§
`
`
`
`a
`co
`~
`™La7Pi
`hee|~“e
`ae!Aeg~~ha
`co
`4A
`a
`‘w
`
`.‘7h2@|fayya"rm
`
`APPL-1013 / Page 1 of 16
`Apple Inc. v. YU et al.
`
`

`

`
`
`CCD ARRAYS, CAMERAS,
`and DISPLAYS
`Second edition
`
`Gerald C. Holst
`
`Copublished by
`
`JCD Publishing
`2932 CoveTrail
`Winter Park, FL 32789
`
`7h
`
`SPIE OpticAL ENGINEERING PREss
`A Publication of SPIE—The International Society for Optical Engineering
`Bellingham, Washington USA
`
`APPL-1013 / Page 2 of 16
`APPL-1013 / Page 2 of 16
`
`

`

`Library of Congress Cataloging-in-Publication Data
`
`Holst, Gerald C.
`CCD arrays, cameras, and displays/ Gerald C. Holst. -- 2nd ed.
`p. cm.
`Includes bibliographical references (p.
`ISBN 0-9640000-4-0 (hardcover)
`1. Charge coupled devices.
`2. Information display systems.
`I. Title
`TK7871.99.C45H65
`621.36°7--de21
`
`) and index.
`
`1998
`
`98-10770
`CIP
`
`Copublished by
`
`JCD Publishing
`2932 Cove Trail
`Winter Park, FL 32789
`Phone: 407/629-5370
`Fax: 407/629-5370
`ISBN: 0-9640000-4-0
`
`SPIE - The International Society for Optical Engineering
`P.O. Box 10
`Bellingham, WA 98227-0010
`Phone: 360/676-3290
`Fax: 360/647-1445
`WWW: http://www.spie.org
`ISBN: 0-8194-2853-1
`
`Notice:
`Reasonable efforts have been made to publish reliable data and information,
`but the Author and Publishers cannot assume responsibility for the validity of
`all materials or the consequences of their use.
`
`Copyright © 1998 Gerald C. Holst
`
`All rights reserved. No part of this book may be reproduced in any form by any
`means without written permission from the copyright owner.
`
`APPL-1013 / Page 3 of 16
`APPL-1013 / Page 3 of 16
`
`

`

`The second edition is dedicated to
`
`Emily, Irma, and Deanne
`
`APPL-1013 / Page 4 of 16
`APPL-1013 / Page 4 of 16
`
`

`

`CCD ARRAYS, CAMERAS,
`and DISPLAYS
`Second edition
`
`APPL-1013 / Page 5 of 16
`APPL-1013 / Page 5 of 16
`
`

`

`TABLE of CONTENTS
`
`1
`1. INTRODUCTION .... 2. se ee ee ee ee ee ee eee ee oe
`2
`1.1. Solid State Detectors... 56 66 ce pe ee ee eee ee
`3
`1.2. Imaging System Applications ... 1... 2-2 see e eee eee
`6
`1.2.1. General Imagery . 2... 226s eee eee eee eee
`7
`1.2.2, Machine Vision ......- 6 eee eee ee eee
`1.2.3. Scientific Applications... 6... ee ee eee 8
`1.2.4. Military Applications
`... 6.6 ee eee eee eee 8
`1.3. Configurations... 0.1 eee ee eee 9
`1.4. Image Quality © o.0c06s we be eee sie eee ee eee 12
`1.5. Pixels, Datels, Disels, and Resels .......-5--- +e sues
`14
`1.6. References
`. 66 os ga ea be ee ee ee ee ee 17
`
`18
`2. RADIOMETRY and PHOTOMETRY ........5555+55 52055
`9.1. Radiative Transfer 2... 1 2c ee ee ee es 19
`2.2. Planck’s Blackbody Law . 1... eee eee eee eee es 21
`2.3, Photometry fice ha cies we SMR Se ee Re Ma ee ee 23
`V3.1, Units on. ce we eee ee ee ee eee ne ee 25
`2.3.2. Typical Illumination Levels «2... - 20s eee eres 27
`Fd SACP ce aeecateceack Rw EXEC eR ee Ee RTS 29
`2.4.1. Calibration Sources 0... 05.2 bee ee ee ees 29
`44.9. Real SOUS:
`4 ks ceed woe ee ere ee oe rece 30
`9.5. Camera Formula .....-2-- cc eee eee ee eee ene 33
`92.6. Normalization 2... 0c ce ee ee ee 37
`2.7. Normalization Issues .. 1... eee ee ee ee ees 39
`7 References: 6 6 hele he BENS oe ewer. EOE RUS 44
`
`45
`3. SOLID STATE ARRAYS 200 ok ois ee eee be eee He Ome
`3.1. Photodetection ¢. 26 6 e665 ee eee ee ee ee ee ees 46
`3.1.1. Photogate 2... ce eee eee eens 47
`3.1.2. Photodiode: 2 2068 wie ee See ee ese ee oe ee
`47
`3.2. CCD Array Operation ... 0.0 ee eee eee 47
`3.3. CCD Array Architecture 2.6.0.0 ee ee ee eee 58
`3.3.1. Limear ATTuyS 2.6. ee ee eee ees 58
`3.3.2. Full Frame Arrays .. 6.6620 eee eee eee 59
`3.3.3. Frame Transfer ... 6c ee ee ee eee ee ee
`61
`3.3.4. Interline Transfer... 1... ee es 63
`3.3.5. Progressive Scan 2... eee ee ee ees 68
`3.3.6. Time Delay and Integration ....- +++ e+e ree eres
`70
`3.4. Charge Transfer Efficiency ..... 666s eee eee eee ees
`74
`
`Xill
`
`APPL-1013 / Page 6 of 16
`APPL-1013 / Page 6 of 16
`
`

