US005276519A
`115)
`United States Patent
`5,276,519
`{11] Patent Number:
`Jan. 4, 1994
`[45] Date of Patent:
`Richardset al.
`
`AAUAACEA
`
`[54] VIDEO IMAGE CAPTURE APPARATUS FOR
`DIGITALLY COMPENSATING
`IMPERFECTIONS INTRODUCED BY AN
`OPTICAL SYSTEM
`
`{75]
`
`Inventors:
`
`John W. Richards, Stockbridge;
`Morgan W.A, David, Farnham, both
`of United Kingdom
`
`[73] Assignee:
`
`Sony United Kingdom Limited,
`Staines, United Kingdom
`[21] Appl. No.: 880,038
`[22] Filed:
`May8, 1992
`[30]
`Foreign Application Priority Data
`Jun. 21, 1991 [GB] United Kingdon................. 9113440
`[51] Unt. CBS .....cesessnesseee HO4N 5/30; HO4N 5/225;
`HO4N 5/213; GO6K 9/32
`[52] U.S. Ch, necsecscsetserteseesensenceoeate 358/209; 358/225;
`358/167; 358/44; 382/44
`(58] Field of Search.............. 358/209, 213.11, 213.15,
`358/225, 43, 44, 167; 382/44, 45, 50
`References Cited
`U.S. PATENT DOCUMENTS
`
`[56]
`
`4,314,281
`
`4,370,641
`1/1983 Kantor etal. once 358/209
`4,549,208 10/1985 Kamejimaet al... 358/108
`
`..........
`« 358/213.15
`4,683,493 7/1987 Taft et al.
`
`4,751,660 6/1988 Hedley..........
`
`4,774,678 9/1988 David et al.
`
`5,023,723
`6/1991 Date etal.
`
`6/1991 Avis...
`+
`5,025,495
`...eessesssccscareeee 382/44
`$,070,465 12/1991 Kato et al.
`
`Primary Examiner—Joseph Mancuso
`Assistant Examiner—Wendy R. Greening
`Attorney, Agent, or Firm—William S. Frommer; Alvin
`Sinderbrand
`
`ABSTRACT
`[57]
`A video image capture apparatus including an image
`sensor for sensing light at a plurality of pixel positions
`received via an optical system, picture storage means
`for temporarily storing pixels derived from the image
`sensor, address generation meansfor applying differing
`write and read addresses to the picture storage to effect
`mapping of input pixels from the image sensor to pro-
`vide output pixels compensating for the effects of im-
`perfections of the optical system is described. A video
`camera producing high quality output images for in-
`stance can thereby be produced.
`
`PETITIONERS EX1017
`Page 1
`
`2/1982 Wiggins et al.
`
`.........csesnseee 382/50
`
`14 Claims, 5 Drawing Sheets
`
`PETITIONERS EX1017
`Page 1
`
`

`

`U.S. Patent
`
`Jan. 4,1994
`
`Sheet 1 of 5
`
`
`
`| 5,276,519
`
`Page 2
`
`PETITIONERS EX1017
`Page 2
`
`

`

`U.S. Patent
`
`Jan, 4, 1994
`
`Sheet 2 of 5
`
`5,276,519
`
`€ZgiOl41JOld
`
`
`
`YOSSIIOUdWol934u0da/vnn)oY
`
`Al
`
`HITIOLNOD7nagHOLI3HH03
`TOHLNOD
`
`
`97JIVIYILNIa
`
`PETITIONERS EX1017
`Page 3
`
`PETITIONERS EX1017
`Page 3
`
`

`

`Sheet 3 of 5
`
`5,276,519
`
`Jan. 4, 1994
`
`U.S. Patent
`
`O|WoJ4
`
`PETITIONERS EX1017
`Page 4
`
`PETITIONERS EX1017
`Page 4
`
`

`

`U.S. Patent
`
`Jan. 4, 1994
`
`Sheet 4 of 5
`
`5,276,519
`
`9S2S0g7Old
`
`
`
`YOLV10du31NI
`
`SSIu00V
`
`INIddVW
`
`AWOWIN
`
`YOLVYINI9
`
`
`
`scquadyNWN109
`
`ONYMOY
`
`
`
`ZZwory
`
`PETITIONERS EX1017
`Page 5
`
`PETITIONERS EX1017
`Page 5
`
`
`
`
`

