Image enhancement system

• US 5,218,649 A
• Filed: 05/04/1990
• Issued: 06/08/1993
• Est. Priority Date: 05/04/1990
• Status: Expired due to Fees

• Current Assignee: Qwest Communications International Incorporated (Lumen Technologies, Inc.)
• Original Assignee: U.S. West Advanced Technologies, Inc.
• Inventors: Kundu, Amlan, Ramamoorthy, V.
• Primary Examiner(s): Razavi, Michael T.
• Assistant Examiner(s): COUSO, YON JUNG
• Application Number:
  US07/519,128
• Time in Patent Office: 1,131 Days
• Field of Search: 382/54, 382/52, 382/22, 358/166, 358/167
• US Class Current: 382/180
• CPC Class Codes: G06T 5/20 using local operators
• CPC Class Codes: H04N 19/80 Details of filtering operations specially adapted for video compression, e.g. for pixel interpolation H04N19/635, H04N19/86 take precedence

Abstract

A system for enhancing the quality of an image that suffers from staircase effects, contouring effects, and granular noise, without affecting the existing method of creating the image is disclosed. The enhancement system identifies the edge and nonedge regions in the decoded image, and then filters those regions in order to reduce the specific type of quality degradation that generally occurs in the specific type of region. The system is integrated into a processing system including means for receiving and displaying digital image data. The system receives the digital image data, differentiates between edge data and the nonedge data within the digital image data, and filters the image data accordingly. The filtered edge data and the filtered nonedge data are combined to form an enhanced image. A multi-pass filtering system is also disclosed.

Claims

1. A method for digital image enhancement using a processing system including means for receiving and displaying image comprised of decoded pixel digital data, the data defining edge and nonedge regions in the image, the method comprising the steps of:

(a) receiving a digital image;

(b) differentiating between edge pixel data and nonedge pixel data in said digital image;

(c) filtering each edge pixel by identifying a first edge image region, said first edge image region including said edge pixel, and determining a filtered edge pixel value based on pixel data in said first edge image region, such that the staircase effects appearing when said filtered edge pixels are displayed are decreased;

(d) filtering each nonedge pixel by identifying a first nonedge image region, said fist nonedge image region including said nonedge pixel, and determining a filtered nonedge pixel value based on pixel data in said first nonedge image region, such that the loss of detail and contouring effects appearing when said filtered nonedge pixels are displayed are decreased; and

(e) combining said filtered edge pixel data and said filtered nonedge pixel data to form an enhanced image.

2. A method as claimed in claim 1, wherein said method further includes the steps of:

refiltering each edge

(a) pixel by identifying a second edge image region, said second edge image region including said edge pixel, and determining a refiltered edge pixel value based on pixel data in said second edge image region, such that the staircase effects appearing in said edge region data when said refiltered edge pixels are displayed are further decreased;

(b) filtering each nonedge pixel by identifying a second nonedge image region, said second nonedge image region including said nonedge pixel, and determining a refiltered nonedge pixel value based on pixel data in said second nonedge image region, such that staircase effects appearing in said nonedge region data when said refiltered nonedge pixels are displayed are further decreased; and

(c) combining said refiltered edge region pixel data and said refiltered nonedge region pixel data to form a second enhanced image.

3. A method as claimed in claim 1, wherein prior to differentiating between the edge and nonedge regions, a subimage of said digital image is formed, such that said digital image is comprised of said subimage and an unfiltered portion of said digital image, said edge and nonedge data of said subimage are differentiated, filtered, and combined to form an enhanced subimage, and said enhanced image is formed by combining said enhanced subimage with said unfiltered portion of said digital image.

4. A method as claimed in claim 1 or 2, wherein said step of differentiating between edge and nonedge regions is carried out by analyzing said digital image via an edge detector.

5. A method as claimed in claim 1, further comprising the step of adding random noise to said nonedge data after said step of filtering said nonedge data.

6. A method as claimed in claim 1 or 2, wherein said step of determining a filtered edge pixel value based on pixel data in an edge image region includes the steps of ordering said pixel data in said edge image region, and selecting a median value from said ordered pixels as said filtered edge pixel value.

7. A method as claimed in claim 1 or 2, wherein said step of determining a filtered nonedge pixel value based on said pixel data in a nonedge image region includes the steps of approximating a mean value for said pixel data in said nonedge image region via a statistical D-filtering process, and selecting said approximated mean value as said filtered nonedge pixel value.

8. A method as claimed in claim 1 or 2, wherein said step of differentiation requires edge identification threshold parameters, and said edge image regions and said nonedge image regions are defined by window parameters, and

said method further includes the step of receiving values for said threshold parameters and said window parameters prior to the steps of differentiating and filtering.

9. An apparatus for enhancing a digital image, said digital image comprised of decoded digital data, the decoded digital data defining edge and nonedge regions in the image, said apparatus comprising:

(a) means for receiving a digital image;

(b) edge detection means for differentiating between the edge pixel data and the nonedge pixel data in said digital image;

(c) nonlinear edge filtering means for filtering each said edge pixel data by identifying a first edge image region including a plurality of pixels such that the staircase effects in said filtered edge pixel data are decreased;

(d) nonlinear nonedge filtering means for filtering each said nonedge pixel data by identifying a first edge image region including a plurality of pixels such that the loss of detail and contouring effects in said filtered nonedge pixel data is decreased; and

(e) combination means for combining said filtering edge pixel data and said filtered nonedge pixel data to produce an enhanced digital image.

10. An apparatus as claimed in claim 9, wherein said nonlinear edge filtering means is a median filter.

11. An apparatus as claimed in claim 9, wherein said nonlinear nonedge filtering means is a statistical D-filter.