Textual representation of an image

• US 10,185,887 B2
• Filed: 02/27/2013
• Issued: 01/22/2019
• Est. Priority Date: 02/27/2013
• Status: Active Grant

• Current Assignee: Longsand Limited (Open Text Corporation)
• Original Assignee: Longsand Limited (Open Text Corporation)
• Inventors: Blanchflower, Sean, Davis, Stephen
• Primary Examiner(s): Robinson, Greta L
• Application Number:
  US14/769,621
• Publication Number:
  US 20160004928A1
• Time in Patent Office: 2,155 Days
• Field of Search: 707756, 707758, 707760, 707915, 707999107, 382173, 382175, 382176, 382177, 382194, 382196, 382232, 382276, 382290
• CPC Class Codes: G06F 16/22 Indexing; Data structures therefor; Storage structures
• CPC Class Codes: G06F 16/5838 using colour
• CPC Class Codes: G06F 16/5846 using extracted text
• CPC Class Codes: G06T 1/0007 Image acquisition
• CPC Class Codes: G06T 2207/10004 Still image; Photographic image
• CPC Class Codes: G06T 2207/20021 Dividing image into blocks, subimages or windows
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06V 10/424 Syntactic representation, e.g. by using alphabets or grammars
• CPC Class Codes: G06V 30/1983 Syntactic or structural pattern recognition, e.g. symbolic string recognition

Abstract

At least a computer-implemented method and an apparatus for processing an image are described. In examples, numeric values for at least one property of the image are determined. These values are then converted into at least one corresponding text character, said conversion being independent of any text content within the image. This enables a text representation of the image to be generated that contains said plurality of text characters. This text representation may be used to index and search for the image.

Claims

1. A computer-implemented method of processing an image comprising:

determining, by a processor of a computing device, a numeric value that represents an average pixel intensity of colors within each of a plurality of regions of the image, the numeric value representing the average pixel intensity for each region of the image being independent of any text content within the region of the image;

converting, by the processor, each of the numeric values for each region of the image into an alphabet representing the average pixel intensity for the region, wherein the alphabet for the region of the image is independent of any text content within the region of the image; and

combining, by the processor, the alphabets for the regions of the image into a string of alphabets to represent the average pixel intensities of the image; and

outputting, by the processor, the string of alphabets of the image,

wherein the string of alphabets of the image is used to construct a search query to be applied to a database storing previously-generated strings of alphabets of different images to locate the image.

2. The computer-implemented method according to claim 1, further comprising:

storing the string of alphabets representing the average pixel intensities of the image in an index of the database.

3. The computer-implemented method according to claim 1, wherein the alphabet for each region includes a string of three alphabets representing average pixel intensities of three different colors in the region.

4. The computer-implemented method according to claim 1, wherein the method further comprises, for each particular region of the image:

determining a text character that represents an identifier of the particular region of the image; and

combining the text character representing the identifier of the particular region and the alphabet representing the average pixel intensity for the particular region into one word for the particular region, wherein a text representation of the image comprises a combination of the words for the regions of the image.

5. The computer-implemented method according to claim 1, further comprising:

determining a mapping table that maps a plurality of ranges of numeric values to corresponding alphabets,

wherein converting each of the numeric values into an alphabet comprises using the mapping table to determine the alphabet corresponding to the numeric value of each region of the image.

6. The computer-implemented method according to claim 1, wherein determining a numeric value for each of the plurality of regions of the image further comprises determining a numeric value for a feature in the region of the image, the feature being representative of a property of the region of the image.

7. The computer-implemented method according to claim 6, wherein converting each of the numeric values for each region of the image into an alphabet comprises combining the alphabet corresponding to the numeric value for the region with one or more of:

a text character identifying a particular feature in the region; and

a text character identifying a location of the particular feature in the region of the image, wherein the alphabet and text characters are combined to generate a word to represent the region of the image and wherein a text representation of the image comprises a combination of the words for the regions of the image.

8. Apparatus for processing an image comprising:

an interface to receive an image;

an image processor communicatively coupled to the interface and to determine a numeric value that represents an average pixel intensity of colors within each of a plurality of regions of the received image, the numeric value representing the average pixel intensity for each region of the image being independent of any text content within the region of the image; and

a text converter communicatively coupled to the image processor to receive the numeric values, convert each of the numeric values for each region of the image into a text character representing the average pixel intensity of the region, wherein the text character for the region of the image is independent of any text content within the region of the image, and output a string of the text characters representing the average pixel intensities of the regions of the image,

wherein the string of the text characters of the image is to be used to construct a search query to be applied to a database storing previously-generated strings of text characters of different images to locate the image.