`

`U.S. Patent
`
`Jan, 4, 1994
`
`Sheet 5 of 5
`
`YOLIIHYOI
`
`YITIOULNOI
`
`5,276,519
`
`USLIVWHO4
`
`“INAS
`
`PETITIONERS EX1017
`Page 6
`
`PETITIONERS EX1017
`Page 6
`
`

`

`1
`
`5,276,519
`
`VIDEO IMAGE CAPTURE APPARATUS FOR
`DIGITALLY COMPENSATING IMPERFECTIONS
`INTRODUCED BY AN OPTICAL SYSTEM
`
`BACKGROUNDOF THE INVENTION
`1. Field of the Invention
`The inventionrelates to video image capture appara-
`tus including means for compensating for the effects of
`optical imperfections.
`2. Description of the Prior Art
`A high definition television system, for example with
`a resolution of 1920 pixels by 1035active lines per frame
`makes very high demands on image capture apparatus
`such as cameras. High accuracy is required in capture
`resolution and also in picture geometryandregistration.
`The resolution limits tend primarily to be a function of
`a sensor arrangement(e.g. a Charged Coupled Device
`(CCD)sensoror a tube target) of the capture apparatus
`whereas the accuracy of a picture geometry and regis-
`tration is primarily determined by the capture appara-
`tus’s optical system through which light has to travel
`before reaching the sensor arrangement.
`Significant distortion of the picture geometry can be
`introduced into a picture output from a camera, for
`instance, even when high quality lenses are used. Mod-
`ern television cameras are typically provided with
`zoom lenses which may work overa relatively large
`optical range and fields of view. Due to the different
`optical and mechanical arrangements of the optical
`system contained in a zoom lens for different zoom
`settings, it is usually only possible to provide a highly
`accurate geometric arrangement at one point in the
`centre of range of operation of the lens. The better the
`lens, the Jess will be the distortion generated at other
`positions. However,
`it
`is not practically possible to
`make a zoom lens whichis optically perfect at all points
`within the optical range. Typically, the design of a lens
`includes compromises to avoid excessive distortion at
`any particular point over its range of zoom operation.
`Similarly, compromises have to be madeto avoid exces-
`sive distortions over the focusing range of a lens,
`whether a zoom lens ora lensof fixed focal length.
`Other distortion effects can occur due to the place-
`ment and alignmentof the image sensor or sensors in a
`camera, for example in a colour camera where a beam
`splitter is employed. These effects occur with fixed
`focal length lenses or zoom lenses.
`Aswell as geometric distortions which can be intro-
`duced in the optical path through a lens (which may
`include many optical elements), chromatic distortions
`may appear at the image sensor. This is because the
`optical elements in the lens affect different light fre-
`quencies differently. Traditionally, a balance has to be
`met between the geometrical and registration errors
`which are considered unacceptable, and the expense of
`the lens. The tighter the specification of the lens(i.e. the
`lowerthe level of geometric, registration and chromatic
`errors which are acceptable), the more expensiveit will
`be. It is to be noted that the cost of a lens in a camera
`system, for example a zoom lens on a video camera,is a
`significant portion of the total cost of the camera sys-
`tem, particularly where a very tight specification is
`required.
`FIGS. 1A, 1B, 1C and 1D of the accompanying
`drawingsillustrate four well known examples of geo-
`
`5
`
`10
`
`20
`
`25
`
`2
