EXHIBIT 1034
`
`U.S. PATENT NO. 4,434,441 TO ISHIZAKI
`
`(“ISHIZAKI”)
`
`
`
`
`
`TRW Automotive U.S. LLC: EXHIBIT 1034
`PETITION FOR INTER PARTES REVIEW
`OF U.S. PATENT NUMBER 8,599,001
`IPR2015-00436
`
`

`
`‘United States Patent
`
`[19]
`
`[11]
`
`4,434,441
`
`Ishizaki et al.
`[45] Feb. 28, 1984
`
`
`
`[54] METHOD FOR DRIVING A CHARGE
`INJECI‘ION DEVICE
`
`[56]
`
`References Cited
`U.S. PATENT DOCUMENTS
`
`[75]
`
`Inventors:
`
`I-Iiroyuki Ishizaki, Akashi; Yoshiki-
`Tsujino, Kakogawa; Masaji Dohi,
`Kobe, all of Japan
`
`[73] Assignee:
`
`Fujitsu Limited, Kawasaki, Japan
`
`[21] Appl. No.: 324,070
`
`[22] Filed:
`
`Nov. 23, 1981
`
`.
`.
`.
`.
`.
`Forelgn Application Pnonty Data
`[30]
`Nov. 27, 1980 [JP]
`Japan ................................ 55-167828
`
`Int. Cl.3 ............................................. .. H04N 3/14
`[51]
`
`[52] U.S. Cl.
`. . . . . . . . . . . . . . . .
`. . . . . . . . . .. 358/213
`
`[58] Field of Search ................................ .. 358/213
`
`............. 250/211
`3,801,820 4/1974 Eichelberger et al.
`
`357/24
`3,882,531
`5/1975 Michon et a1.
`...... ..
`4,322,753
`3/1982 Ishihara ............................... 358/213
`
`Primary Examiner——Richard Murray
`Attorney, Agent, or Firm—Daniel J. Tick
`
`[57]
`
`ABSTRACT
`
`A charge injection device sensing an optical radiation
`pattern is driven at an operation sequence where the
`charge holding mode is inserted between the charge
`storing/readout mode and the charge injection mode.
`The charge holding period is effective for picture ele-
`ments influenced by charges injected into the substrate
`after the readout operation placed spacially far and
`timed. The problem of cross-talk at high speed opera-
`tion is resolved.
`
`6 Claims, 4 Drawing Figures
`
`7/
`
`A/ae/zm/M2 63¢/Ff /€66/5752 M
`
`72 ‘
`
`73
`
`
`
`£1567/20A//C
`SW/7C# 70/
`
`//01?/Z0/VTAA
`P/C72/’P£'
`£11.‘-7!//é’/V7‘
`C04 Z//$4/V
`
`Cl/IP65
`SEIVSI77V5
`
`
`
` V6.67/C44
`E14-‘Cree/v/c
`_ 74
`75-
`P/c/wee
`SW/74/Z/6
`(W)
`EAEMENT
`/‘Z )
`C01/44/V
`
`1034-O01
`
`1034-001
`
`

`
`U.S. Patent
`
`Feb. 28, 1584
`
`Sheet 1 of4
`
`4,434,441
`
`F/6./
`PA’/0/? 437'
`
`//ae/za/V744 5///rr,e£6/sr.a_-‘R /0
`
`P5557
`5/4//76//9
`
`£4E6790/V/C
`Swxrax //
`
`Veer/544
`az/.5 X,
`
`V£A’7/€44
`5/9/F7
`P561575?20
`
`
`
`C//M65
`SE/VS/7/I/E
`I/5\_
`I 4/MPUF/5.? 7
`-_-
`7'
`I
`‘ P075/V7‘/All WELL 3b
`1
`,
`P075/Y7‘/AL wen 3a.
`'
`35/us//va sue/74:6 5a
`- E I C//A26! 4 4
`u
`E 1
`54.