`

`xiv CCD ARRAYS, CAMERAS, and DISPLAYS
`
`3.5. Charge Conversion (Output Structure) ... 6.625 +e ee eee 78
`3.6. Dark Current
`... 0... eee ee ee ee ee eee es 79
`29 Dark Pikely
`34 86 GONE ET eK Rs UN Gare e esi ws
`83
`3.8, Antibloom Draft
`aoc eavecwrsacee es ee eee mene we eee ee
`85
`BS OT!
`ececaverene ace ecm em epee eoe enka bis Ore MGS CTE gie Fie Se
`89
`4:10 CMOS gicceelg G8 Be Pidlesiecste alas oe Bw eee en ee HOO
`91
`3.11. Physical Parameters... 2... ee eee etre eer rere eene 93
`3.11.1. Microlemses:
`< 2 oe cee ea we He Deere We we
`93
`3.11.2. Color Filter Arrays i...0060 66 eee hee ee ee 94
`3.11.3. Number of Detectors... .... 00+ eee eee eee 97
`3.11.4. Video Chip Size... ee ee eee as
`98
`19. FREPErenCeS
`discece as aie os wire Fie wrap He ce eee 100
`
`102
`4, ARRAY PERFORMANCE ...... 0000 e eee eee eee ewes
`103
`AA SIGE ose ceeeeie we ee oe ee oe ae ele ee ed
`104
`4.1.1. Spectral Response
`..-- 2. eee eee eee eee
`4.1.2. Responsivity .. 0... 0 eee ees 111
`4.1.3. Minimum Signal ... 1... cee eee ee ee eee 119
`4.1.4, Maximum Signal’... 20. 6 be ee ee see soe em ei es
`120
`4.15. Dynaitic Range. coe ee eee ee ee eee Le wie HE Bs
`121
`APO NGS ech HE SE OR SEME RRA BG tie ete ee om ae
`123
`13-1. SHGUNGISE eecceve wie ay ee Oe Ree oe es 127
`4.2.2. Reset Noise .. 2... eee ee ees 127
`4.2.3, On-chip Amplifier Noise .. 2... 6 ee eee ee eee 128
`4.2.4, Off-chip amplifier Noise... 1. - eee eee ees 130
`4.2.5. Quantization Noise... 1.6... eee eee 130
`4.2.6, Pattern Noise... 6 60 eee ee he ee ee ee 131
`4.2.7. Photon Transfer... 1... 0c eee eee eee ee ees
`133
`4.3. Array Signal-to-Noise Ratio 2... 0.6 eee eee eee
`139
`4.4. Correlated Double Sampling ..... 2.6 eee eee ees
`141
`4:5. Frame Rates... 0 ec he a ie ee Oe Oe ee 142
`4G, Defects:
`< ccncacucsces ce os we ereceese ne OEE 143
`4.7. References
`.. 0... cc ee ee eee ee eee eee eee 144
`
`& CAMERAS: 34 ox caeswarswn Ba ke ee neuen me ree ee 146
`5.1. Camera Operation
`.. 1... ee eee eee eee eee ees 147
`5:9. Video Formats: sss sece Sees aS fe ee ae ee Se cee
`149
`5.2.1. Video Timing .. 2... 06 eee eee ee ee eee
`151
`5.2.2. Broadcast/Non-broadcast Design ......---+5555 155
`5.2.3. Component/Composite Signals... ..-.++-+++ee> 156
`§.2.4. IRE Units
`. 0.2.22 eet ee ee eee eee ees 157
`5.2.5. Digital Television
`. 2.1... 0c cere eee eee 159
`
`APPL-1013 / Page 7 of 16
`APPL-1013 / Page 7 of 16
`
`

