SYSTEM AND METHODS FOR ARABIC TEXT RECOGNITION BASED ON EFFECTIVE ARABIC TEXT FEATURE EXTRACTION

• US 20100272361A1
• Filed: 04/27/2009
• Published: 10/28/2010
• Est. Priority Date: 04/27/2009
• Status: Active Grant

• Current Assignee: Hussein K. Al-Omari, Khalid M. Alfaifi, Khalid M. Alhazmi, Mohammad S. Khorsheed
• Original Assignee: Hussein K. Al-Omari, Khalid M. Alfaifi, Khalid M. Alhazmi, Mohammad S. Khorsheed
• Inventors: Al-Omari, Hussein K., Alhazmi, Khalid M., Alfaifi, Khalid M., Khorsheed, Mohammad S.
• Granted Patent: US 8,111,911 B2
• Time in Patent Office: Days
• US Class Current: 382/187
• CPC Class Codes: G06V 10/50 by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
• CPC Class Codes: G06V 30/153 using recognition of characters or words
• CPC Class Codes: G06V 30/224 of printed characters having additional code marks or containing code marks
• CPC Class Codes: G06V 30/2276 with probabilistic networks, e.g. hidden Markov models
• CPC Class Codes: G06V 30/293 of characters other than Kanji, Hiragana or Katakana

Abstract

A method for automatically recognizing Arabic text includes digitizing a line of Arabic characters to form a two-dimensional array of pixels each associated with a pixel value, wherein the pixel value is expressed in a binary number, dividing the line of the Arabic characters into a plurality of line images, defining a plurality of cells in one of the plurality of line images, wherein each of the plurality of cells comprises a group of adjacent pixels, serializing pixel values of pixels in each of the plurality of cells in one of the plurality of line images to form a binary cell number, forming a text feature vector according to binary cell numbers obtained from the plurality of cells in one of the plurality of line images, and feeding the text feature vector into a Hidden Markov Model to recognize the line of Arabic characters.

Claims

1. A method for automatically recognizing Arabic text, comprising:

acquiring a text image containing a line of Arabic characters;

digitizing the line of the Arabic characters to form a two-dimensional array of pixels each associated with a pixel value, wherein the pixel value is expressed in a binary number;

dividing the line of the Arabic characters into a plurality of line images;

defining a plurality of cells in one of the plurality of line images, wherein each of the plurality of cells comprises a group of adjacent pixels;

serializing pixel values of pixels in each of the plurality of cells in one of the plurality of line images to form a binary cell number;

forming a text feature vector according to binary cell numbers obtained from the plurality of cells in one of the plurality of line images; and

feeding the text feature vector into a Hidden Markov Model to recognize the line of Arabic characters.

2. The method of claim 1, further comprising:

converting the binary cell number into a decimal cell number;

serializing the decimal cell numbers obtained from the plurality of cells in the one of the plurality of line images to form the string of decimal cell numbers; and

forming the text feature vector in accordance to a string of decimal cell numbers obtained from the plurality of cells in the one of the plurality of line images.

4. The method of claim 1, wherein the two-dimensional array of pixels comprises a plurality of rows in a first direction and a plurality of columns in a second direction, wherein the line of Arabic characters is aligned substantially along the first direction, wherein the plurality of line images are sequentially aligned along the first direction.

5. The method of claim 4, wherein the two-dimensional array of pixels comprises N number of rows of pixels, wherein at least one of the plurality of line images has a height defined by M number of rows in the first direction and a width defined by N number of columns in the second direction, wherein M and N are integers.

6. The method of claim 5, wherein N is in a range between 2 and about 100.

7. The method of claim 6, wherein N is in a range between 3 and about 10.

8. The method of claim 1, wherein the pixel values in the two-dimensional array of pixels are expressed in single-bit binary numbers.

9. The method of claim 1, wherein the pixel values in the two-dimensional array of pixels are expressed in multi-bit binary numbers.