`
`
`
`I I II:
`
`£z.~:c7;ea«v/c_
`sh//raw 3,
`
`1034-002
`
`1034-002
`
`

`
`U.S. Patent
`
`Feb. 28, 1984
`
`?Sheet 2 of4
`
`4,434,441
`
`F/6.2
`
`7/
`- rm)
`
`A/a.e/zaxv/'44 .54//7 1956/5752 /0
`- me)
`- <¢°z>
`— C/MRGE
`
`£2 ECTROA//C
`.5»//7c.4 70/ I
`
`//0.?/Zfl/I/7,44
`,0/trues
`£15/9/E/V7
`C04 0544/
`
`
`
`I
`I
`|
`I?/5 /‘" "‘
`E£5[74?0/V/C
`
`‘X16
`
`0
`
`Z
`
`Vfkr/(.44
`
`COLV/1’?-/’1"’
`
`3/6
`
`.
`
`1034-O03
`
`1034-003
`
`

`
`U.S. Faterit
`
`Feb. 28, 1984
`
`Sheet 3 of4
`
`4,434,441
`
`F/6.3
`
`1034-O04
`
`1034-004
`
`

`
`U.S. Patent
`
`Feb. 28, 1984
`
`Sheet 4 of4
`
`4,434,441
`
`1034-O05
`
`1034-005
`
`

`
`1
`
`4,434,441
`
`METHOD FOR DRIVING A CHARGE INJECTION
`DEVICE
`
`BACKGROUND OF THE INVENTION
`
`1. Field of the Invention
`
`The present invention relates to a method for driving
`a charge injection device. More particularly, the inven-
`tion relates to an improved method for driving a charge
`injection device, hereinafter referred to as a CID,
`which is used as an image sensing device for bi-dimen-
`sional images.
`2. Description of Prior Arts
`A CID is known as a solid state imaging device
`which reads out corresponding electrical image signals
`by sensing the optical radiation pattern.
`A known CID has, in general, picture elements, each
`of which is composed of a pair of insulating electrodes,
`arranged bi-dimensionally on the semiconductor sub-
`strate which is covered with insulating film at the sur-
`face. Corresponding electrodes of each row are con-
`nected in common to vertical X buses and other corre-
`sponding electrodes of each column are connected in
`common to the horizontal Y buses. When an optical
`image is projected onto the sensing surface of the matrix
`under the condition that a voltage of about 10 V is
`applied to the one electrode 2X of picture elements and
`a voltage of about 20 V to the other electrode, the
`charges 4 in the semiconductor substrate generated by
`the optical activation are stored in the potential well 3b
`formed just under the electrode 2Y. Thereafter, said
`charges 4 are shifted at a time for each picture element
`to the well 3a just under the adjacent electrode 2X in
`such a timing that the wells just under the electrode 2Y
`connected to the selected Y bus are disappeared at a
`time by the voltage operation from the shift register 20
`in the vertical direction. The condition during such
`process can be seen in the picture element group of 3rd
`stage in FIG. 1. The image charge is generated on the
`electrode 2X at the moment where such charge 4 is
`shifted, and said image charge is sequentially read out to
`the output terminal of the charge sensitive amplifier 7
`via the electronic switches 11, 12, 13, 14. Here, 10 is the
`horizontal shift register; 9 is the reset switch which
`applies a voltage V3 to the electrode 2X of each picture
`element; X1 to X4, Y1 to Y4 are X bus group and Y bus
`group, respectively.
`When readout for one frame of said CID completes,
`the well to which the charges generated by optical
`activation are shifted can be made to disappear by turn-
`ing ON the electronic switches, whereby the charges
`are injected into the substrate and disappear therein.
`The succeeding charges generated by optical activation
`are then stored in the well newly formed just under the
`other electrode preparing for the image sensing of the
`next single frame.
`Here, the other electrode has the function of storing
`signal charges, so it is called the storing electrode, while
`the adjacent electrode has the function of detecting
`signal charges, so it is called the detecting electrode.
`The peripheral circuits as well asfthe charge sensitive
`amplifier of the aforementioned CID may be integrated
`on the same substrate. However, the upper limit of
`frequency response of the charge sensitive amplifier is
`usually set to a frequency as low as 1 MHz, which
`means that it is difficult to read out the signal within a
`period shorter than one microsecond. It is therefore
`impossible to use a high speed drive having the afore-
`
`5
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
`40
`
`45
`
`50
`
`55
`
`60
`
`65
`
`2
`described known structure for a CID where a plurality
`of picture elements are arranged with high density.
