-
-
公开(公告)号：ES2146599T3
公开(公告)日：2000-08-16
申请号：ES93304413
申请日：1993-06-07
Applicant: IBM
Inventor： EDGAR ALBERT DURR , PENN STEVEN CRAIG
Abstract: A method, system and process for improving image enhancement tools through the application of expert systems. The improvements are accomplished by the operation of a process in the memory of a processor that enhances an image based on a set of criteria entered by a user. The image is enhanced in at least two ways and the two enhanced images (510, 520) are presented to the user. The user selects the image that appears best and the selected image (530) is thereafter used for subsequent image enhancements according to the user criteria until a final choice of images (550, 560) is presented.
-
公开(公告)号：DE69328595D1
公开(公告)日：2000-06-15
申请号：DE69328595
申请日：1993-06-07
Applicant: IBM
Inventor： EDGAR ALBERT DURR , PENN STEVEN CRAIG
Abstract: A method, system and process for improving image enhancement tools through the application of expert systems. The improvements are accomplished by the operation of a process in the memory of a processor that enhances an image based on a set of criteria entered by a user. The image is enhanced in at least two ways and the two enhanced images (510, 520) are presented to the user. The user selects the image that appears best and the selected image (530) is thereafter used for subsequent image enhancements according to the user criteria until a final choice of images (550, 560) is presented.
-
公开(公告)号：SG43723A1
公开(公告)日：1997-11-14
申请号：SG1996000208
申请日：1991-08-29
Applicant: IBM
Inventor： EDGAR ALBERT DURR
Abstract: Analog video data input is received by a video display adapter of a digital video capture system. A method is disclosed for retaining most of the resolution of the captured full interlaced image while eliminating the image shake of fast moving components. In a particular embodiment, a first and second sequential field comprising a full frame of video data are captured and digitised (62). These fields are then processed (64,66,68) to determine areas corrupted by fast movement of components of the image which were captured. In the areas of detected fast movement, an estimate for pels is substituted (72) in one field based upon the other field. Resolution loss is thereby confined to small areas which, due to rapid motion, were not well defined prior to processing.
-
公开(公告)号：DE69218494D1
公开(公告)日：1997-04-30
申请号：DE69218494
申请日：1992-05-19
Applicant: IBM
Inventor： EDGAR ALBERT DURR
Abstract: A method for processing signals to diffuse quantization error spatially or temporally. Positive feedback of the quantization error is used to reduce the relative noise for the low frequencies of the spectrum. According to a preferred implementation in a two-dimensional spatial application involving a base pixel and three adjacent orthogonally arranged pixels, error attributable to quantization at the base pixel location is provided as positive feedback for the quantization of the diagonally adjacent pixel. Offsetting and complementary values of the quantization error are also diffused to the remaining two diagonally positioned pixels. The composite of the four pixels forms a square with a zero net error diffusion along the horizontal, vertical and diagonal orientations.
-
Patent Agency Ranking
公开(公告)号：DE69305807D1
公开(公告)日：1996-12-12
申请号：DE69305807
申请日：1993-07-23
Applicant: IBM
Inventor： EDGAR ALBERT DURR
-
公开(公告)号：DE69019068T2
公开(公告)日：1995-12-21
申请号：DE69019068
申请日：1990-06-07
Applicant: IBM
Inventor： DEACON JOHN JOSEPH , EDGAR ALBERT DURR
-
公开(公告)号：HK139394A
公开(公告)日：1994-12-16
申请号：HK139394
申请日：1994-12-08
Applicant: IBM
Inventor： EDGAR ALBERT DURR , LUBART NEIL DAVID
-
公开(公告)号：ZA918467B
公开(公告)日：1992-07-29
申请号：ZA918467
申请日：1991-10-23
Applicant: IBM
Inventor： EDGAR ALBERT DURR , ALBERT DURR EDGAR
IPC: H04N5/78 , G06T5/00 , H04N1/38 , H04N5/14 , H04N5/44 , H04N5/91 , H04N5/92 , H04N7/18 , H04N , G06F
Abstract: Analog video data input is received by a video display adapter of a digital video capture system. A method is disclosed for retaining most of the resolution of the captured full interlaced image while eliminating the image shake of fast moving components. In a particular embodiment, a first and second sequential field comprising a full frame of video data are captured and digitised (62). These fields are then processed (64,66,68) to determine areas corrupted by fast movement of components of the image which were captured. In the areas of detected fast movement, an estimate for pels is substituted (72) in one field based upon the other field. Resolution loss is thereby confined to small areas which, due to rapid motion, were not well defined prior to processing.
-
公开(公告)号：DE3278955D1
公开(公告)日：1988-09-29
申请号：DE3278955
申请日：1982-02-17
Applicant: IBM
Inventor： BOHRER KATHRYN ANN , EDGAR ALBERT DURR
Abstract: Method of character generation for displaying or printing image data stored sequentially in a compressed font memory. The proper character is addressed during decompression by combining a calculated approximate pointer (25) obtained by multiplying the code of the character by an approximate pel dimension number, with a limited range correction factor (22, 26) stored with the font to obtain an exact start of character pointer utilized to address the image data for the character. Also stored with the correction factor is data (23) indicating the vertical positioning of the character within the character box (10).
-
-
-
-
-
-
-
-
-
Search Results
42
records
(took 0.017 seconds)
