US005592256A
`[11] Patent Number:
`United States Patent 19
`—
`aten
`umber:
`Muramatsu
`[45] Date of Patent:
`
`
`5,592,256
`Jan. 7, 1997
`
`NTAACETATE
`
`1/1989 Burt
`4,797,942
`[54] PHOTOMETRY DEVICE FOR A CAMERA
`...escsessssssessceeserecssesevseneeesses 382/41
`
`
`5,187,754 2/1993 Currin et al.woeeeeeeeneees 382/54
`8/1993 Janda oo...cecseseseesessseeeneees 364/413.13
`5,233,517
`(75]
`Inventor: Masaru Muramatsu, Kawasaki, Japan
`5,249,015
`9/1993 Takagi etal. .
`.
`5,266,984 11/1993 Muramatsu et al.
`5,270,767 12/1993 Hamada et al... eeseessesenes 354/430
`5,319,416
`6/1994 Takagi
`...cssesssceccesscssseseneesseres 354/432
`e
`Primary Examiner—Safet Metjahic
`Assistant Examiner—D. P. Malloy
`
`[21] Appl. No.: 654,998
`[22]
`Filed:
`May 29, 1996
`
`[73] Assignee: Nikon Corporation, Tokyo, Japan
`
`Related U.S. Application Data
`
`[63] Continuation of Ser. No. 246,424, May 20,
`.
`oo.
`_.
`[30]
`Foreign Application Priority Data
`Jun, 8, 1993
`[JP]
`Japan cecssssecsssssecseesesreceete 5-163919
`
`1994, aban
`
`P
`Tmt, C0 neecececcecseeseeceeeseteoeeceeeeeeeneess GO03B 7/08
`(51)
`[52] U.S. Ch eeeceneeeneeene 396/225; 348/362; 348/366;
`396/234
`[58] Field of Search 000...cece 354/430, 432;
`348/362, 366
`
`[56]
`
`References Cited
`U.S. PATENT DOCUMENTS
`7/1983 Terasita .
`10/1984 Fukuhara et al. .
`
`4,395,099
`4,476,383
`
`[57]
`ABSTRACT
`A photometric device for use in a camera includes,butis not
`limited to, a segmented brightness measuring unit which
`segments a photographic scene of a camera into multiple
`areas and which outputs corresponding multiple photometric
`values. A spectral analysis unit performs spectral analysis of
`the spatial frequency of a photographic subject using the
`corresponding multiple photometric values output by the
`segmented brightness measuring unit. Also, a photometric
`computation unit computes photometric values based on the
`spectral pattern of the photographic subject output by the
`spectral analysis unit. The spectral analysis unit may be
`disposed within the photometric device, or may be located
`elsewhere in a camera’s body.
`
`20 Claims, 4 Drawing Sheets
`
` ODOOOOOROOOOO0GOH OOOOOOOH
` OHOBOoOoONo
` OOOoOOO00DOOOOOOo0
`OOOOQOO000OOOOOOOO
`
`
`
`SPECTRAL
`
`ANALYSIS UNIT
`
`
`MULTI—PATTERN
`PHOTOMETRIC UNIT
`
`
`
`
`
`
`EXPOSURE.
`VALUE
`CONTROL
`
`HUAWEI EX. 1212 - 1/10
`
`Canon Exhibit 1212
`Page 1
`
`Canon Exhibit 1212
`Page 1
`
`HUAWEI EX. 1212 - 1/10
`
`

`

`5,592,256
`
`Sheet 1 of 4
`
`Jan. 7, 1997
`
`U.S. Patent
`
`FIG. 2
`
`OOOOOUOO
`
`MOOUOUOO
`
`MOUWOBUUOO
`
`OOOOOOUO
`
`OOOOOUUUU
`
`OOOUOUUOUO
`
`OOOWOUOU
`
`DOOOOUOO
`
`™OO
`
`HUAWEI EX. 1212 - 2/10
`
`Canon
`
`Exhibit 1212
`Page 2
`
`Canon Exhibit 1212
`Page 2
`
`HUAWEI EX. 1212 - 2/10
`
`
`