`

`TABLE of CONTENTS xv
`
`5.2.6. HDTVIATS ... 00 ccc ee ee be ele ae Oe we we es 160
`5.3. Consumer/Broadcast Cameras
`.. 1-1-0 ee ee ee ee es 163
`5.3.1. The Knee
`. 0... ee ee ee 164
`§ 3.9. Color Correction:
`.6.655 Gao ela eke ie ee Se a 165
`5.3.3. Gamma Compensation... 1... ee eee eee ee ees
`169
`5.3.4. Aperture Correction ©... eee eee eee eee 172
`5.4. Industrial/Scientific Cameras ... 2.06 eee eee ee ees 172
`5.4.1. Analog-to-Digital Converters... 6.6 ee ee ener
`174
`5.4.2. Intensified Solid State Cameras
`.. 1... eee ee nee
`176
`5.5. ReferenceS 2.0.66 cc ck ee ee ee ee ee es 179
`
`6. CAMERA PERFORMANCE .......- ee eee eee eres 182
`6.1, Nomenclature:
`< 56 Gcice eee See win ee ie ie eee oe 8 183
`6.2. Camera Metrics. <.ces: aac ee ee ee ae ee te ee 185
`6.2.1. Camera SNR ... 0.25 eee ee ee ee ee eee
`186
`6.2.2. Camera Dynamic Range
`...-.-- eee ee ees 187
`6.2.3. Maximum Signal .. 0.6 ee ee es 189
`6.2.4. Minimum Signal
`.. 2... 6-50 ee eee eee ees 189
`6.3. Intensified CCD Camera... 2-2 eee ee 195
`6:3.1; ICCD Signal: «os oa we x ere sce eee et 196
`6.3.2. ICCD Noise .... 6. eee ee ee eee 197
`6:4.3..TOCD SNR fs ic bie bos See ee oe owe eee erence
`198
`GA References
`cig eceecesd wu wea woe eee Ee eK ne ee ee Ee TCS
`200
`
`7. CRT-BASED DISPLAYS «2.0006 ce ee ee ee eee es 201
`7 4 "The ODSELVEL cece eee ew ere ee eon ee ee ee Ee ee 203
`7%. CRT QVGGVIEW ccc ee Oe ee ee Se le ee 205
`7.2.1. Monochrome Displays... . 2-2-0 + eee ere eee 207
`7.2.2. Color Displays ....- ee ee ese eet eee nnees
`207
`IIIT ce wie we AAD} lle Woe TE WR 210
`7.3. Spot Size 4 64 05 84 oy dew cree Ge eee 210
`Pa FRB es os eek KR dee MR EER Ee ue eee re Bs
`213
`7.5. Resolution ......00 0 ee ee ee ee eee eee ee ees 218
`7 5.1. Vertical Resolution... 6-60 ee eee eee ee ne ees
`219
`7.5.2. Theoretical Horizontal Resolution..........---++-.
`219
`7.5.3. TV Limiting Resolution .... 6... 502+ ee eee ees
`220
`9-8 0 WATER cocci eae eee a oe Be HE ee ee ee 221
`7.6. Addressability 2.0... 02 cece een eee eee ne ee ne es
`222
`7:7, Shades Of Gray 0.6050 6s ee we SG ke ew ee Oe 227
`7.8. Character Recognition .. 0... eee eee eee 227
`79. Contrast... .. ee eee ee ee ee 228
`TAY References o. is GRA ED PRON ae eee ee eae se aE
`230
`
`APPL-1013 / Page 8 of 16
`APPL-1013 / Page 8 of 16
`
`