`metric distortions which can be introducedinto a pic-
`ture by an optical system.
`FIG.1Aillustrates the so called pin-cushioneffect.
`FIG.1B illustrates a skew effect.
`FIG.1C illustrates barrel distortion.
`FIG. 1D illustratestilt distortion.
`The distortions which occur in a lens system can be
`characterised to a first approximation by combination of
`linear distortion (tilt and/or skew) and second order
`parabolic distortion (barrel or pin-cushion distortion).
`Tube cameras are known whereelectronic meansare
`employed to approximate a correction to a picture dis-
`tortion by applying inverse signals to the horizontal and
`vertical scan drives of the tube to cancel the distortion.
`In other words, rather than attempting to arrange for
`the horizontal scans to be as nearto linear as possible,
`correction signals are used to modify the output of the
`horizontal and vertical scan drives to compensate for
`the distortion encountered. For example, with regards
`to FIG.1A,if pin-cushion distortion is caused by the
`optical system, then the horizontal and vertical scan
`drives of the tube can be controlled so that the electron
`beam traces a pattern to compensate for the distortion,
`so that signals can be output from the tube camera
`which would reflect the original
`image without the
`distortions introduced by the optical system. Such elec-
`tronic means are widely used in current tube cameras.
`However, these cancelling signals are normally gener-
`ated in analogue circuitry and cannot compensate for
`distortions which are not characterised by combinations
`of first and second order effects. Moreover, the tech-
`niques currently used are only applicable to cameras
`with image sensors which employ a scanning beam.
`Theobject of the invention is to provide a video
`image capture apparatus such as a video camera with
`means for compensating for the effects of imperfections
`of the optical system of an image capture apparatus
`which is not limited in application to cameras with
`image sensors employing scanning beams. Forinstance,
`the image sensor or sensors of the video camera might
`include CCD chips.
`
`45
`
`50
`
`60
`
`65
`
`SUMMARYOF THE INVENTION
`In accordancewith the invention,there is provided a
`video image capture apparatus comprising:
`at least one image sensor for sensing light at a plural-
`ity of pixel positions received via an optical system;
`at least one picture storage means, each picture stor-
`age meansassociated with a respective image sensor for
`temporarily storing pixels derived from the respective
`imagesensor; and
`address generation means for applying differing write
`and read addresses to each picture storage means to
`effect mapping of input pixels from each Tespective
`image sensor to provide output pixels associated with
`each image sensor compensatedfortheeffects ofimper-
`fections of the optical system.
`- Preferably the address generation means provides
`write addresses for storing the input pixels from each
`image sensorin each respective picture storage meansat
`addresses directly related to the sensed inputpixel posi-
`tions, and wherein the address generation means pro-
`vides read addresses associated with each picture stor-
`age means for mapping the stored pixels from each
`picture storage meansso as to form the output pixels
`associated with each image sensor which are compen-
`sated for the effects of imperfections of the optical sys-
`tem.
`
`PETITIONERS EX1017
`Page 7
`
`PETITIONERS EX1017
`Page 7
`
`