`In order to solve such problem,
`it is sufficient to
`sequentially read out in parallel all signals, in the se-
`quence of columns, for example, sent from the picture
`elements arranged in a line in the vertical direction.
`However, it is always very difficult to inject all picture
`element charges into the substrate after completing
`parallel readout in the sequence of columns and store
`the optically activated charges by immediately prepar-
`ing new wells. In other words, if the lifetime of charges
`injected into the substrate is comparatively long such
`as, for example, one microsecond, if the next storing
`operation is started while the charges injected first into
`the substrate have notyet disappeared and remain, such
`remaining charges partly enter the newly formed well
`drastically degrading the readout image. The partial
`entering of the remaining -charges into the new formed
`well is known as the cross—talk phenomenon.
`In order to solve such cross-talk problem, sufficient
`idle time may be prepared for the disappearance of
`injected charges after the stored charges of all the pic-
`ture elements are injected into the substrate. This pre-
`vents the cross-talk perfectly, but results in the frame
`time becoming long due to the setting of idle time, thus
`making the realization of high speed operation difficult.
`The principal object of the invention is to provide a
`method for driving a CID which solves the abovemen-
`tioned problem caused by the injected charges.
`An object of the invention is to provide a method for
`driving a CID which successfully solves the problem of
`residual image and deterioration of resolution resulting
`from the injected charges.
`Another object of the invention is to provide a
`method for driving a CID which realizes high speed
`sequential readout without any cross-talk.
`Still another object of the invention is to provide a
`high speed driving method for a CID which assures
`high quality image signals.
`SUMMARY OF THE INVENTION
`
`In the method of the invention, the basic operation
`sequence, essentially including the operations of charge
`storing, readout and injection, is repeated for each pic-
`ture element considered as the readout unit and the
`' specified charge holding period is set between the read-
`out period and injection period. This results in the real-
`ization of the drive for a charge injection device with-
`out permitting the frame time to become long and pre-
`venting cross-talk phenomenon.
`BRIEF DESCRIPTION OF THE DRAWINGS
`
`In order that the invention may be readily carried
`into effect, it will now be described with reference to
`the accompanying drawings, wherein:
`FIG. 1 is a block diagram of a CID of the prior art;
`FIG. 2 is a block diagram of a preferred embodiment
`of the parallel readout type CID of the invention;
`FIG. 3 is the principal part of a time chart explaining
`the method for driving a CID of the invention; and
`FIG. 4 is the entire time chart explaining the CID
`drive method of the invention.
`
`DETAILED DESCRIPTION OF THE
`PREFERRED EMBODIMENT
`
`The known CID of FIG. 1 has, in general, picture
`elements 1, each of which is composed of a pair of
`
`1034-O06
`
`1034-006
`
`

`
`4,434,441
`
`3
`insulating electrodes 2X and 2Y, arranged bi-dimension-
`ally on the semiconductor substrate which is covered
`with insulating film at the surface. The corresponding
`one of the electrodes 2X of each row are connected in
`common to vertical X buses X1 to X4 and the other
`correspondingelectrodes 2Y of each column are con-
`nected in common to the horizontal Y buses Y1 to Y4.
`When an optical image is projected onto the sensing
`surface 50 of the matrix under the condition that a volt-
`age of about 10 V is applied to the one electrode 2X of
`the picture elements and a voltage of about 20 V is
`applied to the other electrode 2Y, the charges 4 in the
`semiconductor substrate generated by the optical acti-
`vation are stored in the potential well 3b formed just
`under the electrode 2Y. Thereafter, the charges 4 are
`shifted at a time for each picture element to the poten-
`tial well 150 just under the adjacent electrode 2X with
`such timing that the wells just under the electrode 2Y
`connected to the selected Y bus disappear at a time due
`to voltage operation from a shift register 20 in the verti-
`cal direction. The condition during such process is
`shown in the picture element group of the third stage of
`FIG. 1. The image charge is generated on the adjacent
`electrode 2X at the moment that the charge 4 is shifted,
`and said image charge is sequentially read out to the
`output terminal of a charge sensitive amplifier 7 via
`electronic switches 11, 12, 13 and 14. Also provided are
`a horizontal shift register 10 and a reset switch which
`applies a voltage V5 to the electrode 2X of each picture
`element. The X bus group includes X1 to X4 and the Y
`bus group includes Y1 to Y4.