`

`U.S. Patent
`
`Jan. 7, 1997
`
`Sheet 2 of 4
`
`5,592,256
`
`
` SPECTRAL
`
`FIG. 3 OOOOOOOoOBEDOOHROOOOO0O000OOOOOO0HOOOOOOOOOOOOORRDONOODOROoCOOOOOnn
`ANALYSIS UNIT
`
`
`FIG. 4
`
`HUAWEI EX. 1212 - 3/10
`
`Canon Exhibit 1212
`Page 3
`
`Canon Exhibit 1212
`Page 3
`
`HUAWEI EX. 1212 - 3/10
`
`

`

`US. Patent
`
`Jan. 7, 1997
`
`Sheet 3 of 4
`
`5,592,256
`
`FIG. 5
`
`HI
`
` DCMD(DC)
`
`DCBG(DC)
`
`
`
`FIG. 6
`
`FIG. 7
`
`FIG. 8
`
`0
`
`;
`
`0
`
`;
`
`0
`
`LOSL(LO)
`
`LOBG(LO)
`
`DC
`
`LO
`
`HISL(HI)
`
`HIBG(HI)
`
`HI
`
`HUAWEI EX. 1212 - 4/10
`
`Canon Exhibit 1212
`Page 4
`
`Canon Exhibit 1212
`Page 4
`
`HUAWEI EX. 1212 - 4/10
`
`

`

`U.S. Patent
`
`Jan. 7, 1997
`
`Sheet 4 of 4
`
`5,592,256
`
`START
`
`si
`
`a[n,m]=1/64-£ X’(B[x,y]>cos(2 x nx/B) *cos(2+ my/8))
`
`
`
`
`82
`
`x=0 yO
`
`n:0..4 m:0..4
`
`
`
`b[n,m]=1/64= E(ols)sinaxnx/8) *sin(2 « my/8))
`
`
`
`s4
`
`P[n,mJ=a[n,m]-a{n,m] + b[n,m]-b[n,m]
`n:0..4 m:0..4
`
`FIG. 9
`
`26
`
`7
`
`s5~ DC=P[0,0]
`
`Lo=x’ y*Pinm] - DC
`n=O m=0
`
`HI=£* £*P[nm] -LO - DC
`n=0 m=0
`
`K{1]=Min(DCSL(DC), LOSL(LO), HISL(HI))
`
`i
`
`i
`
`i
`
`(
`
`(
`
`;
`
`1)
`
`0
`
`0
`
`DC)
`
`DC)
`
`C)
`
`(
`
`LO)
`
`)
`
`)
`
`LO),
`
`;
`
`HISL(HE))
`
`(
`
`LO
`
`i
`
`)
`
`H
`
`1)
`
`DC
`
`)
`
`28~]
`
`“
`
`H3
`
`816
`
`K[9]=Min(DCBG(DC), LOSL
`(
`i
`=
`(
`
`(LO),
`LO)
`
`HIBG(HI)
`
`)
`
`a7
`
`ras
`
`i
`
`(DC
`
`)
`
`)
`
`)
`
`)
`
`
`819—“]K[12]=Min(DCBG(DC), LOBG(LO), HIBG(HI
`)
`),
`=
`(
`~—
`LO
`
`
`
`
`
`s20~]
`
`BP[nJ=W[n,1]-DC + W[n,2]+LO + Mino]H + C[n]
`
`al
`
`n:
`
`321
`
`—7]80=E,(KIn]+BPIn})/EKin]
`
`HUAWEI EX. 1212 - 5/10
`
`Canon Exhibit 1212
`Page 5
`
`Canon Exhibit 1212
`Page 5
`
`HUAWEI EX. 1212 - 5/10
`
`