`

`3,276,519
`
`_ 5
`
`20
`
`25
`
`40
`
`3
`4
`Thus, advantageously, even if the apparatus com-
`efficient and effective way. The optical system may
`prises more than one image sensor, the address genera-
`have varying parameters. Forinstance, the optical sys-
`tion means provides only one write address to each
`tem may form part of a zoom lens. Even if the optical
`picture storage means. However,in the case of an appa-
`system is not part ofa zoomlens, buta fixed focal Jength
`ratus comprising more than one image sensor, as each
`lens, it may be focused over a range of distances.
`image sensor may be subject to differing effects of im-
`Advantageously apparatus according to the inven-
`perfections of the optical system, the address generation
`tion can form part of a colour camera. For instance
`meansprovides different read addresses to eachpicture -
`image sensors, each sensitive to light of a different pri-
`storage means for mapping the stored pixels.
`mary colour, may senselight passing through the opti-
`Preferably each picture storage means has a pixel
`cal system. Distortions to an imageat any ofthe sensors,
`interpolator connected to it for interpolating between
`caused by imperfections of the optical system, can be
`stored input pixels where the mapping does not provide
`separately compensated for by separate digital control
`one-to-one mapping between an input pixel position and
`signals. If desired, apparatus accordingto the invention
`an output pixel position. Advantageously, the address
`could comprise any numberofsensors.
`generation means provides read addresses having a
`Preferably, if the apparatus comprises more than one
`major address portion for addressing each picture stor-
`sensor, but only a single optical lens system, then an
`age meansandaresidual address portion for controlling
`optical beam splitter means is provided in the optical
`each pixel interpolator. In this way, output pixels are
`system before the sensors. The beam splitter divides an
`generated associated with each image sensor compen-
`image received by the optical system into a number of
`sating for the image distorting effects of imperfections
`identical separate images and passes each separate
`of the optical system in a way that is independent of the
`image to a respective sensor. In this way, a number of
`form of the distortion. Thatis: the distortion may be, for
`sensors, each sensitive to light of different frequencies,
`instance,first order skew ortilt, second order barrel or
`can each receive the image received by the optical
`pin-cushion, or a higher order distortion, or even a
`system. The separate images sensed by each sensor may
`distortion that cannot be represented by an analytic
`function.
`later be combined. Forinstance, in the case of a colour
`camera comprising three sensors that are each sensitive
`In the case of an apparatus comprising two or more
`to light of a different primary colour, the three primary
`sensors, aS an alternative to the address generation
`colour images sensed by the sensors may be combined
`means applying the same write addresses, and differing
`to formafull colour image.
`read addresses to each sensor, an apparatus can be pro-
`vided in which the address generation means applies
`Optical distortions may also be introduced by the
`write addresses associated with each picture storage
`positioning of the beam splitter and by the beamsplitter
`means for mapping the input pixels from each image
`itself. Further, the position of each sensor may giverise
`sensor onto each respective picture storage means so as
`to a distortion of an image andso this position could
`to arrange the pixels to compensate for the effects of
`also be a varying parameter. Preferably, the address
`imperfections of the optical system and wherein the
`generation meansis responsive to control signals repre-
`address generation means applies read addresses for
`sentative of current parameters to apply the address to
`directly outputting the stored pixels from the picture
`eachpicture storage meansto provide the outputpixels
`storage meansto form the output pixels associated with
`compensating for the effects of imperfections of the
`each image sensor compensating for the effects of im-
`optical system. Advantageously, the control signals can
`perfections of the optical system. In this caseit is prefer-
`be supplied by a corrector controller. The corrector
`able that a pixel interpolator is connected between the
`controller senses, for instance, the lens type, its current
`or each image sensor and each respective picture stor-
`focusing position and, in the case of a zoom lens,its
`age means for interpolating between the sensed image
`currentfocal length. These parameters may be constant
`pixels where the mapping does not provide one-to-one
`or continuously varying. The corrector controller can
`mapping between an input pixel position and an output
`supply the address generation means with controlsig-
`pixel position.
`nals according to the current parameters ofthe optical
`Preferably, the write addresses have a major address
`system and/or each sensor.
`portion for addressing each picture storage means and a
`The above, and other objects, features and advan-
`residual address portion for controlling each respective
`tages of this invention will be apparent from the follow-
`pixel interpolator.
`ing detailed description of illustrative embodiments
`Such an address generation meansis preferably capa-
`which is to be read in connection with the accompany-
`ble of generating address independently of any control
`ing drawings.
`signals from the corrector controller and so output
`pixels associated with each image sensor maybe pro-
`vided even if the optical system and/or each image
`sensor is unable to generate an information signal ac-
`cording to any varying parameters.
`In a further embodimentof the invention, an interpo-
`lator is provided between the optical system and each
`picture storage means for motion adaptive progressive
`scan conversion. Such a video image capture apparatus
`is able both to compensate for the effects of imperfec-
`tions in the optical system while taking account of
`movements within the image.
`A video image capture apparatus according to an
`embodimentof the invention can compensatefor distor-
`tion to an image created by the optical system in an
`
`BRIEF DESCRIPTION OF THE DRAWINGS
`FIGS. 1A-1Dillustrate typical geometric distortions
`which can be introduced by the optical system of an
`image capture apparatus;
`FIG.2 is a schematic block diagram giving an over-
`view of a video image capture apparatus in accordance
`with the invention;
`FIG.3 is a schematic block diagram ofa first example
`of a geometry,registration, and chromatic error correc-
`tor for the image capture apparatus of FIG.2;
`FIG. 4 is a schematic block diagram of an address
`generator forming part of the corrector unit of FIG.3;
`and
`
`—
`
`45
`
`50
`
`55
`
`60
`
`65
`
`PETITIONERS EX1017
`Page 8
`
`PETITIONERS EX1017
`Page 8
`
`