`When readout is completed for one frame of the CID,
`the well 3a to which the charges generated by optical
`activation are shifted can be made to disappear by tum-
`ing ON the electronic switches S1 to S4, whereby the
`charges are injected into the substrate and disappear
`therein. The succeeding charges generated by optical
`activation are then stored in the well newly formed just
`under the electrode 2Y preparing for the image sensing
`of the next single frame.
`Here, the insulated electrode 2Y has the function of
`storing signal charges, so it is called the storing elec-
`trode, while the electrode 2X has the function of detect-
`ing signal charges, so it is called the detecting electrode.
`The peripheral circuits as well as the charge sensitive
`amplifier 7 of the aforedescribed CID may be integrated
`on the same substrate. However, the upper limit of
`frequency response of the charge sensitive amplifier 7 is
`usually set to a frequency as low as 1 MHz, which
`means that it is difficult to read out the signal within a
`period shorter than one microsecond. It is therefore
`impossible to use a high speed drive having the afore-
`described known structure, shown in FIG. 1, for a CID
`where a plurality of picture elements are arranged with
`high density.
`FIG. 2 shows a preferred embodiment of the CID of
`the invention, of 16X 16 picture element parallel read-
`out type where the charge sensitive amplifiers 301 to
`316 are connected to the horizontal buses X1 to X15,
`respectively,
`in view of realizing high speed readout
`operation. In the parallel readout operation, the picture
`element group of vertical one column, enclosed by the
`broken line in FIG. 2, is called the vertical picture ele-
`ment column R1 to R15, while the picture element group
`enclosed by the broken line for horizontal one column is
`called the horizontal picture element column C1 to C16.
`A vertical shift register is not included in the parallel
`readout type CID having the circuit shown in FIG. 2.
`
`l0
`
`15
`
`20
`
`25
`
`30
`
`35
`
`40
`
`45
`
`50
`
`55
`
`60
`
`65
`
`4
`However, a first group of electronic switches 101 to 116 .
`and a second group of electronic switches 201 to 216 are
`provided instead of such shift register. In addition, the
`pulse <]>c2 is applied to the common gate terminal 71 of
`the first group of electronic switches and on the other
`hand, the pulse cbcz, which is the reverse phase pulse of
`4:02 is applied to the common gate terminal 72 of the
`second group of electronic switches. Additionally, the
`power source V1; for biasing the detecting electrode 2X
`is connected to the termiyl 74 via a charge sensitive
`amplifier. When the pulse qbcz becomes high level H, the
`pulse dacz simultaneously becomes low level L. There-
`fore, the high level voltage V11 supplied via the terminal
`74 biases the detecting electrode 2X of each picture
`element via the input terminals of the charge sensitive
`amplifiers 301 to 316 and the electronic switches 201 to
`216 in the ON condition and the comparatively shallow
`well is formed thereunder. On the contrary, wlgn the
`pulse d>c2 becomes high level H and the pulse d>c2 be-
`comes L level, the electronic switches 101 to 116 be-
`come ON, while the electronic switches 201 to 216
`become OFF. As a result, the horizontal buses X1 to
`X16 are disconnected from the charge sensitive amplifi-
`ers 301 to 316, and are simultaneously connected,
`in
`turn, to a low level (or ground potential) of the terminal
`75. Therefore, the detecting electrode 2X is biased to a
`low level voltage and the well just under the electrode
`2X disappears.
`On the other hand, the drive voltage having the pre-
`determined waveform is sequentially applied selectively
`to the storing electrode 2Y in the vertical picture ele-
`ment columns R1 to R15 toward the direction of the
`buses Y1 to Y115 via the vertical buses Y1 to Y15 from the
`shift register 10 in such a manner that the storing period
`and readout period, charge sustaining period and injec-
`tion period are set as hereinafter explained. In this case,
`the deep well formed previously just under the storing
`electrode 2Y disappears once in the readout period, the
`signal charges generated therein are shifted to the shal-
`low well formed just under the detecting electrode 2X
`said signal charges are read by the charge sensitive
`amplifiers 301 to 316 via the horizontal buses X1 to X16
`and the electronic switches 201 to 216 which are turned
`ON, and said signal charges are output in parallel at the
`output terminal of each charge sensitive amplifier.