`

`SOLID STATE ARRAYS 93
`
`attractive feature for man-portable, battery-operated devices. It appears at this
`time, CMOSwill compete with CCDsin the general video marketplace where
`weight, size, and power consumption are factors.
`
`3.11. PHYSICAL PARAMETERS
`
`All detector arrays are fabricated for specific applications. Microlenses are
`used to increase the optical fill factor. Color filter arrays employ filters that are
`placed over
`the detectors to create specific spectral
`responses that
`lend
`themselves to color imagery. For general video applications, the number of
`detectors is matched to the video standard to maximize bandwidth and image
`sharpness. The overall array size follows the convention used with vidicons.
`
`3.11.1. MICROLENSES
`
`Optical fill factor may be less than 100% due to manufacturing constraints
`in full transfer devices. In interline devices, the shielded vertical transfer register
`can reduce the fill factor to less than 20%. Microlens assemblies (also called
`microlenticular arrays or lenslet arrays) increase the effective optical fill-factor
`(Figure 3-42). But it may not reach 100% due to slight misalignment of the
`microlens assembly, imperfections in the microlens itself, nonsymmetric shielded
`areas, and transmission losses. As shown by the camera formula (Equation 2-16,
`page 35), the output is directly proportional to the detector area. Increasing the
`optical fill factor with a microlens assembly increases the effective detector size
`and, therefore, the output voltage.
`
`The photosensitive area is below the gate structure and the ability to collect
`the light depends upon gate thickness. The cone oflight reaching the microlens
`depends upon the f-number of the primary camera lens. Figure 3-42 illustrates
`nearly parallel rays falling on the microlens. This case is encountered with high
`f-number lens systems. Low f-number primary camera lenses increase the cone
`angle and the effective fill-factor decreases with decreasing f-number.”’
`Microlenses are optimized for most practical f-numbers. As the array size
`grows, off-axis detectors do not obtain the same benefit as on-axis detectors.”
`
`Hyper-HADis a Sony trademark whichindicates the presence of a microlens
`assembly over a hole accumulation diode. While many camera manufacturers
`use the Hyper-HAD detectors,
`they do not identify the detector by Sony’s
`trademark. This leads the user to believe that there are many different detectors
`and hence many different cameras on the market.
`
`APPL-1013 / Page 9 of 16
`APPL-1013 / Page 9 of 16
`
`

`

`94 CCD ARRAYS, CAMERAS, and DISPLAYS
`
`(a)
`
`wa
`
`(a) With no
`Figure 3-42. Optical effect of a microlens assembly.
`microlens, a significant amount of photon flux is not detected. (b) The
`microlens assembly can image nearlyall the flux onto the detector when
`a high f-numberlens is used. These lenslets can either be grown on the
`array during the fabrication process or manufactured out of a material
`such as quartz and placed on the array surface during packaging.
`
`3.11.2. COLOR FILTER ARRAYS
`
`The subjective sensation of color can be created from three primary colors.
`By adjusting the intensity of each primary (additive mixing), a full gamut
`(rainbow) ofcolors is experienced. This approach is used onall color displays.
`They have red, green, and blue phosphors that, when appropriately excited,
`produce a wide spectrum of perceived colors, The CIE committee standardized
`a color perception model for the human observerin 1931. Theliterature” is rich
`with visual data that form the basis for color camera design.
`
`APPL-1013 / Page 10 of 16
`APPL-1013 / Page 10 of 16
`
`

`

`SOLID STATE ARRAYS 95
`
`The "color" signals sent to the display must be generated by three detectors,
`each sensitive to a primary or its complement. The primary additive colors are
`red, green, and blue (R, G, B) and their complementary colors are yellow, cyan
`and magenta (Ye, Cy, Mg). For high quality color imagery,
`three separate
`detectors (discussed in Section 5.3.3., Color Correction) are used whereas for
`consumer applications, a single array is used. The detectors are covered with
`different filters that, with the detector response, can approximate the primaries
`or their complements (Figure 3-43). A single array with filters is called a color
`filter array (CFA).
`
`Response
`
`
`
`RelativeResponse
`
`Relative
`
`Wavelength (um)
`
`(b)
`
`Figure 3-43. Desired spectral response for the three primaries and their
`complements. (a) The primaries and (b) their complements.
`
`APPL-1013 / Page 11 of 16
`APPL-1013 / Page 11 of 16
`
`