`

`DESCRIPTION OF THE PREFERRED
`EMBODIMENTS
`
`5,276,519
`
`20
`
`25
`
`30
`
`35
`
`5
`FIG. 5 is a schematic block diagram of a second
`example of a corrector for the image capture apparatus
`of FIG.2.
`
`6
`buffer memories 30 and 32, each address generators 38
`and 40 addressing the memories 30 and 32in an alternat-
`ing manner during successive periods such that, during
`odd periods, the first buffer memory 30 receives write
`addresses for receiving input pixel data from the ana-
`logue to digital convertor 16 and the second buffer
`FIG.2 is a schematic block diagram providing an
`memory 32 receives read addresses for outputting out-
`overview of an image capture apparatus in accordance
`put pixel data and, during even periods, the second
`with the inventionin the form of a video camera system.
`buffer memory 32 receives write addresses for receiving
`The camera system has a camera head 10 which com-
`input pixel data from the analogueto digital converter
`prises an optical system 12, which includes a zoom lens ©
`16 and the first buffer memory 30 receives read ad-
`and associated control circuitry, an image sensor 14 for
`dresses for outputting outputpixel data, thereby provid-
`registering light received through the optical system 12,
`ing the continuous video output. The write address
`and signal amplification and camera head control cir-
`generator 38 operates by providing continuouslinearly
`cuitry (not shown). An amplified output of the image
`incrementing row and column addresses (row by row
`sensor 14 is passed to an analogueto digital converter 16
`and within each row column by column) to the field
`which converts analogue image signals produced by an
`buffer memory currently being written to for storage of
`image on the image sensor 14 into digital signals for
`the picture in that memoryas it is received from the
`processing. The digital signal output of the analogue to
`analogue to digital convertor 16. In other words, a
`digital convertor 16 comprises image pixel data whichis
`direct mapping of imagesignals from the image sensor
`passed to a geometry, registration and chromatic error
`14 to addresses in the picture storage 28 is provided.
`corrector 18 which compensates for the effects of opti-
`The pixel data in the form of digital signals in the pic-
`cal imperfections of the zoom lens. The corrector 18is
`ture storage 28 therefore reflects the effects of any dis-
`under the control of a corrector controller 22. The
`tortions and/or imperfections caused by the optical
`corrector controller 22 is connected to the lens control
`system formed from the image signals.
`circuitry via a connection 15 to receive information
`The read address generator 40, however, reads the
`about the current zoom and focus position, and the
`information from the picture storage 28 in accordance
`identity of the lens. From this information, the correc-
`with a mapping characteristic representative of the
`tor controller 22 is able to provide appropriate control
`inverse of the distortion incorporated in the captured
`signals to the corrector 18 for performing the appropri-
`images. In other words, the read address generator 40,
`ate corrections to the pixel data. An interface 26is pro-
`under control of the corrector controller 22, provides
`vided for inputting external information to program the
`selective addressing ofthe picture storage 28 in orderto
`corrector controller 22 to provide the appropriate con-
`generate the digital signal output in the form ofpixel
`trol signals according to various lens positions and the
`data for forming the rows and columns of an output
`image.
`lens identity. The interface 26 can be a socket to enable
`the connection of the corrector controller 22 to a com-
`In order to provide high quality output video, the
`puter or the like for inputting the programming infor-
`tead address generator 40 calculates addresses to sub-
`mation. The output of the corrector 18 is supplied to a
`pixel accuracy. In other words, there may not be an
`conventional Camera Control Unit (CCU)processor 23
`exact mapping from a pixel in the picture storage 28 on
`for conventional processing of the image data. The
`to a particular outputpixel. In order to correctly gener-
`CCU processor 23 operates in a conventional manner
`ate the output pixel it may be necessary to interpolate
`under control of a CCU control unit 24.
`between a numberofpixels in the picture storage 28.
`FIG.3 is a schematic block diagram of one example
`Accordingly, a pixel interpolator 34 is provided for this
`of a geometry, registration and chromatic error correc-
`purpose. The read address generator 40 produces ad-
`tor 18 for incorporation in the image capture apparatus
`dresses comprising a major, or integer portion which is
`illustrated in FIG. 2. The corrector 18 illustrated in
`used for addressing the pixel locations in the picture
`FIG.3 is able to process one image signal channel. In
`storage 28 and a fractionalor residual proportion which
`other words, the apparatus shown in FIG.3 is suitable
`is used for controlling the interpolator 34. The output of
`for a monochrome video image cameraor for forming
`the pixel interpolator 34 is supplied to a synchronisation
`one channel of a colour camera.
`formatter 36 for output to the CCU processor 23.
`The output from the analogueto digital convertor 16
`For synchronising the operation of the various ele-
`is supplied to a picture storage 28. The picture storage
`ments ofthe corrector 18, a synchronisation decoder 43
`28 is arranged in a form of a pair of field memories 30
`extracts synchronisation information from the digital
`and 32 which are written to and read from onafield
`signal output from the analogueto digital converter 16
`alternating basis so as to provide a continuous video
`and supplies control signals to the various elements of
`output. A write address generator 38 is provided for
`the corrector 18 to ensure an accurate periodic alterna-
`determining write addresses for the first and second
`tion between the operation of the two field memories 30
`and 32.
`field memories 30 and 32 for the receipt of input pixel
`data from the analogueto digital convertor 16.
`FIG. 4 is a schematic block diagram of the read ad-
`A read address generator 40 provides read addresses
`dress generator 40. The read address generator 40 com-
`to the first and second field memories 30 and 32 for the
`prises a simple row and column address generator 50
`outputting of corrected pixel data from those memories.
`which generates pixel addresses by column by column
`A selector 42 controls switching of the video output
`and, within each column, row by row. These addresses
`from field memory 30 and field memory 32 and switch-
`are used to access a mapping memory 52 which contains
`ing of video input pixel data to the field memories 30
`the picture storage address which needs to be accessed
`and 32 by the provision of appropriate enable signals as
`in orderto select a pixel or pixels from the picture stor-
`will be apparent to one skilled in the art. In other words,
`age 28 in order to producean output pixel for display at
`the picture storage 28 comprises first and second field
`the address produced by the simple address generator
`
`40
`
`45
`
`50
`
`60
`
`65
`
`PETITIONERS EX1017
`Page 9
`
`PETITIONERS EX1017
`Page 9
`
`

