-
公开(公告)号:JP2003153290A
公开(公告)日:2003-05-23
申请号:JP2002256394
申请日:2002-09-02
Applicant: ST MICROELECTRONICS SRL
Inventor: BOSCO ANGELO , MANCUSO MASSIMO
Abstract: PROBLEM TO BE SOLVED: To provide a method for preserving picture detail and applying filter processing to noise from complicated image files. SOLUTION: A filter is provided for a digital camera includes image sensors sensitive to light, a color filter placed over sensitive elements of the sensors and patterned according to a Bayer mosaic pattern layout, an interpolation algorithm joining together the digital information provided by differently colored adjacent pixels in the Bayer pattern. The filter (10) is adaptive and includes a noise level computation block (26) thus removing noise while simultaneously preserving picture detail.
-
公开(公告)号:JPH07298101A
公开(公告)日:1995-11-10
申请号:JP29325394
申请日:1994-11-28
Applicant: ST MICROELECTRONICS SRL , CONS RIC MICROELETTRONICA
Inventor: MANCUSO MASSIMO , POLUZZI RINALDO , RIZZOTTO GIANGUIDO
Abstract: PURPOSE: To attain noise reduction through a filter for filtering only the noise- containing signal sections in digital video signals. CONSTITUTION: Each circuit means is operated by a fuzzy logic and provided with inference circuits (look-up tables) C1 and C2 and digital subtracters S1-S4. Each subtracter has an input terminal and is provided for receiving an individual digital video signal for reporting information concerning the pixel to be processed. The number of circuit means included in a filter 1 depends on the size of an image to be processed, and secondly, depends on the number of directions under the operation of a line operator. While using fuzzy type logic rules, a noise calculation circuit 4 compares trigger levels determined in the circuits C1 and C2 with such a trigger level. Thus, noise contained in the digital video signal are reduced effectively.
-
公开(公告)号:JPH11312239A
公开(公告)日:1999-11-09
申请号:JP37282398
申请日:1998-12-28
Applicant: ST MICROELECTRONICS SRL
Inventor: COLLA FEDERICA , MANCUSO MASSIMO , POLUZZI RINALDO
Abstract: PROBLEM TO BE SOLVED: To provide a digital image color correcting device for correcting facial tone image parts of a digital video image by adopting fuzzy logic. SOLUTION: This device includes a pixel fuzzy fine unit 1 for receiving a pixel stream belonging to a sequence of correlated frames of a digital video image and for calculating a multilevel value representing the membership of each pixel with respect to a skin color class, a global parameter estimator 2 for receiving each pixel and a relative membership value and for calculating first and second parameters regulating the feature of an image belonging to the skin color class, a processing unit 3 for correcting each of the pixels of the part of the image belonging to the skin color class according to the first global parameter, and a processing switch 4 for outputting pixels or corrected pixels in accordance with the second global parameter.
-
公开(公告)号:JPH07240858A
公开(公告)日:1995-09-12
申请号:JP23539494
申请日:1994-09-29
Applicant: ST MICROELECTRONICS SRL , CONS RIC MICROELETTRONICA
Inventor: MANCUSO MASSIMO , POLUZZI RINALDO , RIZZOTTO GIANGUIDO
Abstract: PURPOSE: To provide a filter architecture optimizable and easily executable by using fuzzy logic. CONSTITUTION: This filter is provided with a filter circuit 2 for separating high band components and low band components from input signals for video images, a brilliance estimating circuit 6 for obtaining the mean brilliance value of respective plural sections for which the video images are divided, first and second video characteristic adjustment circuits 4 and 5 for changing the high band components and the low band components in response to the mean brilliance value and a totaling circuit 36 for combining the changed high band components and low band components and generating filtered video signals.
-
公开(公告)号:JP2003199119A
公开(公告)日:2003-07-11
申请号:JP2002323033
申请日:2002-11-06
Applicant: ST MICROELECTRONICS SRL , SGS THOMSON MICROELECTRONICS
Inventor: GUARNERA MIRKO , MANCUSO MASSIMO , HURWITZ JONATHAN , SMITH STEWART , FINDLATER KEITH
Abstract: PROBLEM TO BE SOLVED: To reduce computing time and energy consumption in a method of processing digital images in devices for acquiring both individual images and image sequences. SOLUTION: The method of processing digital images comprises the steps of: acquiring images in CFA (Color Filter Array) format; and reducing the resolution of the images acquired. The resolution-reduction step processes the images directly in the CFA format. COPYRIGHT: (C)2003,JPO
-
公开(公告)号:JPH07298301A
公开(公告)日:1995-11-10
申请号:JP10261595
申请日:1995-04-26
Applicant: CONS RIC MICROELETTRONICA , ST MICROELECTRONICS SRL
Inventor: MANCUSO MASSIMO , POLUZZI RINALDO , RIZZOTTO GIANGUIDO
Abstract: PURPOSE: To provide high resolution of video and to eliminate the defects of conventional techniques by providing a calculation block which is connected to an input terminal and operated by a fuzzy-logic based processing method and which executes a switch between at least two different interpolation processing methods. CONSTITUTION: For a TV signal scanning converting device 1, luminance/ chrominance signals CFy and CFuv and signals PFy and PFuv corresponding to image fields are inputted to an interface(IF) 2. The IF 2 is operated with a fuzzy logic and supplies TV signals components Xy and Piy to input terminals Pi and X of a filtering block 3. A calculation block CALC1 is operated with a fuzzy logic processing method, receives a luminance component Y and detects the front and motion of a TV signal, while utilizing information in two continuous image fields. On the other hand, an arithmetic block 5 inputs an average parameter, which is calculated based on pixels (Pi , Pj and x) of image fields, to the calculation block CALC1 . A block MAX calculates and outputs the peak value of plural signals to be inputted.