`

`96 CCD ARRAYS, CAMERAS, and DISPLAYS
`
`While CFAscan havefilters with the appropriate spectral transmittance to
`create the primaries, it is more efficient to create the complementary colors.
`Complementary filters have higher transmittances than the primary filters. The
`luminance channel
`is identical to the green channel and is often labeled Y.
`"White" (no colorfilter) is represented by W=R+G+B.
`
`A linear set of equations relates the primaries to their complements. These
`equations are employed (called matrixing) in cameras to provide either or both
`output formats. Matrixing can convert RGB into other color coordinates:
`Ye=- R+G-W-B
`
`Mg - R+B-W-G
`
`(3-12)
`
`Cy =-G+B-W-R.
`The arrangement of the color filters for a single array system is either a
`stripe or a mosaic pattern (Figure 3-44), The precise layout of the mosaic varies
`by manufacturer. One basic CFA patent” was granted to Bryce E, Bayerat
`Eastman Kodak in 1976.
`
`
`
`er|verve
`
`er[veorfve
`
`Mosaics
`
`Figure 3-44. Representative stripe and mosaic arrays. Although shown
`as a full frame device for clarity, they typically are interline transfer
`devices. The layout depends upon the manufacturer’s philosophy and
`cleverness in reducing color aliasing.
`
`APPL-1013 / Page 12 of 16
`APPL-1013 / Page 12 of 16
`
`

`

`SOLID STATE ARRAYS 97
`
`In many sensors, the number of detectors devoted to each coloris different.
`The basic reason is that the human visual system (HVS) derives its detail
`information primarily from the green portion of the spectrum. That
`is,
`luminance differences are associated with green whereas color perception is
`associated with red and blue. The HVS requires only moderate amounts of red
`and blue to perceive color. Thus, many sensors have twice as many green as
`either red or blue detector elements. An array that has 768 horizontal elements
`may devote 384 to green, 192 to red, and 192 to blue. This results in an unequal
`sampling of the colors. A birefringent crystal (discussed in Section 10.3.4.,
`Optical Anti-alias Filter), inserted between the lens and the array, accommodates
`the different sampling rates (discussed in Section 8.3.2., Detector Array Nyquist
`Frequency). The video signal from the CFA is embedded in the architecture of
`the CFA. The color video from a single chip CFA must be decoded
`(unscrambled) to produce usable R, G, and B signals.
`
`3.11.3. NUMBER OF DETECTORS
`
`Each detector array is designed for a specific application (discussed in
`Section 5.2.2., Broadcast/Non-Broadcast Design) The EIA 170 video standard
`supports 485 lines (discussed in Section 5.2.1., Video Timing). However, one
`line is split between the even and odd fields so that
`there are only 484
`continuous lines. Thus detector arrays designed for EIA 170 compatibility tend
`to have 484 detectors in the-Vertical direction. For convenience, this has been
`reduced to 480 detectors. Image processing algorithms are more efficient with
`square pixels. With a 4:3 aspect ratio, the desired numberof detectors is 640 x
`480. Less expensive arrays will have a submultiple (for easy interpolation) such
`as 320 x 240.
`
`For video applications, several rows or columns are devoted to dark current
`and for light leakage. Therefore an array may be 650 x 492 but
`the light
`sensitive part may be 640 x 480.
`It
`is manufacturer dependent whether to
`specify the array size by the numberof active pixels or the total number of
`pixels (which includes the dark pixels).
`
`images. To avoid beat
`frequencies are very obvious in color
`Beat
`frequencies,
`the color pixel rate should be a multiple of the chrominance
`subcarrier frequency (3.579545 MHz). This results in arrays that contain 384,
`576, or 768 horizontal detectors. While CFAsare designed for NTSC operation,
`the same chip can be used for monochrome video without the CFA. Therefore,
`many monochromearrays also contain 384, 576, or 768 horizontal detectors.
`These arrays do not have square pixels. But, from a manufacturing point-of-
`view, this reduces the inventory of arrays offered.
`
`APPL-1013 / Page 13 of 16
`APPL-1013 / Page 13 of 16
`
`