`

`5,276,519
`
`7
`50. In other words, the mapping memory52 is in the
`form ofa translation table defining the mapping which
`needs to be performed in order to correct for effects of
`the distortions in the optical system.
`The row and column address generator 50 and the
`mapping memory 52 need not operateat the full resolu-
`tion of the picture storage 28. For example, they may
`contain locations corresponding to a grid density of 3th
`horizontal and 3th vertical resolution. The map grid
`density depends on the geometric accuracy of the cor-
`rection required. If the distortion is smooth, then sub-
`sampling by much greater numbers maybepractical.
`The output of the mapping storage 52 is fed to an
`address interpolator 54, which produceslinearly inter-
`polated full resolution pixel and line addresses. The
`integer part of each of the horizontal and vertical ad-
`dresses is used to address the picture storage 28, whilst
`the fractional part of those addressesis fed to the inter-
`polator 34.
`Theinformation relating to a current zoom and focus
`position and the identity of the lens is downloaded to
`the mapping memory52 during a non-active scan time,
`in other words during a time when the read address
`generator 40 is not active.
`Calibration of the mapping memory 52 is performed
`manually by providing a test card with a regular square
`grid pattern and manually adjusting the parameters
`stored to produce a corrected pattern. This could be
`done for example by displaying the camera output ona
`display containing a superimposedgraticule.
`As mentioned above, a geometry, registration and
`chromatic error corrector 18 provides for the correc-
`tion of one channelof input video information.In a full
`colour (e.g. a RGB) system, three channels of correc-
`tion need to be provided. However, it is not necessary
`to provide all of the elements in FIG. 3 for each of the
`channels. Several parts of the corrector 18 could be
`commonto all three channels. Specifically, there need
`only be one synchronisation decoder 43, one write ad-
`dress generator 38, and one synchronisation formatter
`36. The remaining components of FIG. 3 enclosed by
`the dotted line 44 in that Figure are, however provided
`separately for each of the separate channels. FIG. 5
`illustrates this arrangement with separate red, blue and
`green processors 44R, 44B, and 44G, respectively.
`Each one of the processors 44R, 44B, and 44Gre-
`ceives digital image signals from a respective image
`sensor 14R, 14B, and 14G via a respective analogue to
`digital converter 16R, 16B and 16G. Each image sensor
`14R, 14B and 14G registers light through an optical
`system 12’ which includesan optical beam splitter 13 for
`producing three light beams.
`There has been described a video capture apparatus
`including means for compensating for the effects of
`optical imperfections. It will be appreciated that the
`particular embodiments described are illustrative and
`that many additions or modifications are possible within
`the scope of the invention.
`For instance, an alternative to the arrangements of
`elements in the embodimentsof the invention described
`above would be to have the write address generator,
`rather than the read address generator responsive to the
`corrector controller. In this case the interpolator would
`be located prior to the picture storage.
`It would be possible to incorporate known progres-
`sive scan conversion techniques into the apparatus. For
`instance a motion adaptive progressive scan converter
`(MAPSC) could be placed between the analogue to
`
`8
`digital converter and the corrector to receive digital
`signals and pre-process them before they are corrected.
`The MAPSC would receive the image pixel data from
`the analogue to digital converter in an array compre-
`hendible form. In wholly static picture areas inter-field
`interpolation is applied whilst if significant motion is
`detected intra-field interpolation is performed by the
`MAPSC.Theprocessed pixel data from the MAPSC
`would then be passed to the corrector. In this way, a
`capture apparatus according to the invention could
`provide output pixels compensating for the effects of
`imperfectionsofthe optical system whenthereis signifi-
`cant motion in the image.
`Although the means for compensating for the effects
`of optical imperfections herein described is well suited
`to a high definition television standard system,it is also
`equally applicable to a lower definition system, such as
`one with a resolution of 625 or 525 active lines per
`frame. In both cases the apparatus provides a higher
`quality video output from a video image capture appa-
`ratus by mitigating the effects of optical imperfections.
`Also, the correction circuitry could be placed after,
`rather than before the CCU.
`Although the embodimentof the invention described
`aboveis a camera system comprising a camera head and
`signal processing circuitry, incorporating the correc-
`tion circuitry for compensating for the effects of imper-
`fections in the optical system, separate from the camera
`head, the correction circuitry and/or othersignal pro-
`cessing circuitry could be incorporated in the camera
`head. Indeed the correction circuitry could be incorpo-
`rated in a camera or camera/recorder designed for
`stand alone operation.
`.
`The corrector controller in the embodiments de-
`scribed aboveis responsive to signals from the optical
`system representing its current parameters to produce
`control signals for the address generation means. In the
`case wherethe optical system is used by more than one
`image sensor, it can include an optical beam splitter.
`The beam splitter may also distort an image being
`passed throughit and the corrector controller could be
`madesensitive to control information from it. Further,
`the positioning of the individual image sensors within
`the capture apparatus may bevariable and may giverise
`to imagedistortion. In that case the corrector controller
`could be madesensitive to control signals from the or
`each image sensor.
`Although illustrative embodiments of the invention
`have been described in detail herein with reference to
`the accompanying drawings,it is to be understood that
`the invention is not limited to those precise embodi-
`ments, and that various changes and modifications can
`be effected therein by one skilled in the art without
`departing from the scope and spirit of the invention as
`defined by the appended claims.
`Weclaim:
`1. A video image capture apparatus comprising:
`at least one imagesensorfor sensing light captured by
`an optical system at a plurality of pixel positions
`and providing respective inputpixels, said optical
`system exhibiting optical imperfections;
`at least one picture storage means, each picture stor-
`age meansassociated with a respective image sen-
`sor for temporarily storing the input pixels derived
`from the respective image sensor; and
`address generation meansfor applying differing write
`and read addressesto each picture storage means to
`map the input pixels from each respective image
`
`30
`
`40
`
`45
`
`60
`
`65
`
`PETITIONERS EX1017
`Page 10
`
`PETITIONERS EX1017
`Page 10
`
`