`

`5,592,256
`
`1
`PHOTOMETRYDEVICE FOR A CAMERA
`
`This application is a continuation of application Ser. No.
`08/246,424,filed May 20, 1994, now abandoned.
`
`BACKGROUND OF THE INVENTION
`
`1. Field of the Invention
`
`The present invention relates to a camera photometric
`device and, more particularly,
`to a camera photometric
`device which segments a photographic scene into multiple
`areas in which to perform photomeiry.
`2. Description of the Related Art
`In the past, camera photometric devices included multi-
`pattern photometric units which divided or segmented a
`photographic scene into multiple areas in which to perform
`classification of a photographic subject. The multiple areas
`were derived by examining the maximum brightness levels
`and the brightness differences among sections of a photo-
`graphic subject. The multiple areas were producedas a result
`of analysis of respective photometric outputs based on the
`aforementioned brightness levels which were selected from
`among several well-known methods, including an exposure
`value computation method which was appropriate for each
`pattern with respect to a photographic scene which corte-
`sponds to a respective pattern. The aforementioned outputs
`were then usedto instruct cameracircuitry to perform proper
`exposure control. For a discussion of exemplary brightness
`detecting photometric systems, refer to U.S. Pat. No. 4,395,
`099 to Terasita and U.S. Pat. No. 4,476,383 to Fukuharaet
`al.
`
`Jt is important to note that prior photometric systems were
`designed so that in cases in which the maximum brightness
`of the photometric output was high and the differences in
`brightness levels were large, such photometric systems
`performed exposure control with an emphasis on low bright-
`ness conditions as if a photographic scene included the sun
`or sun light.
`In terms of the problems associated with conventional
`photometric devices of the type described above, there are
`cases in which different exposures are required even in
`photographic scenes classified as the same pattern. For
`example, conventional photometric devices were often
`designed to recognize and detect a scene with a high
`maximum brightness and large differences in brightness
`levels (e.g., a case such as that of a person being rear-lit by
`the sun). In such situations, an exposure which emphasized
`low-brightness levels of a photographic subject was set,
`whichresulted in a problem in that the portions on which the
`sun was shining were overexposed. Accordingly, the con-
`ventional classification methods described above resulted in
`scenes which could not be sufficiently classified and which
`ultimately resulted in poor picture quality.
`
`SUMMARYOF THE INVENTION
`
`Accordingly, it is an object of the present invention to
`solve the above-mentioned problems associated with exist-
`ing photometric devices.
`It is another object of the present invention to provide a
`photometric device for a camera which can perform a
`classification of a photographic scene having complex pat-
`terns and brightness variances with high accuracy and which
`makes possible the selection of a photometric computation
`methodthat accurately conforms to a photographic subject.
`
`10
`
`15
`
`20
`
`35
`
`40
`
`45
`
`50
`
`55
`
`60
`
`65
`
`2
`It is still a further object of the present invention to
`provide a photometric device for a camera which utilizes a
`spatial frequency value transformation operation to prepare
`exposure control commands.
`It is still another object of the present invention to provide
`a photometric device for a camera which utilizes a Fourier
`transformation operation to prepare exposure control com-
`mands.
`
`It is yet another object of the present invention to provide
`a photometric device which can becost-effectively imple-
`mented in a camera device.
`
`These and other objects of the present invention are
`achieved by providing a photometric device for use in a
`camera which includes, but is not limited to, a segmented
`brightness measuring unit which segments a photographic
`scene to be captured by the camera into multiple areas and
`which outputs corresponding multiple photometric values.
`Moreover, the photometric device includes a spectral analy-
`sis unit which performs spectral analysis of the spatial
`frequency of a photographic subject using the corresponding
`multiple photometric values output by the segmented bright-
`ness measuring unit. Also, the photometric device includes
`a photometric computation unit which computes photomet-
`tic values based on the spectral pattern of the photographic
`subject output by the spectral analysis unit.
`
`BRIEF DESCRIPTION OF THE DRAWINGS
`
`The above-mentioned and other objects and advantages of
`the present invention will become apparent and morereadily
`appreciated from the following description of the preferred
`embodiment, taken in conjunction with the accompanying
`drawings, of which:
`FIG. 1 is a block diagram which depicts an embodiment
`of a photometric device according to the present invention.
`FIG. 2 is a diagram which depicts a divisional configu-
`ration of the photometric device depicted in FIG.1.
`FIG. 3 is a block diagram which depicts a configuration
`of a photometric computation device part of the embodiment
`depicted in FIG. 1.
`FIG. 4 is a graph which depicts an example of a two-
`dimensional power spectra P.
`FIG. 5 is a diagram which depicts a condition in which a
`subject is classified by spectra.
`FIG. 6 is a line graph which depicts a membership
`function which relates to a direct current component DC.
`FIG. 7 is a line graph which depicts a membership
`function which relates to a low-frequency component LO.
`FIG. 8 is a line graph which depicts a membership
`function which relates to the high-frequency componentHI.
`FIG.9 is a flow chart which illustrates the operations of
`the photometric device depicted in FIG. 1.
`
`DETAILED DESCRIPTION OF A PREFERRED
`EMBODIMENT
`
`Reference will now be made to a detailed description of
`a preferred embodiment of the present invention with ref-
`erence to the drawing figures which were briefly described
`above. Like parts are referred to with like reference numer-
`als.
`
`Referring now to FIG. 1, therein depicted is a block
`diagram of a single-lens reflex (“SLR”) camera built into
`which is a preferred embodiment of a photometric device
`according to the present invention.
`
`HUAWEI EX. 1212 - 6/10
`
`Canon Exhibit 1212
`Page 6
`
`Canon Exhibit 1212
`Page 6
`
`HUAWEI EX. 1212 - 6/10
`
`