9. Apparatus according to claim 8, wherein the text converter is to determine a text character that represents an identifier of each region of the image, the text converter to combine the text character representing the identifier of the region of the image with the text character converted from the numeric value representing the average pixel intensity of the region into one word, wherein a text representation of the image output by the text converter comprises a combination of the words for the regions of the image.

10. An indexing system comprising:

an interface for receiving an image;

an image processor communicatively coupled to the interface and to determine a numeric value that represents an average pixel intensity of colors within a plurality of regions of the received image, the numeric value representing the average pixel intensity for each region of the image being independent of any text content within the region of the image;

a text converter communicatively coupled to the image processor to convert each of the numeric values for each region of the image into a text character representing the average pixel intensity of the region, wherein the text character for each region of the image is independent of any text content within the image, and output a string of the text characters representing the average pixel intensities of the regions of the image; and

a database to store previously generated strings of text characters representing average pixel intensities of different images,

wherein the string of text characters representing the average pixel intensities of the image is used to construct a search query to be applied to the database to locate the image.

11. The indexing system of claim 10, wherein the text character for each region is an alphabet.

12. The indexing system of claim 10, wherein the text converter is to:

determine a text character that represents an identifier of each particular region of the image; and

combine the text character converted from the numeric value for the particular region and the text character representing the identifier of the particular region into one word, wherein a text representation of the image comprises a combination of the words for the regions of the image.

13. A system for image matching comprising:

a search query interface for receiving an image to be matched;

an image processor communicatively coupled to the search query interface to determine a numeric value that represents an average pixel intensity of colors within each of a plurality of regions of the received image, the numeric value for each region of the image being independent of any text content within the region of the image;

a text converter communicatively coupled to the image processor to convert each of the numeric values for each region of the image into an alphabet representing the average pixel intensity of the region, wherein the alphabet for the region of the image is independent of any text content within the region of the image, and output a string of the alphabets representing the average pixel intensities of the regions of the image;

a text database storing a plurality of strings of alphabets representing average pixel intensities of a plurality of images; and

a search query processor communicatively coupled to the text converter and the text database to use the string of alphabets output by the text converter to construct a search query to be applied to the text database to match the string of alphabets of the image output by the text converter against the plurality of strings of alphabets stored in the text database.

14. The system according to claim 13, wherein the search query processor is to probabilistically match the string of alphabets of the image output by the text converter against the plurality of strings of alphabets stored in the text database.

15. The system according to claim 13, further comprising:

an image database storing the plurality of images; and

an image matching processor to match numeric features of the image received at the search query interface against the plurality of images stored in the image database, wherein the output of the image matching processor and the search query processor are combined to match the image received at the search query interface.

16. A computer program product comprising a non-transitory computer-readable storage medium having computer readable instructions stored thereon, the computer readable instructions being executable by a computerized device to cause the computerized device to:

receive an image;

determine a numeric value that represents an average pixel intensity of colors within each of a plurality of regions of the image, the numeric value representing the average pixel intensity for each region of the image being independent of any text content within the region of the image;

convert each of the numeric values for each region of the image into an alphabet, wherein the alphabets for the regions of the image are independent of any text content within the regions of the image; and

output a string of the alphabets representing the average pixel intensities of the image,

wherein the string of the alphabets of the image is used to construct a search query to be applied to a database storing previously-generated strings of alphabets of different images to locate the image.

17. The computer program product of claim 16, wherein the computer readable instructions are to cause the computerized device to:

store the string of alphabets of the image in an index of the database.

18. The computer program product of claim 16, wherein the computer readable instructions to convert each of the numeric values for each region into an alphabet include computer readable instructions that are to cause the computerized device to:

convert each of the numeric values for each region of the image into an alphabet based on a mapping table that maps ranges of pixel intensity values to corresponding alphabets.

19. The computer program product of claim 16, wherein the computer readable instructions are to cause the computerized device to:

determine a text character that represents an identifier of each particular region of the image; and

combine representing the identifier of the particular region and the alphabet for the particular region into one word for the region, wherein a text representation of the image comprises a combination of the words for the regions.