-
公开(公告)号:JP2003199115A
公开(公告)日:2003-07-11
申请号:JP2002331847
申请日:2002-11-15
Applicant: ST MICROELECTRONICS SRL
Inventor: CASTORINA ALFIO , MANCUSO MASSIMO , BATTIATO SEBASTIANO
Abstract: PROBLEM TO BE SOLVED: To provide an exposure correction method for improving the quality of a digital image acquired with a no-optimal exposure and taking into account the face that regions, which are important from the perceptive or content point of view, exist in the image. SOLUTION: The method, which is applied directly to the image in CFA (Color Filter Array) format, identifies the regions of the image that are most important from the perceptive or content point of view by means of simple statistical measures. The image is then transformed by inverting then response function is such a manner as to optimize the quality of these regions. COPYRIGHT: (C)2003,JPO
-
公开(公告)号:JP2003189317A
公开(公告)日:2003-07-04
申请号:JP2002250623
申请日:2002-08-29
Applicant: ST MICROELECTRONICS SRL
Inventor: GUARNERA MIRKO , MANCUSO MASSIMO , PULIAFITO ANTONIO , ARENA ANTONINO , VILLARI MASSIMO
IPC: H04N5/91 , H04N9/04 , H04N9/07 , H04N9/79 , H04N101/00
Abstract: PROBLEM TO BE SOLVED: To provide a DSC (Digital Still Camera) system capable of reducing the configuration and energy consumption and producing an image with a small loss on a network. SOLUTION: A digital camera includes: a sensor 205 for sensing an image and producing a first signal; a Bayer pattern producer 210 coupled to the sensor and structured to produce a Bayer pattern from the first signal; a splitter 212 that splits the Bayer pattern into separate color channels; a compressor 220 for compressing the color channels into a compressed image; and an output interface 226 for transmitting the compressed image over a communication channel. The digital camera also includes: a color interpolator 214 that is structured to perform a modification on each of the color channels produced by the splitter 212 prior to transmission of the color channels to the compressor 220. A multimedia packer 222 packs multimedia data in a unique format and transmits the result to the output interface 226 to be connected to a wireless network. COPYRIGHT: (C)2003,JPO
-
9.
公开(公告)号:JPH10336490A
公开(公告)日:1998-12-18
申请号:JP12835998
申请日:1998-05-12
Applicant: ST MICROELECTRONICS SRL
Inventor: MANCUSO MASSIMO , D ALTO VIVIANA , SIRTORI DANIELE , POLUZZI RINALDO
Abstract: PROBLEM TO BE SOLVED: To suppress noise affected on a specific area of a field scanned by the unit of lines by replacing a pixel having noise with an average pixel value of a nearby pixel belonging to a prescribed processing window. SOLUTION: A difference calculation circuit calculates a difference Di (D1-D7) between luminance values Xi (X1-X7) of a pixel adjacent to a processing pixel value X of a processing window. Then a Dmax &Dmin detection block identifies a maximum value and a minimum value of a received difference Di. Upon the receipt of the Dmax and Dmin by an adaptive threshold level detection processing block adopting a 1st fuzzy logic, they are referenced as a membership 1 and corresponding values to the threshold levels Th1, Th2 are generated. Then a local adaptive noise smoothing processing block adopting a 2nd fuzzy block generates a set of output values Ki based on the threshold level parameters Th1, Th2 referenced by a membership 2 and generated before. A final filter block generates a pixel Xout as a sum of weighting factors.
-
公开(公告)号:JPH11168703A
公开(公告)日:1999-06-22
申请号:JP11564598
申请日:1998-04-24
Applicant: ST MICROELECTRONICS SRL
Inventor: MANCUSO MASSIMO , D ALTO VIVIANA , POLUZZI RINALDO , MOLINARI LUCA
Abstract: PROBLEM TO BE SOLVED: To obtain a compensation field rate up method by evaluating an error function for each pair of image blocks, evaluating the extent of homogeneity property for each pair of estimation motion vectors, applying a fuzzy rule which has a start level for each pair of estimation motion vectors, deciding the best estimation motion vector of the pair and deciding an estimation motion vector of the image blocks. SOLUTION: A block C BUILD is supplied according to the output of a block FUZZY, an updating set is supplied to a component of a candidate motion vector C, and eight updating motion vectors are decided. An appropriate image block which is positioned by a motion vector that is selected by a block ADDGEN is decided, and the block ADDGEN is supplied by an output of a block VPROC and an output of the block C BUILD. Error function value that is related to eight updating motion vectors which are evaluated by a block SAD 2 is supplied to a block CSEL, and a motion vector related to image block is selected as an estimation motion vector.
-
-
-
-
-
-
-
-
-