`

`5,592,256
`
`4
`
`aln, m] = 1/64 - 5, E> (Blx, y] - cos(2mnx/8) - cos(2rmy/8)
`
`[1]
`
`Where: n and m are numbers from 0 to 4;
`
`z,
`
`expresses a sum from x=0 to 7;
`
`D>
`
`10
`
`expresses a sum from y=0 to 7.
`
`bin, m] = 1/64 - Z4 Z4 (Bix, y] - sin(QQanx/8) - sin(2nmy/8))
`
`[2]
`
`Where:
`
`5
`
`expresses a sum from x=0 to 7;
`
`Ey
`
`Pin, m)=a[n, m]-a[n, m]+b[n, m)-b[n, m)}
`
`[3]
`
`Referring now to FIG.4, therein depicted is a graph which
`illustrates an example of the two-dimensional power spectra
`P output by spectral analysis unit 81 (FIG. 3). A total of 25
`power spectra from P[0, 0] to P[4,4] are indicated.
`The multi-pattern photometric unit 82 is divided into three
`frequency ranges: the direct current component DC of P[0,
`0] in FIG.4, 8 low-frequency ranges which are indicated by
`A and 16 high-frequency ranges which are indicated by B.
`With the three-mentioned ranges, photometric computation
`unit 8 computes the low-frequency component LO which is
`the total value of the power spectra of low-frequency range
`A, and the high-frequency component HI whichisthetotal
`value of the high-frequency range B. Photometric compu-
`tation unit 8 computes the direct current component DC, the
`low-frequency component LO andthe high-frequency com-
`ponent HI according to the following equations:
`
`20
`
`25
`
`30
`
`35
`
`|
`
`45
`
`50
`
`55
`
`60
`
`3
`Light L reflected from a photographic subject whichis to
`be photographed is directed to a focusing screen 4 after it
`passes through a picture-taking or objective lens 2. There-
`after, light L is reflected by a quick return mirror 3 to be
`observed as a scene by a photographer via a pentagonal
`prism 5 and conventional eyepiece optical system.
`Additionally, the image of the photographic subject which
`is produced by light L and which is formed on focusing
`screen 4 is directed to a photometric sensor 7 by a photo-
`metric image reforming lens 6. Photometric sensor 7 is a
`sensor with multiple segments and is shown in FIG. 2.
`Photometric sensor 7 produces photometric output signals
`which describe or otherwise model the brightness distribu-
`tion of the image of the photographic subject. The signals
`produced by photometric sensor 7 are sent to a photometric
`computation unit 8.
`Photometric computation unit 8 is configured to include,
`but should not be limited to, a microcomputer or a micro-
`processor. Photometric computation unit 8 computes expo-
`expresses a sum from y=0 to 7.
`sure control values based on the photometric outputsignals
`If the number of segments in the horizontal and vertical
`produced by photometric sensor 7.
`directions is a power of 2, the photometric sensor 7 can use
`Referring now to FIG. 2, therein depicted is a diagram
`a so-called fast Fourier transform (FFT) algorithm,so it can
`which illustrates a photographic scene of a segmented
`execute the computations in equations [1] and [2] at a high
`photometric device accordingto the principlesof the present
`speed using conventional techniques.
`invention. As shown in FIG. 2, segmented photometric
`sensor 7, in effect, segments the photographic scene into
`The two-dimensional power spectrumPis as follows:
`eight sections in both the horizontal and vertical directions
`and outputsa total of sixty-four individual signals indicative
`of scene brightness data. That is,
`the segments forming
`photometric sensor 7 are arranged rectangularly (e.g., a
`square), but the invention should not be so limited. In fact,
`other arrangements and quantities of segments may be
`possible. For purposes of explanation, the respective pho-
`tometric outputs are individually considered as the value
`B[x, y] where B represents BRIGHTNESS andthe coordi-
`nate pair [x,y] indicates a particular segment of the photo-
`metric sensor 7.
`
`Segmented photometric device 7 is arranged so that a
`sample is 8 points (e.g., segments)x8 points (e.g., segments).
`Thus, a high-frequency component
`in which the spatial
`frequency is 4 or more cycles per 8 points, shades image
`reforming lens 6 and performs division using an optical
`low-pass filter in order to accurately detect low-frequency
`components.
`therein depicted is a block
`Referring now to FIG. 3,
`diagram which illustrates the configuration of photometric
`computation unit 8 for the preferred embodimentaccording
`to the present invention.
`The photometric computation device 8 is configured to
`include a spectral analysis unit 81 which implements a
`discrete two-dimensional Fourier transform operation. The
`discrete two-dimensional Fourier transform operation is a
`common and well-known technique which by way ofthe
`present invention can now be applied to determining bright-
`ness distribution of a photographic subject from a photo-
`metric sensor.
`In the context of the present
`invention,
`however, the two-dimensional Fourier transform is used to
`obtain a power spectrum of the spatial frequencies of a
`photographic subject. A multi-pattern photometric unit 82
`performs multi-pattern photometry using the power spec-
`trum of the spatial frequency of the photographic subject and
`also outputs an exposure control value.
`A Fourier coefficient “a” of the cos componentof the
`two-dimensional spectrum of the photometric output B[x,y]
`and a Fourier coefficient “b” of the sin component are
`described as follows:
`
`DC=P(0, 0]
`
`LO=
`
`Zs; Le Pin, m] — DC
`
`P[n, m]-DC +1#m[5]
`
`Where:
`
`Zs
`
`(4)
`
`{5]
`
`expresses a sum from n=0 to 2; and
`
`Xs
`
`65
`
`expresses a sum from m=0 to 2.
`
`HI = 4 Eg Pin, m] — LO - DC
`
`[6]
`
`HUAWEI EX. 1212 - 7/10
`
`Canon Exhibit 1212
`Page 7
`
`Canon Exhibit 1212
`Page 7
`
`HUAWEI EX. 1212 - 7/10
`
`

