IPC분류정보
국가/구분 |
United States(US) Patent
등록
|
국제특허분류(IPC7판) |
|
출원번호 |
US-0737513
(2000-12-15)
|
발명자
/ 주소 |
- Fan, Zhigang
- Balasubramanian, Thyagarajan
|
출원인 / 주소 |
|
대리인 / 주소 |
Fay, Sharpe, Fagan, Minnich & McKee, LLP
|
인용정보 |
피인용 횟수 :
35 인용 특허 :
9 |
초록
▼
A document processing system and a method for classifying an input image or region thereof as either a synthetic graphic or a natural picture, is disclosed. The system includes an image input subsystem, a processing subsystem for processing image data provided by the image input subsystem, and softw
A document processing system and a method for classifying an input image or region thereof as either a synthetic graphic or a natural picture, is disclosed. The system includes an image input subsystem, a processing subsystem for processing image data provided by the image input subsystem, and software/firmware means operative on the processing subsystem for a) low-pass filtering image data representative of the input image or region thereof to produce low-pass filtered pixel values; b) determining a smoothness value for each of a plurality of low-pass filtered pixel values; c) generating histogram data from the smoothness values; d) determining a texture metric for the input image or region thereof from a subset of the histogram data; and e) thresholding the texture metric to classify the input image as either a synthetic graphic or a natural picture.
대표청구항
▼
1. A method for classifying an input image or region thereof as either a synthetic graphic or a natural picture, the method comprising:a) low-pass filtering image data representative of the input image or region thereof to produce low-pass filtered pixel values;b) determining a smoothness value for
1. A method for classifying an input image or region thereof as either a synthetic graphic or a natural picture, the method comprising:a) low-pass filtering image data representative of the input image or region thereof to produce low-pass filtered pixel values;b) determining a smoothness value for each of a plurality of low-pass filtered pixel values;c) generating histogram data from the smoothness values;d) determining a texture metric for the input image or region thereof from a subset of the histogram data; ande) thresholding the texture metric to classify the input image as either a synthetic graphic or a natural picture. 2. The method of claim 1, wherein step b) includes determining a smoothness value for each low-pass filtered pixel value below a predetermined threshold. 3. The method of claim 1, wherein step e) includes classifying the input image as a natural image when the texture metric is above a predetermined threshold, and classifying the input image as a synthetic image when the texture metric is below the threshold. 4. The method of claim 1, further including:f) executing a downstream image processing operation on the image data based on the resulting image classification. 5. The method of claim 1, wherein step a) includes low-pass filtering image data representative of a luminance component of the input image or region thereof to produce low-pass filtered pixel values. 6. A method for classifying an input image or region thereof as one of two different types, the method comprising:a) low-pass filtering image data representative of the input image or region thereof to produce low-pass filtered pixel values;b) determining a smoothness value for each of a plurality of low-pass filtered pixel values, wherein each smoothness value t(m,n) is determined from: t ( m,n )=| P lpf ( m,n )−([ P lpf ( m+d,n )+ P lpf ( m−d,n )+ P lpf ( m,n+d )+ P lpf ( m,n−d )]/4)|;c) generating histogram data from the smoothness values;d) determining a texture metric for the input image or region thereof from at least a portion of the histogram data; ande) thresholding the texture metric to classify the input image or region thereof as one of the two different types. 7. A method for classifying an input image or region thereof, the method comprising:a) low-pass filtering image data representative of the input image or region thereof to produce low-pass filtered pixel values;b) determining a smoothness value for each of a plurality of low-pass filtered pixel values, wherein each smoothness value is a measure of an absolute difference between a low-pass filtered pixel value and an average of a plurality of other pixel values proximate the low-pass filtered pixel value; and,c) classifying the input image or region thereof based upon at least a portion of the determined smoothness values. 8. A method for classifying an input image or region thereof as one of two different types, the method comprising:a) low-pass filtering image data representative of the input image or region thereof to produce low-pass filtered pixel values;b) determining a smoothness value for each of a plurality of low-pass filtered pixel values;c) generating histogram data from the smoothness values;d) determining a texture metric for the input image or region thereof from at least a portion of the histogram data; wherein the textue metric (T) is determined from: T=Σt 2 ( m,n )/( N−M ); ande) thresholding the texture metric to classify the input image or region thereof as one of the two different types. 9. A method for classifying an input image or region thereof, the method comprising:a) low-pass filtering image data representative of the input image or region thereof to produce low-pass filtered pixel values;b) determining a smoothness value for each of a plurality of low-pass filtered pixel values;c) generating histogram data from the smoothness values;d) determining a texture metric for the input image or re gion thereof from a subject of the histogram data excluding histogram data associated with edge pixels from the texture metric determination; ande) thresholding the texture metric to classify the input image or region thereof. 