`FIG. 3 shows an enlarged part of the signal charge
`readout time chart of the CID of FIG. 2. The operation
`mode of the vertical picture element columns R1, R2,
`R3, R4, .
`.
`. is the operation sequence composed of the
`charge storing period To, readout period T1, charge
`holding period T2, injection period T3. More particu-
`larly, the storing electrode 2Y of the first picture ele-
`ment column R1 is set to the high level potential H and
`operates in the storing mode during the period from too
`to toz, set to the low level potential L, and transfers -the
`signal charge to the well at the side of the detecting
`electrode 2X. Its resultant readout, during the period
`from toz to to3, is set to the high level H again and again
`fetches the signal charges and sustains them during the
`period from to3 to t13, and finally sets the gain to the low
`1evelL and injects the charges into the substrate during
`the period from t13 to t27. Both high and low levels of
`each voltage waveform shown in FIG. 3 as R1 to R4
`show the levels of voltages applied sequentially to the
`storing electrode 2Y of the corresponding vertical pic-
`ture element comumns from the shift register 10.
`The first vertical picture element column R1 is now
`considered. A high level H voltage is applied to the
`
`1034-O07
`
`1034-007
`
`

`
`4,434,441
`
`5
`detecting electrode 2X of the horizontal picture element
`columns C1 to C15 from the terminal 74, as shown in C1
`to C15 of FIG. 3 at the time to1 just before the voltage to
`the storing electrode 2Y drops to the low level L. The
`shallow well is thereby formed just under the detecting
`electrode 2X, waiting to transfer charges from the deep
`well formed just under the storing electrode 2Y. If the
`voltage of the storing electrode 2Y of the vertical pic-
`ture element column R1 drops to the low level L at the
`time tm and the well just under said storing electrode
`disappears, the shallow well just under the detecting
`electrode 2X of the pertinent picture element column
`receives the signal charges from the deep well. How-
`ever, since the image charges are generated on the de-
`tecting electrode, said image charges are detected by
`the charge sensitive amplifier, thus completing the read-
`out operation.
`Succeedingly, the voltage of the storing electrode 2Y
`of the vertical picture element column R1 comes to the
`H level again in the charge holding period from to3 to
`t13 and the charges which are once transferred to the
`detecting electrode 2X are returned again to the well
`formed under the pertinent storing electrode 2Y, setting
`the holding condition. The reason why the holding
`period is provided in this case is hereinafter explained.
`However, during this charge holding period, since the
`well is formed just under the storing electrode 2Y, the
`charges generated by the incident light are stored, but
`these stored charges are not read out as the signal
`charges and are injected into the substrate in the injec-
`tion period from t13 to t27 and disappears.
`The vertical picture element column R2 is provided
`with the same operation sequence as explained, with a
`time difference of to5——to2 and the vertical picture ele-
`ment column R3 is also provided with the same opera-
`tion sequence with a further time difference of to3—to5.
`In the same way, the succeeding vertical picture ele-
`ment columns R4, R5, R5, R7,
`. are operated with the
`same time difference and the same operation _s§quence
`as explained. In FIG. 3, the signal indicated by ¢c2 is the
`waveform for the ON and OFF control of the elec-
`tronic switches 201 to 216, corresponding to the afore-
`mentioned voltage waveforms C1 to C15 applied to the
`detecting electrode 2X of each horizontal picture ele-
`ment column. In FIG. 3, the signal indicated by 4:1; is
`the waveform of the reset pulse for the charge sensitive
`amplifiers 301 to 316. This signal is changed to high and
`low levels a little more rapidly than the aforementioned
`readout start time of the vertical picture element col-
`umns because the charge sensitive amplifiers discharge
`the charges previously stored in the feedback capaci-
`tance of said amplifiers before said amplifiers connected
`to the horizontal picture element columns detect the
`image charges. In FIG. 3, where the time t is plotted on
`the horizontal axis, voltage applied to the horizontal
`picture element columns C1 to C15 are often switched to
`high and low levels corresponding to $c2. Thus, when
`the vertical picture element column R1 enters the injec-
`tion mode, for example, at the time t=t13, the well is
`formed just under the detecting electrode 2X if the
`voltage of said detecting electrode of the horizontal
`picture element columns C1 to C15 is at the high level,
`and therefore no charge is injected into the substrate.