`

`5,592,256
`
`Where:
`
`x,
`
`expresses a sum from n=0 to 2; and
`
`Lp
`
`expresses a sum from m=0 to 2.
`Referring now to FIG. 5, therein depicted is a diagram
`which illustrates classification of a photometric subject.
`Classification is performed using the above-mentioned
`direct current component DC, low-frequency component LO
`and high-frequency component HI.
`In FIG. 5, the photographic subjectis classified into a total
`of twelve (12) patterns, with three (3) patterns from the size
`of the direct current component DC, two (2) patterns from
`the size of the low-frequency component LO, and two (2)
`patterns from the size of the high-frequency component HI.
`Referring now to FIGS. 6-8, therein depicted are line
`graphs whichillustrate the membership function for obtain-
`ing a degree K[n] of conformity between the photographic
`subject and a spectral pattern depicted in FIG. 5. The degree
`of conformity is determined with reference to the direct
`current component DC, the low-frequency component LO
`and the high-frequency component HI.
`FIG. 6 shows membership function DCSL(DC)such that
`the direct current component DC of a photographic subject
`expresses a small degree of conformity. Membership func-
`tion DCMD(DC) expresses a medium degree of conformity
`and membership function DCBG(DC) expresses a large
`degree of conformity.
`FIG. 7 shows membership function LOSL(LO)such that
`the low-frequency component LO ofthe subject expresses a
`small degree of conformity and membership function LOB-
`G(LO)expresses a large degree of conformity.
`FIG. 8 shows membership function HISL(HI) such that
`the high-frequency componentHI ofthe subject expresses a
`small degree of conformity and membership function HIB-
`G(HI) expresses a large degree of conformity.
`By using the membership functions discussed above, the
`degree K[n] to which there is conformance between the
`photographic subject and the twelve (12) spectral patterns in
`FIG. 5 is obtained as follows:
`Where: Min(n,, no, ..., Ny) ate coefficients which return the
`minimum value from among n,, N2,..., Ny.
`
`K[1}=Min(DCSL@C), LOSL(LO), HISLGH))
`
`K{2}=Min(DCMD(DC), LOSL(LO), HISL(HD)
`
`K[3]=Min(DCBG@C), LOSL(LO), HISLCI))
`
`K[4]=Min(DCSL@C), LOBG(LO), HISLGHW)
`
`K[5]=Min(DCMD(DC), LOBG(LO), HISL())
`
`K(6}=Min(DCBG(DC), LOBG(LO), HISL(HD)
`
`K[7|=Min(DCSL(DC), LOSL(LO), HIBG(HD)
`
`K[8}=Min(DCMD(DC), LOSL(LO), HIBG(HI))
`
`K[9]=Min(DCBG(DC), LOSL(LO), HIBG(HD)
`
`K[10]=Min(DCSL(DC), LOBG(LO), HIBG(HD))
`
`K[11]=Min(DCMD(DC), LOBG(LO), HIBG(HD)
`
`K[12]=MinDCBG(C), LOBG(LO), HIBG(HD)
`
`[7}
`
`[8]
`
`19]
`
`[10]
`
`(11)
`
`(12)
`
`(13)
`
`[14]
`
`[15]
`
`[16]
`
`(17}
`
`[18]
`
`6
`patterns shownin FIG.5 using the direct current component
`DC, the low-frequency component LO and the high-fre-
`quency component HI. Moreover, BP[n] is computed as
`follows:
`
`BP{nJ=W[n, 1]-DC+W{n, 2]-LO+WIn, 3)-H+-C[n}
`
`[19}
`
`Where: n is a number from 1 to 12.