`

`5,276,519
`
`9
`sensor into output pixels associated with the re-
`spective image sensor which are compensated for
`the imperfections introduced by the optical system.
`2. A video image capture apparatus as claimed in
`claim 1 wherein the address generation means provides
`write addresses for storing the input pixels from each
`imagesensorin each respective picture storage meansat
`addressesdirectly related to the sensed input pixel posi-
`tions, and wherein the address generation means pro-
`vides read addresses associated with each picture stor-
`age means for mapping the stored pixels from each
`picture storage means so as to form the output pixels
`associated with each image sensor compensated for the
`imperfections introduced by the optical system.
`3. A video image capture apparatus as claimed in
`claim 2 further comprising at least one pixel interpola-
`tor, each pixel interpolator connected to a respective
`picture storage meansfor interpolating between stored
`input pixels where the mapping between an input pixel
`position and an output pixel position is other than a
`one-to-one mapping.
`4. A video image capture apparatus comprising:
`at least one image sensorfor sensing light at a plural-
`ity of pixel positions received via an optical system;
`at least one picture storage means, each picture stor-
`age means associated with a respective image sen-
`sor for temporarily storing pixels derived from the
`respective image sensor and for providing the
`stored pixels as output pixels;
`at least one pixel interpolator, each pixel interpolator
`connected to a respective picture storage means for
`interpolating between said output pixels in re-
`sponse to an interpolation control signal; and
`address generation means for applying write ad-
`dresses to each picture storage means for storing
`the input pixels from each image sensor in each
`respective picture storage means at addresses di-
`rectly related to the sensed input pixel positions,
`and for applying read addresses to each picture
`storage means for mapping the stored pixels from
`each picture storage means so that said output
`pixels are associated with