10. A method for automatically recognizing Arabic text, comprising:

acquiring a text image containing a line of Arabic characters;

digitizing the line of the Arabic characters to form a two-dimensional array of pixels each associated with a pixel value;

dividing the line of the Arabic characters into a plurality of line images;

downsizing at least one of the plurality of line images to produce a downsized line image;

serializing pixel values of pixels in each column of the downsized line image to form a string of serialized numbers, wherein the string of serialized numbers forms a text feature vector; and

feeding the text feature vector into a Hidden Markov Model to recognize the line of Arabic characters.

11. The method of claim 10, wherein the two-dimensional array of pixels comprises a plurality of rows in a first direction and a plurality of columns in a second direction, wherein the line of Arabic characters is aligned substantially along the first direction, wherein the plurality of line images are sequentially aligned along the first direction.

12. The method of claim 11, wherein the two-dimensional array of pixels comprises N number of rows of pixels, wherein at least one of the plurality of line images has a height defined by M number of rows in the first direction and a width defined by N number of columns in the second direction, wherein M and N are integers.

13. A method for automatically recognizing Arabic text, comprising:

acquiring a text image containing a line of Arabic characters;

digitizing the line of the Arabic characters to form a two-dimensional array of pixels each associated with a pixel value expressed in a binary number, wherein the two-dimensional array of pixels comprises a plurality of rows in a first direction and a plurality of columns in a second direction;

counting frequencies of consecutive pixels of a same pixel value in a column of pixels;

forming a text feature vector using the frequency counts obtained from the column of pixels; and

feeding the text feature vector into a Hidden Markov Model to recognize the line of Arabic characters.

14. The method of claim 13, wherein the frequencies of consecutive pixels of a same pixel value are counted up to a predetermined cut-off transition number.

15. The method of claim 14, wherein the predetermined cut-off transition number is six.

16. The method of claim 13, wherein the pixel values in the two-dimensional array are expressed in single-bit binary numbers.

17. The method of claim 16, wherein the step of counting frequencies comprises starting to count number of pixels having a pixel value of “

1”

in the column of pixels.

18. The method of claim 17, wherein a frequency count is assigned to be “

0”

if a first pixel in the column of pixels has a pixel value of “

0”

.

19. A computer program product comprising a computer useable medium having computer readable program code functions embedded in said medium for causing a computer to:

acquire a text image containing a line of Arabic characters;

digitize the line of the Arabic characters to form a two-dimensional array of pixels each associated with a pixel value, wherein the pixel value is expressed in a binary number;

divide the line of the Arabic characters into a plurality of line images, wherein the two-dimensional array of pixels comprises a plurality of rows in a first direction and a plurality of columns in a second direction, wherein the line of Arabic characters is aligned substantially along the first direction, wherein the plurality of line images are sequentially aligned along the first direction and have a width between 2 and 100 columns of pixels;

define a plurality of cells in one of the plurality of line images, wherein each of the plurality of cells comprises a group of adjacent pixels;

serialize pixel values of pixels in each of the plurality of cells in one of the plurality of line images to form a binary cell number;

form a text feature vector according to binary cell numbers obtained from the plurality of cells in one of the plurality of line images; and

feed the text feature vector into a Hidden Markov Model to recognize the line of Arabic characters.

20. The computer program product of claim 19, wherein the computer readable program code functions embedded in said medium is configured to cause the computer to:

to convert the binary cell number into a decimal cell number;

to serialize the decimal cell numbers obtained from the plurality of cells in the one of the plurality of line images to form the string of decimal cell numbers; and

to form the text feature vector in accordance to a string of decimal cell numbers obtained from the plurality of cells in the one of the plurality of line images.

21. The computer program product of claim 19, wherein the two-dimensional array of pixels comprises N number of rows of pixels, wherein at least one of the plurality of line images has a height defined by M number of rows in the first direction and a width defined by N number of columns in the second direction, wherein M and N are integers.