`In equation [19], W and C are weighting coefficients.
`Preferably, weighting coefficients are chosen from those
`which have been studied and optimized through analysis of
`subject patterns extracted in advance.
`An exposure control value BO is obtained based on
`equation [20] (listed below) by taking a weighted mean of
`BP[n] using the degree of conformance K{n]:
`
`BO = EZ, (Kin] - BP[n])/Zy Kin]
`Where:
`
`23
`
`expresses a sum from n=0 to 12; and
`
`X5
`
`[20}
`
`is
`
`expresses a sum from n=0 to 12.
`Referring now to FIG. 9, therein depicted is a flow chart
`whichoutlines the operation of the photometric device of the
`present embodiment. Moreover,
`the flowchart shows the
`operational flow from when photometric value B is obtained
`to when the exposure control value BO is obtained. It is to
`be understoodthat the operational flow depicted in FIG. 9 is
`to be carried out by photometric computation unit 8 using
`well-known software and computer programming concepts
`and constnicts.
`In step sl, photometry is performed and B[x, y]
`obtained.
`In step s2, Fourier coefficient a of the cos component of
`the two-dimensional spectrum of the photometric output
`B[x, y] is computed, and in step s3, Fourier coefficient b of
`the sin component is computed.
`In step s4, the power spectrum P[n, m] is computed from
`Fourier coefficient a and Fourier coefficientb.
`In step s5, the direct current component DC is obtained
`from the power spectrum P[n, mJ]. In step s6,
`the low-
`frequency component LO is obtained.
`In step s7, the high-frequency componentHI is obtained.
`Steps s8 through s19 obtain the degrees K[n] of conform-
`ance between a photographic image and the twelve (12)
`spectral patterns using the direct current component DC,the
`low-frequency component LO,the high-frequency compo-
`nent Hi, and membership functions DCSU DCMD, DCBG,
`LOSL, LOBG, HISL and HIBGas such were described
`above.
`In step s20, BP[n] is calculated using the direct current
`component DC, the low-frequency componentLO,the high-
`frequency component HI, and coefficients W and C as such
`were described above.
`In step s21, the weighted mean of BP[n] is taken using the
`degree of conformance K[n] to obtain the exposure control
`value BO, thereby completing the operational flow.
`In the present embodiment, only the power spectra were
`used in the spectral pattern classification, but it is also
`possible to classify using the respective frequency phases.
`Where respective frequency phases are used, the photo-~
`graphic subject can be even more accurately classified.
`The phases © at the respective frequencies are as follows:
`
`25
`
`30
`
`35
`
`40
`
`45
`
`50
`
`55
`
`60
`
`65
`
`The operation expression indicated by equation [19]
`(listed below)is prepared for each ofthe twelve (12) spectral
`
`Q[n, m]=tan™ (b[n, m]/a[n, m])
`
`[21]
`
`HUAWEI EX. 1212 - 8/10
`
`Canon Exhibit 1212
`Page 8
`
`Canon Exhibit 1212
`Page 8
`
`HUAWEI EX. 1212 - 8/10
`
`