10. A document processing system for classifying an input image or region thereof as either a synthetic graphic or a natural picture, the system comprising:an image input subsystem;a processing subsystem for processing image data provided by the image input subsystem; andsoftware/firmware means operative on the processing subsystem for:a) low-pass filtering image data representative of the input image or region thereof to produce low-pass filtered pixel values;b) determining a smoothness value for each of a plurality of low-pass filtered pixel values;c) generating histogram data from the smoothness values;d) determining a texture metric for the input image or region thereof from a subset of the histogram data; ande) thresholding the texture metric to classify the input image as either a synthetic graphic or a natural picture. 11. The system of claim 10, wherein b) includes determining a smoothness value for each low-pass filtered pixel value below a predetermined threshold. 12. The system of claim 10, wherein e) includes classifying the input image as a natural image when the texture metric is above a predetermined threshold, and classifying the input image as a synthetic image when the texture metric is below the threshold. 13. The system of claim 12, wherein the predetermined threshold has a value of about 30. 14. The system of claim 10, wherein the software/firmware means is further operative on the processing subsystem for:f) executing a downstream image processing operation on the image data based on the resulting image classification. 15. The system of claim 10, wherein a) includes low-pass filtering image data representative of a luminance component of the input image or region thereof to produce low-pass filtered pixel values. 16. The system of claim 10, wherein the document processing system is a xerographic document processing system. 17. A system for classifying an input image or region thereof as one of two types, the system comprising:an image input subsystem;a processing subsystem for processing image data provided by the image input subsystem; andsoftware/firmware means operative on the processing subsystem for:a) low-pass filtering image data representative of the input image or region thereof to produce low-pass filtered pixel values;b) determining a smoothness value for each of a plurality of low-pass filtered pixel values, wherein each smoothness value t(m,n) is determined from: t ( m,n )=| P lpf ( m,n )−([ P lpf ( m+d,n )+ P lpf ( m−d,n )+ P lpf ( m,n+d )+ P lpf ( m,n−d )]/4)|;c) generating histogram data from the smoothness values;d) determining a texture metric for the input image or region thereof from at least a portion of the histogram data; ande) thresholding the texture metric to classify the input image or region thereof as one of the two different types. 18. A system for classifying an input image or region thereof, the system comprising:an image input subsystem;a processing subsystem for processing image data provided by the image input subsystem; andsoftware/firmware means operative on the processing subsystem for:a) low-pass filtering image data representative of the input image or region thereof to produce low-pass filtered pixel values;b) determining a smoothness value for each of a plurality of low-pass filtered pixel values, wherein each smoothness value is a measure of an absolute difference between a low-pass filtered pixel value and an average of a plurality of other pixel values proximate the low-pass filtered pixel value; and,c) classifying the input image or region thereof based upon at least a portion of the determined smoothness values. 19. A system for classifying an input image or region thereof as one of the two dif ferent types, the system comprising:an image input subsystem;a processing subsystem for processing image data provided by the image input subsystem; andsoftware/firmware means operative on the processing subsystem for:a) low-pass filtering image data representative of the input image or region thereof to produce low-pass filtered pixel values;b) determining a smoothness value for each of a plurality of low-pass filtered pixel values;c) generating histogram data from the smoothness values;d) determining a texture metric for the input image or region thereof from at least a portion of the histogram data, wherein the texture metric (T) is determined from: T=Σt 2 ( m,n )/( N−M ); ande) thresholding the texture metric to classify the input image or region thereof as one of the two different types. 20. A system for classisfying an input image or region thereof, the system comprising:an image input subsystem;a processing subsystem for processing image data provided by the image input subsystem; andsoftware/firmware means operative on the processing subsystem for:a) low-pass filtering image data representative of the input image or region thereof to produce low-pass filtered pixel values;b) determining a smoothness value for each of a plurality of low-pass filtered pixel values;c) generating histogram data from the smoothness values;d) determining a texture metric for the input image or region thereof from a subset of the histogram data excluding histogram data associated with edge pixels from the texture metric determination; ande) thresholding the texture metric to classify the input image or region thereof.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.