Systems and methods for content-based document image enhancement
First Claim
1. An image processing system, the system comprising a system processor that is programmed or adapted to:
- a) receive image data comprising a plurality pixels each having a pixel value;
b) selectively mark a current pixel as a potential text element;
c) determine a distance from the current pixel to a previous pixel marked as a potential text element; and
d) selectively mark all pixels in between the current and previous pixels as text pixels or image pixels based upon the determined distance.
4 Assignments
0 Petitions
Accused Products
Abstract
A system can include a system processor that is programmed or adapted to classify pixels in a document image, and in some instances, to enhance the document image based upon such classification. Image data including pixels is received. For each pixel in some subset of the image data, a feature vector is generated. Each pixel in the subset is classified as text or image based upon its associated feature vector. The classifications can be outputted for storage and/or further processing. Further processing can include pixel level enhancement processing. Such enhancement processing can include application of a sharpening filter to pixels classified as text and of a smoothing filter to pixels classified as image. In some instances, background suppression can be performed upon a subset of the image data prior to classification.
-
Citations
43 Claims
-
1. An image processing system, the system comprising a system processor that is programmed or adapted to:
-
a) receive image data comprising a plurality pixels each having a pixel value;
b) selectively mark a current pixel as a potential text element;
c) determine a distance from the current pixel to a previous pixel marked as a potential text element; and
d) selectively mark all pixels in between the current and previous pixels as text pixels or image pixels based upon the determined distance. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35)
-
-
36. An image processing method, the method comprising the steps of:
-
a) receiving image data comprising a plurality pixels each having a gray level value;
b) generating a feature vector for each pixel, wherein each feature vector comprises a vertical edge parameter equal to |I(x, y+1)−
I(x, y−
1)| and a horizontal edge parameter equal to |I(x+1, y)−
I(x−
1, y)|, wherein I(x, y) corresponds to the gray level value of the pixel addressed by coordinates x and y;
c) classifying each pixel as text or image by;
i) calculating a norm η
of each generated feature vector χ
, wherein
and wherein {xi|i=1,2, . . . , d} are elements of χ
.ii) for each feature vector norm η
greater than a selected threshold,1) marking the pixel associated with the respective feature vector as a potential text element; and
2) if the spatial distance between two consecutive potential text element pixels is below a specified distance threshold, marking all pixels in between the two consecutive potential text element pixels as text pixels; and
if not, marking all pixels between the two consecutive potential text element pixels as image pixels; and
iii) for each feature vector norm η
not greater than the selected threshold, marking the pixel associated with the respective feature vector as an image pixel; and
d) applying a filter to one or more pixels based upon each pixel'"'"'s classification as text or image. - View Dependent Claims (37, 38, 39, 40, 41)
-
-
42. An image processing system, the system comprising:
-
a) imaging means for scanning a source document to produce image data comprising a plurality of pixel values each having a gray level value;
b) receiving means for receiving the image data from the imaging means;
c) background suppression means for suppressing background in the image data;
d) feature vector calculating means for generating a feature vector for each pixel, wherein each feature vector comprises a vertical edge parameter equal to |(x,y+1)−
I(x,y−
1)| and a horizontal edge parameter equal to |I(x+1, y)−
I(x−
1, y)|, wherein I(x, y) corresponds to the gray level value of the pixel addressed by coordinates x and y;
e) classification means for classifying each pixel of the image data as text or image by;
i) calculating a norm η
of each generated feature vector χ
, wherein
and wherein {xi|i=1,2, . . . , d} are elements of χ
.ii) for each feature vector norm 17 greater than a selected threshold, 1) marking the pixel associated with the respective feature vector as a potential text element; and
2) if the spatial distance between two consecutive potential text element pixels is below a specified distance threshold, marking all pixels in between the two consecutive potential text element pixels as text pixels; and
if not, marking all pixels between the two consecutive potential text element pixels as image pixels; and
iii) for each feature vector norm η
not greater than the selected threshold, marking the pixel associated with the respective feature vector as an image pixel; and
f) filtering means for applying a filter, K, to classified pixels, wherein for each pixel classified as text and wherein for each pixel classified as image.
-
-
43. An image processing system for suppressing background in image data, the image processing system comprising a system processor that:
-
a) receives image data comprising a plurality pixels each having a gray level value;
b) generating a histogram distribution of frequency of occurrence of gray values from the first ten scan lines of the image data;
c) determines a background peak gray value for the image data based upon the generated histogram distribution;
d) for each pixel value below the determined peak gray value less two standard deviations, replaces its pixel value with a new pixel value by mapping its current pixel value into a range of pixel values corresponding to a full gray scale spectrum by multiplying the current pixel value by a pixel value corresponding to white and dividing by the determined peak gray value less two standard deviations; and
e) for each pixel value not below the determined peak gray value less two standard deviations, replacing its pixel value with a new pixel value corresponding to white.
-
Specification