公开(公告)号：JPH1127672A

公开(公告)日：1999-01-29

申请号：JP14425098

申请日：1998-05-26

Applicant: ST MICROELECTRONICS SRL

Inventor： MANCUSO MASSIMO ,  POLUZZI RINALDO

IPC: G06T9/00 ,  H04N1/41 ,  H04N7/26 ,  H04N7/30 ,  H04N7/24

Abstract: PROBLEM TO BE SOLVED: To reduce image quality deterioration by correcting a pixel value according to both an global feature of a video image block to which the pixel belongs and a local feature of video areas in the vicinity of the pixel so as to attenuate grating noise and stepwise noise. SOLUTION: A global feature GF and a local feature LF of an input video image In evaluated by a feature extract block FE is fed to a fuzzy processing block FUZZY as an input, and a parameter FA proper to decision of a type of filter processing is decided according to a fuzzy criterion. A filter pattern FP decided by a filter compensation block FC is fed to a processing block PROC to which the input video image In is directly supplied and the block PROC applies filter processing of the input video image In based on the filter pattern FP. Thus, low pass filter processing is applied to video areas in the vicinity of grating noise and an edge to reduce both grating noise and stepwise noise.

公开(公告)号：DE60227643D1

公开(公告)日：2008-08-28

申请号：DE60227643

申请日：2002-04-23

Applicant: ST MICROELECTRONICS SRL

Inventor： MESSINA GIUSEPPE ,  BATTIATO SEBASTIANO ,  MANCUSO MASSIMO

IPC: G06T3/40

公开(公告)号：DE69605106T2

公开(公告)日：2000-07-20

申请号：DE69605106

申请日：1996-07-09

Applicant: ST MICROELECTRONICS SRL

Inventor： THOMAS JOSEPH ,  D ALTO VIVIANA ,  MANCUSO MASSIMO

IPC: H04N9/68 ,  H04N9/64

公开(公告)号：DE69605018D1

公开(公告)日：1999-12-09

申请号：DE69605018

申请日：1996-02-13

Applicant: ST MICROELECTRONICS SRL

Inventor： MARQUES PEREIRA RICARDO ALBERT ,  MANCUSO MASSIMO ,  POLUZZI RINALDO

IPC: H04N5/208 ,  G06T5/20 ,  G06T7/60 ,  H03H17/00 ,  H04N5/44 ,  H04N7/01

Abstract: An interpolation filter for video signals comprises: first circuit means (R1-R3,L1-L3,C,7-9) supplied by discrete image elements (p1-p10) for detecting an image edge; second circuit means (10) supplied by the first circuit means for generating a first signal (DDA) corresponding to an average of the discrete image elements (p1-p10) along a direction of the image edge; third circuit means (11,12,13) supplied by the first circuit means for detecting a texture image area wherein an image edge cannot be univocally determined and for generating a second signal (K) depending on a degree of existance of the image edge; fourth circuit means (14), supplied by the first signal (DDA), the second signal (K) and a third signal (Mout), for generating an output signal (SWout) obtained by combining the first signal (DDA) with the third signal (Mout) in a proportion determined by the second signal (K), and multiplexing means (15) controlled by a control signal (CNT) for selectively coupling the third signal (Mout) to fourth signal (A), corresponding to an average of the discrete image elements (p1-p10) along a prescribed direction, or to a fifth signal (PF) corresponding to a previously received image element value.

公开(公告)号：DE60135559D1

公开(公告)日：2008-10-09

申请号：DE60135559

申请日：2001-11-19

Applicant: ST MICROELECTRONICS SRL

Inventor： CASTORINA ALFIO ,  MANCUSO MASSIMO ,  BATTIATO SEBASTIANO

IPC: G06T5/50

Abstract: A method of processing digital source images, each represented by pixel matrices, to obtain from two or more source images, representing one and the same real scene and acquired by utilizing different exposure levels, a final digital image capable of reproducing the real scene with an exposure latitude greater than that of each of the source images. The method, which can be advantageously used in digital still cameras, produces the final image by combining the source images with the help of a weighted mean constructed pixel by pixel. Thanks to a special filtering to which the weighting coefficients are subjected before the weighted mean operation, the method obtains a final image in which the source images are harmoniously combined with each other. A variant of the method can also be applied directly to digital images in CFA (Colour Filter Array) format.