`This phenomenon is also true to the other vertical pic-
`ture element columns R2, R3, R4, .
`.
`.
`. Thus, as shown
`in FIG. 3, the well is not formed just under the detect-
`ing electrode 2X when the voltage of said detecting
`electrode of the horizontal picture element columns C1
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
`45
`
`50
`
`55
`
`65
`
`6
`to C15 at the time t=t12, a little before the time t=t13, is
`set at the low level and as a result the used charges can
`be injected into the substrate as required. This reason
`also causes the voltage of the detecting electrode 2X of
`the horizontal picture element columns C1 to C15 at the
`times t=t15, t=t19 and t=t22 to the change to low leve-
`l.In other words, the vertical picture element column
`R1 carries out actual charge injection in each period t12
`to H4, t15 to 1:17,
`t19 to tzo,
`t22 to t23,
`t25 to t2(, and the
`vertical picture element R2 in each period t15 to t17, t19
`to tzo, 1:22 to t23, 1:25 to t25, 1:23 to t29, the vertical picture
`element column R3 in each period t19 to t2o, t22 to t23, t25
`to t25, t23 to t29, t31 to t32 and the vertical picture element
`column R4 in each period t22 to t23, t25 to t25, t23 to t29,
`t31 to t32, t34 to t35.
`FIG. 4 shows the entire part of the time chart of the
`CID drive of the invention, and the same portions as
`shown in FIG. 3 are identified by the same symbols.
`The reason for providing the charge holding time T2
`in the CID of the invention is explained with reference
`to the time chart of FIG. 4. Here, attention is focused on
`the point Q of the 8th vertical picture element column
`R3 in FIG. 4 where it comes to the low level at the time
`t=T13 and carries out charge injection. The 8th picture
`element column R3 carries out charge injection to the
`substrate at the time T13 and injected charges enter the
`well if the well is formed in the periphery of the 8th
`picture element column.
`However, the 7th picture element column R7, which
`is nearest to the picture element column R3, the next
`nearest 6th picture element, and the 5th and 4th picture
`element columns R5 and R4, respectively at the low
`level at the charge injection time T13. More particu-
`larly, no well is formed under the storing electrodes of
`the vertical picture element columns from the 7th to the
`4th columns so that cross-talk does not occur in said
`picture element column group.
`Since the voltage of the storing electrode of the third
`picture element column R3 comes to the high level at
`the time T13 and the voltage of the storing electrodes of
`the second and first picture element columns R2 and R1,
`respectively, are switched to the high level before the
`time T13, the well is formed under the storing electrodes
`of the vertical picture element columns R1, R2 and R3
`ready for accepting the charges injected.
`However, the picture element columns from the 8th
`column to the first, second and third columns R1, R2
`and R3, respectively, are considerably isolated and the
`injected charges disappear by recoupling until
`they
`reach the wells of said first, second and third picture
`element columns and never migrate into said wells.
`When attention is focused on the 9th and 10th picture
`element columns R9 and R10, respectively, the voltages
`applied to these picture element columns are at the high
`level and, obviously, wells are formed under the storing
`electrodes. Therefore,
`the charges injected into the
`substrate from the 8th picture element at the time T13
`flow into the wells just under the storing electrodes
`formed by the 9th and 10th picture element columns R9
`and R10, respectively. Instead,
`the injected charges
`which enter the well under the detecting electrode of
`the 11th picture element column R11 at the time T13 in
`the readout mode are trapped at the wells of the 9th and
`10th picture element columns, so that only a small
`amount remain and therefore there is no deterioration of
`the picture quality.
`'
`On the other hand, the injected charge of the 8th
`picture element column flows into the wells of the 9th
`
`1034-O08
`
`1034-008
`
`

`
`4,434,441
`
`8
`will occur to those skilled in the art without departing.
`from the spirit and scope of the invention.
