• CPC Class Codes: G06V 20/625 License plates
• CPC Class Codes: G06V 20/63 Scene text, e.g. street names

Abstract

Disclosed is a method and apparatus for computation and processing of an image for image matching. The apparatus here is configured to pre-process plurality of images for creating an image template. Next, the test image is extracted and pre-processed for assessing the degree of match between the test image components and the image components of the images in the image template, based a position based matching score, a feature based matching score or both.

Claims

1. A method for image matching, comprising:

receiving plurality of pre-processed images and deriving at least one image template therefrom;

receiving at least a pre-processed test image, said pre-processed test image comprising one or more image components along with one or more features associated with the image components; and

employing a matching process to determine a matching degree between the test image component and the image components of the plurality of pre-processed images of the image template, the matching process, comprising steps of;

computing a position based matching score;

computing a feature based matching score; and

determining if the matching degree determined from at least one of the position based matching score, the feature based matching score or a combination thereof, exceeds a predetermined threshold degree.

2. The method of claim 1, wherein the image template comprises of the plurality of pre-processed images, the image components thereof along with one or more features associated with the image components.

3. The method of claim 1, wherein the plurality of pre-processed images and the pre-processed test image are pre-processed in steps of:

binarizing an image;

de-skewing the binarized image;

noise cleaning the de-skewed image;

size normalizing the noise cleaned image; and

extracting plurality of features from the size normalized image.

4. The method of claim 2, wherein the extraction of plurality of features comprises extracting vertical projection, horizontal projection, contour and a stroke direction from the image.

5. The method of claim 1, wherein the position based matching score is obtained by:

defining boundary co-ordinates of the image component associated with the pre-processed test image and the plurality of pre-processed images of the image template,

computing relative co-ordinates of the image components associated with the pre-processed test image and the plurality of pre-processed images of the image template;

computing similar relative co-ordinates to make the image components resolution independent; and

deriving the position based matching score based on matching of the similar relative coordinates of the image components associated with the first pre-processed image and the plurality of pre-processed images of the image template.

6. The method of claim 1, wherein the feature based matching score is obtained by matching the one or more features associated with the image components of the test image against the one or more features associated with the image components in the image template.

7. An image matching apparatus, the apparatus comprising:

a processor;

a memory coupled to the processor, wherein the processor is capable of executing a plurality of modules stored in the memory, and wherein the plurality of module comprising;

a receiving module to receive

at least one pre-processed test image, said pre-processed test image comprising one or more test image components along with one or more features associated with the image components;

a plurality of pre-processed images for creating an image template;

a template creation module to create at least one image template from the plurality of pre-processed images; and

a matching module, configured to compute

a position based matching score;

a feature based matching score or a combination thereof, test image component and the image components of the plurality of pre-processed images of the image template.

8. The image matching apparatus of claim 7, further comprising a capturing device including:

a capturing module to capture one or more image;

a pre-processing module to extract one or more image components along with one or more features associated with the image components; and

transmitting one or more image components along with one or more features associated with the image components, to the memory.

9. The image matching apparatus of claim 7, wherein the pre-processed image is obtained in steps of:

binarizing the image;

de-skewing the binarized image;

noise cleaning the de-skewed image;

size normalizing the noise cleaned image; and

extracting the plurality of features from the size normalized image.

10. The image matching apparatus of claim 7, wherein the image template comprises of the plurality of pre-processed images, the image components thereof along with one or more features associated with the image components.

11. The image matching apparatus of claim 7, wherein the matching module is configured to compute the position based matching score is derived in steps of:

defining boundary co-ordinates of the image component associated with the pre-processed test image and the plurality of pre-processed images of the image template,

computing relative co-ordinates of the image components associated with the pre-processed test image and the plurality of pre-processed images of the image template;

computing similar relative co-ordinates to make the image components resolution independent; and

deriving the position based matching score based on matching of the similar relative coordinates of the image components associated with the first pre-processed image and the plurality of pre-processed images of the image template.

12. The image matching apparatus of claim 7, wherein the matching module is configured to compute the feature based matching score by matching the one or more features associated with the image components of the test image against the one or more features associated with the image components in the image template.