公开(公告)号：DE60130671D1

公开(公告)日：2007-11-08

申请号：DE60130671

申请日：2001-12-24

Applicant: ST MICROELECTRONICS SRL

Inventor： CASTORINA ALFIO ,  MANCUSO MASSIMO ,  BATTIATO SEBASTIANO

IPC: G06T5/00 ,  G06T5/40 ,  H04N1/407 ,  H04N5/235 ,  H04N9/04 ,  H04N101/00

公开(公告)号：DE60203166D1

公开(公告)日：2005-04-14

申请号：DE60203166

申请日：2002-06-12

Applicant: ST MICROELECTRONICS SRL

Inventor： MESSINA GIUSEPPE ,  MANCUSO MASSIMO ,  BATTIATO SEBASTIANO

IPC: G06T3/40 ,  G06T5/00 ,  G06T5/50 ,  H04N3/15 ,  H04N1/40 ,  H04N7/01

Abstract: A method for obtaining a high-resolution digital image from a plurality of starting images P,S formed by pixel matrices and acquired at a lower resolution. The method comprises a phase of combining the plurality of starting images P,S to generate a provisional high-resolution image HR formed by a pixel matrix. The method also comprises a calculation phase to associate a respective error with at least a part of the pixels of the provisional image HR , said calculation phase providing a first error associated with at least one first pixel, and a phase of at least partially processing the provisional image by modifying the pixels of this image on the basis of the respective errors associated with them. The method is characterized by the fact that the calculation phase includes the phase of: calculating a second error to associate with at least one second pixel situated in the vicinity of said first pixel in the matrix (HR ), said second error being calculated by using the first error associated with said at least one first pixel.

公开(公告)号：DE60008716T2

公开(公告)日：2005-02-10

申请号：DE60008716

申请日：2000-07-10

Applicant: ST MICROELECTRONICS SRL

Inventor： BRUNA ARCANGELO ,  MANCUSO MASSIMO ,  GALLUZZO AGOSTINO

IPC: G06T9/00 ,  H04N19/126 ,  H04N19/14 ,  H04N19/149 ,  H04N19/18 ,  H04N19/186 ,  H04N19/196 ,  H04N19/60 ,  H04N7/30

Abstract: A method (400) of compressing a digital image including a matrix of elements each one consisting of a plurality of digital components of different type representing a pixel, the method comprising the steps of splitting (455) the digital image into a plurality of blocks and calculating, for each block, a group of DCT coefficients for the components of each type, and quantizing (460-465) the DCT coefficients of each block using a corresponding quantization table scaled by a gain factor for achieving a target compression factor; the method further comprises the steps of determining (440) at least one energy measure of the digital image, and estimating (441-450) the gain factor as a function of the at least one energy measure, the function being determined experimentally according to the target compression factor.

公开(公告)号：DE69530566T2

公开(公告)日：2004-03-25

申请号：DE69530566

申请日：1995-01-13

Applicant: ST MICROELECTRONICS SRL

Inventor： MANCUSO MASSIMO ,  MORETTI PAOLO ,  POLUZZI RINALDO ,  RIZZOTTO GIANGUIDO

IPC: G06T7/60 ,  G06K9/46

Abstract: A fuzzy method for the recognition of geometric shapes in images, the particularity whereof is that it comprises the following steps: recognition of at least one edge of a curve of an image; recognition of the gradient of the edge; correction of the gradient by virtue of first fuzzy means; and determination of the number of points that belong to the curve by virtue of second fuzzy means. The process according to the present invention is performed by a fuzzy device for the recognition of geometric shapes in images that comprises: an edge detector, adapted to recognize at least one edge of a curve of an image and adapted to recognize the gradient of the edge; a fuzzy gradient corrector, adapted to correct said gradient; a fuzzy vote attributor, adapted to determine the number of points that belong to the curve.

公开(公告)号：DE69720192T2

公开(公告)日：2003-12-04

申请号：DE69720192

申请日：1997-05-30

Applicant: ST MICROELECTRONICS SRL

Inventor： MANCUSO MASSIMO ,  POLUZZI RINALDO

IPC: G06T9/00 ,  H04N1/41 ,  H04N7/26 ,  H04N7/30

Abstract: A post-processing method for reducing artifacts in block-coded digital images, characterized by providing for: dividing an input image (In) into a plurality of image blocks (IB); for each image block (IB), estimating global features (GF) of said image block providing information on an average content of image edges along the horizontal and vertical directions of said image block (IB); for each pixel (Px) of an image block (IB) under examination, estimating local features (LF) for said pixel (Px) providing information on the content of image edges along the horizontal and vertical directions of an image area near said pixel (Px); modifying the value of said pixel (Px) according to both said global features (GF) of the image block (IB) to which said pixel (Px) belongs and said local features (LF) of the image area near said pixel (Px).