`We claim:
`
`7
`and 10th picture element columns, but since readout of
`the charges of these wells has been completed, if the
`injected charge flows partly flows into the walls, they
`are injected again into the substrate at the times T19 and
`T110. As a result, cross-talk does not occur in the read-
`out picture. Since the llth picture element column R11
`is spatially isolated substantially from the vertical pic-
`ture element column Rg via the double well column, the
`injected charge is considered not to flow into the well
`of the 11th picture element column R11 just formed at
`the time T13. More particularly, the influence of in-
`jected charges can be minimized by utilizing a charge
`holding period, corresponding to the two picture ele-
`ment column readout period in the FIG., which is
`longer than the time until the readout for the adjacent
`one. At least one picture element column then termi-
`nates between the readout period and the injection per-
`iod.
`
`Furthermore, in the 4th and 5th picture element col-
`umns, for example, the voltages of these picture ele-
`ments come to the high level at the times T19 and T110
`which are delayed from the time T13.
`Additionally, the picture element columns R4 and R5
`are also considerably isolated from the 8th picture ele-
`ment column. Such isolation of timing and distance
`causes the charges injected from the 8th picture element
`column to disappear due to re-coupling until they reach
`the wells generated at the times T19 and T110 of the 4th
`and 5th picture element columns R4 and R5, respec-
`tively. As a result these charges do not enter the wells,
`so that cross-talk does not result.
`The phenomenon where injected charges migrate
`into the wells formed in the other picture element col-
`umns at the periphery of the picture element column
`where charges are injected can be prevented by provid-
`ing the specified charge holding period T2 between the
`readout period T1 and the injection period T3. As a
`result, the cross-talk of signal charges can be suppressed
`and the problem of deteriorating picture quality of the
`read out image can be eliminated.
`The foregoing explanation is based on parallel read»
`out for convenience of understanding. The invention is
`not particularly limited only to parallel readout and also
`provides excellent results with a series readout system.
`As hereinbefore explained, the readout system for the
`CID of the invention assures a read out picture without
`cross-talk and provides great practical application re-
`sults, because there is no fear of permitting the frame
`time to become longer.
`While the invention has been described by means of a
`specific example and in a specific embodiment, we do
`not wish to be limited thereto, for obvious modifications
`
`1. A method for driving a charge injection device
`having picture elements, each consisting of a storing
`electrode and a detecting electrode, arranged in the
`form of a matrix on a semiconductor substrate, the stor-
`ing electrodes of the picture elements being connected
`in common for each column or row and the detecting
`electrodes being connected in common for each row or
`column, said charge injection device having a basic
`operation sequence essentially containing each opera-
`tion of charge storing,
`readout and injection, said
`method comprising the steps of sequentially
`applying the basic operation sequence for each pic-
`ture element which is considered as the readout
`unit with a constant time delay; and
`inserting the specified charge holding period between
`the readout period and the injection period.
`2. A method for driving a charge injection device as
`claimed in claim 1, wherein said charge holding period
`is longer than the period required for completing read-
`out operation of at least one adjacent readout unit pic-
`ture element.
`
`3. A method for driving a charge injection device as
`claimed in claim 1, wherein said charge holding period
`is provided at a potential well just under said storing
`electrode.
`4. A method for driving a charge injection device as
`claimed in claim 1, wherein the picture element group
`which is the unit of the readout operation includes pic-
`ture elements connected in common in a row or col-
`umn.
`~
`
`5. A method for driving a charge injection device as
`claimed in claim 4, further comprising the steps of pro-
`viding a column bus connecting the storing electrodes
`of each column in common, a row bus connecting the
`detecting electrodes of each row in common, connect-
`ing charge sensitive amplifiers corresponding to said
`row bus, sequentially applying voltage waveforms con-
`sisting of the storing mode, readout mode, charge hold-
`ing mode and charge injection mode repeatedly to said
`column bus, and reading out the signal charges in the
`unit of the column direction picture element array in
`parallel.
`6. A method for driving a charge injection device as
`claimed in claim 5, further comprising the step of peri-
`odically changing the operation voltage applied to said
`detecting electrodes through said row bus to high and
`low levels to cause the potential well just under the
`pertinent electrode to disappear prior to the start of the
`charge injection mode for each said column bus.
`93¢
`*
`fit
`Ii
`4!
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
`40
`
`45
`
`50
`
`55
`
`60
`
`65
`
`1034-009
`
`1034-009