`

`98 CCD ARRAYS, CAMERAS, and DISPLAYS
`
`Scientific array size tends to a power of 2 (e.g., 512 x 512, 1024 x 1024)
`for easy image processing. There is a perception that "bigger is better" both in
`terms of array size and dynamic range. Arrays may reach 8192 x 8192 with a
`dynamic range of 16 bits. This array requires (8192)(8192)(16) or 1.07 Gbits
`of storage for each image. Image compression schemes may be required if
`storage space is limited. The user of these arrays must decide which imagesare
`significant and through data reduction algorithms, store only those that have
`value, Otherwise, he will be overwhelmed with mountains of data.
`
`While large format arrays offer the highest resolution, their use is hampered
`by readoutrate limitations. For example, consumer camcorder systemsoperating
`at 30 frames/s have a data rate of about 10 Mpixels/s. An array with 5120 x
`5120 elements operating at 30 frames/s has a data rate of about 768 Mpixels/s.
`Large arrays can reduce readout rates by having multiple parallel ports servicing
`subarrays. Each subarray requires separate vertical and horizontal clock signals.
`The tradeoff is frame rate (speed) versus number of parallel ports (complexity
`of CCD design) and interfacing with downstream electronics. Because each
`subarray is serviced by different on-chip and off-chip amplifiers, the displayed
`image of the subarrays may vary in contrast and level. This is due to differences
`in amplifier gains and level adjustments.
`
`3.11.4. VIDEO CHIP SIZE
`
`television
`Vidicon vacuum tubes were originally used for professional
`applications. These were specified by the tube diameter. To minimize distortion
`and nonuniformities within the tube,
`the recommended*!
`image size was
`considerably less than the overall tube diameter. When CCDsreplaced the tubes,
`the CCD industry maintained the image sizes but continued to use the tube
`format nomenclature (Table 3-2).
`
`Table 3-2
`ARRAYSIZE for STANDARD FORMATS
`
`STANDARDIZED ARRAY SIZE
`(H x V)
`
`3.2 mm X 2.4 mm
`
`64mm x 4.8 mm
`
`1/4 inch
`
`APPL-1013 / Page 14 of 16
`APPL-1013 / Page 14 of 16
`
`

`

`SOLID STATE ARRAYS 99
`
`Although each manufacturer supplies a slightly different array size and pixel
`size, nominal sizes for a 768 x 480 array are given in Table 3-3. With interline
`transfer devices, approximately one-half of the pixel width is devoted to the
`shielded vertical transfer register. That is, the detector active width is one-half
`of the pixel width. Thus the active area of the pixel is rectangular in interline
`transfer devices. This asymmetry does not appear to affect image quality in
`consumer video products significantly.
`
`Table 3-3
`NOMINALPIXEL SIZE for a 768 x 480 ARRAY
`Sizes vary by manufacturer. Detector sizes are smaller.
`
`FORMAT
`
`
`
`
`
`
`
`
`
`
`
`
`a.ee
`(H x V)
`
`
`
`The decrease in optical format is related to cost. The price of CCD arrays
`is mainly determined by the cost of processing semiconductor wafers. As the
`chip size decreases, more devices can be put on a single wafer and this lowers
`the price of each individual device. The trend of going to smaller devices will
`probably continue as long as the optical and electrical performance of the
`imagers does not change. However, smaller pixels reduce the charge well size.
`For a fixed flux level and lens f-number,
`the smaller arrays have reduced
`sensitivity.
`
`Smaller chips make for smaller cameras. However, to maintain resolution,
`pixels can only be made so small. Here, the tradeoff is among pixel size, optical
`focal length, and overall chip size. Further unless the lens f-number is reduced
`as the chip size is reduced, the system will move from being detector-limited to
`optically-limited (discussed in Section 10.4., Optical-Detector Subsystem). As
`this happens, the system MTF will change and smaller detectors will provide a
`less sharp image.
`
`APPL-1013 / Page 15 of 16
`APPL-1013 / Page 15 of 16
`
`

`

`This completely revised book includes
`OSDeB iconsie OODENSots
`» CID and CMOStechnology
`> Signal-to-noise ratio
`> Minimum illumination definitions
`
`|
`|
`
`|
`
`ONTTSONY0
`and DISPLAYS6 second edition
`4 }
`
`Topics include
`
`SLUEobe:taile
`Frame transfer
`Progressive scan
`Pioutrm netS(oe
`ONS
`Radiometry
`aileCoyinloamy
`
`Sampling and aliasing
`Dark current
`Antibloom drain
`Electronic shutter
`Noise analysis
`Time-delay and integration
`Minimum resolvable contrast
`
`i.
`
`.
`i”
`|
`"
`|
`:
`
`TTeeee(60@MeLt
`
`APPL-1013 / Page 16 of 16
`
`Also by Gerald C. Holst
`Sampling, Aliasing, and Data Fidelity
`Electro-Optical Imaging System Performance
`Testing and Evaluation of Infrared Imaging Systems
`
`CCD Arrays, Cameras, and Displays
`
`second edition
`
`im
`
`JCD Publishing
`LeveeiM ret
`Winter Park, FL 32789
`SNPeee
`
`SPIE PRESS
`Oa
`Bellingham, WA 98227
`ISBN: 0-81294-2853-1
`
`a a
`i”
`2.
`
`i
`
`5
`
`'
`
`rig
`
`APPL-1013 / Page 16 of 16
`
`