`

`5,592,256
`
`7
`For example, ©[1, 1] expresses the phase of the wave of
`the first cycle in the vertical and the horizontal directions,
`and if @[1, 1] is 0 degrees, it will be observed that the center
`section is dark and in a rear-lit condition. If ©f[1, 1] is 180
`degrees, it will be observed that the center section is bright
`and in a front-lit condition.
`The description above indicates that if classification is
`performed using O[1, 1], it is possible to classify the subject
`into front-lit and rear-lit conditions. In this way, spectral
`classification is not limited to the methods described above.
`To the contrary, if classification is done using the power
`spectrum and the phase, it is possible to determine the
`exposure of the subject more accurately.
`Moreover, the spectral analysis device should be under-
`stood as limited to use of the Fourier transform. To the
`contrary, other spectral analysis methods may also be used.
`For example, a subject can be classified using spatial fre-
`quency operations, or even a discrete cosine transform
`(DCT) which only uses a cosine function. Moreover, a
`discrete sine transform, which only uses a sine function, or
`a Walsh transform, which uses a Walsh function which is a
`binary function of +1 and —1, may be used. In a Walsh
`transform, in particular, since only the two values of +1 or
`—1 are used,
`it
`is possible to perform spectral analysis
`computationsat a high speed even with a small-scale device.
`The optimum spectral analysis unit can be operated from
`among the above-mentioned methods, but should not be so
`limited. As new and different methods are developed and
`miniaturized such methods may also be used. All that is
`necessary is that a particular method be able to provide
`required performance levels and be capable of implemen-
`tation in a camera’s photometric device.
`Although a preferred embodimentof the presentinvention
`has been shown and described,it will be readily appreciated
`by those of ordinary skill in the art that many changes and
`modifications may be made to the embodiment without
`departing from the spirit or scope of the present invention
`which is defined in the appended claims and in equivalents
`thereof.
`Whatis claimed is:
`1. A photometric device for use in a camera, the photo-
`metric device comprising:
`segmented brightness measuring means for segmenting a
`photographic scene of the camera into multiple areas
`and for outputting corresponding multiple photometric
`values;
`spectral analysis means for performing spectral analysis
`of the spatial frequency of a photographic subject using
`the corresponding multiple photometric values output
`by said segmented brightness measuring means; and
`photometric computation means for performing photo-
`metric computations based on the spectral pattern ofthe
`photographic subject output by said spectral analysis
`means to produce exposure control values.
`2. The photometric device according to claim 1, wherein
`said spectral analysis means performsspectral analysis by
`using a Fourier transformation operation.
`3. The photometric device according to claim 1, wherein
`said segmented brightness measuring means includes a
`plurality of segments corresponding to sections of the pho-
`tographic scene of the camera.
`4. The photometric device according to claim 3, wherein
`said plurality of segments is a power of 2.
`5. A photometric device for use in a camera, the photo-
`metric device comprising:
`a segmented brightness measuring unit segmenting a
`photographic scene of the camera into multiple areas
`
`_ 0
`
`15
`
`20
`
`25
`
`30
`
`35
`
`40
`
`45
`
`50
`
`60
`
`8
`and outputting corresponding multiple photometric val-
`ues;
`
`:
`
`a spectral analysis unit performing spectral analysis of the
`spatial frequency of a photographic subject using the
`corresponding multiple photometric values output by
`said segmented brightness measuring unit; and
`a photometric computation unit computing exposure con-
