System and method for characterizing handwritten or typed words in a document
View Patent ↗A method of characterizing a word image includes traversing the word image stepwise with a window to provide a plurality of window images. For each of the plurality of window images, the method includes splitting the window image to provide a plurality of cells. A feature, such as a gradient direction histogram, is extracted from each of the plurality of cells. The word image can then be characterized based on the features extracted from the plurality of window images.
1. A method of characterizing a word image comprising:
traversing the word image stepwise with a window to provide a plurality of window images;
for each of the plurality of window images:
splitting the window image to provide a plurality of cells, wherein when only a portion of the window image contains active pixels, the splitting comprises splitting only a rectangular area in the window image containing the active pixels into an array of cells, whereby a portion of the window image containing no active pixels is excluded from the splitting;
extracting a feature from each of the plurality of cells, the feature comprising a gradient orientation histogram; and
characterizing the word image based on the features extracted from the plurality of window images; and
wherein the method is performed using a computer or processor.
2. The method of claim 1 , further comprising:
for each window image, determining a features vector based on the extracted features of each of the plurality of cells; and
characterizing the word image based on the features vectors of the plurality of window images.
3. The method of claim 2 , wherein the computing of the features vector comprises concatenating the extracted features.
4. The method of claim 1 , wherein the array comprises an M×N array where N is a number of cells arranged in a horizontal direction and M is a number of cells arranged in a vertical direction.
5. The method of claim 4 , wherein M is at least 2 and N is at least two.
6. The method of claim 1 wherein the cells are rectangular.
7. The method of claim 1 , wherein the cells of a window image are of the same size.
8. The method of claim 1 , wherein the extraction of a feature comprises computing a gradient direction histogram for the pixels in the cell.
9. The method of claim 1 , wherein the word image comprises a bitmap acquired by segmenting a document image.
10. The method of claim 1 , wherein the word image is characterized without characterization of individual characters of a character string within the word image.
11. A computer program product embodied on a non-transitory recording medium encoding instructions, which when executed on a computer causes the computer to perform the method of claim 1 .
12. A processing system which executes instructions stored in memory for performing the method of claim 1 .
13. A method of characterizing a word image comprising:
traversing the word image stepwise with a window to provide a plurality of window images;
for each of the plurality of window images:
splitting the window image to provide a plurality of cells;
extracting a feature from each of the plurality of cells, comprising computing a gradient direction histogram for the pixels in the cell, comprising computing a gradient magnitude m and direction θ for each pixel with coordinates (x,y) in the cell as:
m
(
x
,
y
)
=
G
x
2
+
G
y
2
θ
(
x
,
y
)
=
a
tan
2
(
G
y
G
x
)
,
where atan2 is a function that gives the angle of the vector (G x , G y ) in the range [−π,π]; and
characterizing the word image based on the features extracted from the plurality of window images.
14. The method of claim 13 , wherein the splitting the window image comprises splitting at least a portion of the window image into an array of cells.
15. The method of claim 14 , wherein the portion of the window image that is split into cells bounds at least the active pixels in the window image.
16. A computer program product embodied on a non-transitory recording medium for encoding instructions, which when executed on a computer causes the computer to perform the method of claim 13 .
17. A processing system which executes instructions stored in memory for performing the method of claim 13 .
18. A method of characterizing a word image comprising:
traversing the word image stepwise with a window to provide a plurality of window images;
for each of the plurality of window images:
splitting the window image to provide a plurality of cells;
extracting a feature from each of the plurality of cells, the feature comprising a gradient orientation histogram; and
for each window image, computing a features vector based on the extracted features of each of the plurality of cells, wherein the computing of the features vector comprises concatenating the extracted features and normalizing the features vector such that all of the component values sum to a fixed value;
characterizing the word image based on the features vectors of the plurality of window images; and
wherein the method is performed using a computer or processor.
19. A computer program product, embodied on a non-transitory recording medium, encoding instructions, which when executed on a computer causes the computer to perform the method of claim 18 .
20. A processing system which executes instructions stored in memory for performing the method of claim 18 .
21. A method of characterizing a document image comprising:
segmenting the document image to identify word images;
for an identified word image, traversing the word image stepwise with a window to provide a plurality of window images;
for each of the plurality of window images:
splitting the window image to provide a plurality of cells;
extracting a feature from each of the plurality of cells comprising computing a gradient direction histogram for the pixels in the cell; and
computing a features vector based on the extracted features;
normalizing the features vector such that all of the component values sum to a fixed value;
characterizing the word image based on the features vectors of the plurality of window images; and
characterizing the document based on the characterization of at least one of the identified word images;
wherein the method is performed using a computer or processor.
22. The method of claim 21 , further comprising classifying the characterized word image with a classifier trained to identify a keyword.
23. The method of claim 22 , wherein the document image is an image of a handwritten document.
24. A computer program product, embodied on a non-transitory recording medium, encoding instructions, which when executed on a computer causes the computer to perform the method of claim 21 .
25. A processing system which executes instructions stored in memory for performing the method of claim 21 .
26. A processing system comprising:
a document segmentor which processes an input document image to identify word images;
a features extractor which extracts features of an identified word image and computes features vectors therefrom, the features extractor executing instructions for traversing the word image stepwise with a window to provide a plurality of window images and, for each of the plurality of window images,
excluding from splitting, a region of the window image having no active pixels, splitting a remaining portion of the window image comprising active pixels to provide an M×N array of cells where N is a number of cells arranged in a horizontal direction and M is a number of cells arranged in a vertical direction, and where N and M each have the same value for each of the window images,
extracting a feature from each of the plurality of cells, and
computing a features vector based on the extracted features; and
a classifier which classifies the word image based on the computed features vectors of the window images.
27. The processing system of claim 26 , wherein the classifier comprises a hidden Markov model trained to identify at least one keyword.