`trol values based on the spectral pattern of the photo-
`graphic subject output by said spectral analysis unit.
`6. The photometric device according to claim 5, wherein
`said spectral analysis unit performs spectral analysis using a
`Fourier transformation operation.
`7. The photometric device according to claim 5, wherein
`said segmented brightness measuring unit includes a plural-
`ity of segments corresponding to sections of the photo-
`graphic scene of the camera.
`8. The photometric device according to claim 7, wherein
`said plurality of segments is a power of 2.
`9. The photometric device according to claim 8, wherein
`said plurality of segments comprises 64 segments arranged
`rectangularly.
`10. A photometric device for use in a camera, the photo-
`metric device comprising:
`a segmented brightness measuring unit segmenting a
`photographic scene of the camera into multiple areas
`and outputting corresponding multiple photometric val-
`ues; and
`a photometric computation unit computing exposure con-
`trol values based on a spatial frequency distribution of
`said corresponding multiple photometric values.
`11. The photometric device according to claim 10, further
`comprising:
`a spectral analysis unit performing spectral analysis of the
`spatial frequency of a photographic subject using the
`corresponding multiple photometric values output by
`said segmented brightness measuring unit to produce
`spectral output values, said photometric computation
`unit computing said exposure control values based on
`said spectral output values.
`12. The photometric device according to claim 11,
`wherein said spectral analysis unit performs spectral analy-
`sis using a Fourier transformation operation.
`13. The photometric device according to claim 11,
`wherein said spectral analysis unit performs spectral analy-
`sis using a discrete cosine transform (DCT) operation.
`14. The photometric device according to claim 11,
`wherein said segmented brightness measuring unit includes
`a plurality of segments corresponding to sections of the
`photographic scene of the camera.
`15. The photometric device according to claim 14,
`wherein said plurality of segments is a power of 2.
`16. A photometric device for use in a camera, the photo-
`metric device comprising:
`to segment a
`a segmented brightness measuring unit
`photographic scene of the camera into multiple areas,
`and in response, outputting corresponding multiple
`photometric values;
`a spectral analysis unit to perform a spectral analysis of
`the corresponding photometric values to obtain a power
`spectrum of the spatial frequencies of a subject; and
`a multi-pattern photometric unit to determine a direct
`current component, a low-frequency component, and a
`high frequency current component of the power spec-
`trum, to obtain degrees of conformance between the
`photographic scene and a plurality of spectral patterns
`using the direct current, low-frequency and high-fre-
`
`HUAWEI EX. 1212 - 9/10
`
`Canon Exhibit 1212
`Page 9
`
`Canon Exhibit 1212
`Page 9
`
`HUAWEI EX. 1212 - 9/10
`
`

`

`5,592,256
`
`9
`quency components and a plurality of membership
`functions, and to obtain an exposure control value by
`taking a weighted mean using the degrees of conform-
`ance.
`
`17. The photometric device according to claim 16,
`whercin the spectral analysis unit performs the spectral
`analysis by using a Fourier transform operation.
`18. The photometric device according to claim 16,
`wherein the spectral analysis unit performs the spectral
`analysis by using a discrete cosine transform operation.
`
`10
`19. The photometric device according to claim 16,
`wherein the spectral analysis unit performs the spectral
`analysis by using a discrete sine transform operation.
`20. The photometric device according to claim 16,
`wherein the spectral analysis unit performs the spectral
`analysis by using a Walsh transform operation.
`
`HUAWEI EX. 1212 - 10/10
`
`Canon Exhibit 1212
`Page 10
`
`Canon Exhibit 1212
`Page 10
`
`HUAWEI EX. 1212 - 10/10
`
`