TWI253852B - Solid photographing apparatus, digital camera and digital video camera - Google Patents

Abstract

A solid photographing apparatus is provided. After releasing signals of the photodiode in each cell, the photodiode accumulates the signals. After the accumulation, each signal is read from the photodiode. On the semiconductor substrate, each cell includes a photodiode, a reading gate, and a detection section arranged in a two-dimensional way. The reading gate reads signals in the photo-diode. The detecting section detects the read-out signals. The solid photographing apparatus has a circuit, which applies a first pulse signal used for releasing signals from the photo-diode on one horizontal line, to the corresponding cells, during a first horizontal line scanning period for several horizontal lines, and which applies a second pulse signal used for releasing signals from the photo-diode on one horizontal line, to the corresponding cells, during a second horizontal line scanning period for several horizontal lines.

Description

1253852 15763pif.doc IX. Description of the Invention: [Technical Field] The present invention relates to a solid-state imaging device, a digital camera, and a digital camera. [Prior Art] Fig. 23 shows the circuit configuration of a CMOS image sensor. Such a shirt sensor is controlled by a control unit (CPU or DSP) C. One unit (the elemental part) in the image sensing area of the image is composed of four transistors (Ta, Db Tc'Td) and a photo-polar body (pd). This unit (ceu) is configured in 2 dimensions (2-dimensional space). In the lower part of the photographing area, the source transistor (s_e) is used to store the load transistor TL in the horizontal direction. Further, in the photographing area, a column type AD conversion circuit 2 including an ig bit of a noise canceling circuit is disposed in the upper portion. • (4): The apostrophe from unit 1 is input to the AD conversion circuit 2, and the AD is sequentially output by the horizontal shift register 3 in a 10-bit manner. In the ad conversion circuit 2, a comparison number is used based on the level of the triangular wave obtained by the triangular wave controller for each unit ^H. It is converted into a 10-bit digital signal of 〇~1023 level 4, water is translated again, AD conversion circuit 2, triangular wave generation circuit, temporary storage circuit 8 and bias & 2 action, timing production circuit 8 must be configured. The stage generating circuit 9 is connected to the timing generating electric source 1253852 15763pife|j〇c ^1 in the water-light-optical body (PD) selected by the vertical shift register (ES) 5 Is discharged. The ESDATH set in the control unit c outside the unit i corresponds to the horizontal line of the vertical shift register_6, the vertical shift register (10), the water = straight; r Qing) 5 In the horizontal line selected by the vertical shift register (2:2: ί, horizontal line, vertical shift register (es) 5 selected ^ line pair by vertical shift temporary storage) The number of lines, whereby the signal amount accumulated in each photodiode (pd) is accumulated, and the light dipole in the horizontal line selected by the vertical shifting canister is read in each unit 1 (PD) signal.乂, the sequence generating circuit 8 reads the signal accumulated in the photodiode (PD) so that the gate voltage of the amplifying transistor Tb, that is, the detecting portion (DN) = falls, becomes the reference voltage (level tlevel) , (DRESET pulse wave ^ is ON (ON) 'The detection part (dn) is set to the limit by the pulse wave selector 72 (4) ^1) ° At this time, the reset level is output to the vertical signal 2 The reset level is accumulated in the capacitor for the noise canceller in the AD conversion circuit 2. For a long time, the sequence generation circuit 8 is turned on (Φ) by the pulse wave selector 72, and the transistor Td is turned on. The photodiode (PD) is known to read the charge of the ^. Go out to the detection department ((10)). Moreover, in order to select one horizontal line in the i-level effective scanning period during the vertical effective scanning period, when the φadres pulse is turned on, the corresponding row selection transistor is made via the pulse wave selector 72. ^通通, 1253852 15763pif.doc ^ In the way of the vertical signal line, the day °AD conversion circuit 2 is set in the noise canceller action L. The green mark is removed, (10) the signal enters AD, and the exchange Output 1 〇 bit. When the image sensor is operated in 30 Hz (Hz) VGA mode, the e-data of the Hz pulse and the 15.7 Hz drop are used to generate the CK pulse and the accumulation time control data. 'Man to 5 (four) sequence generation circuit 8. The timing generation circuit 8 causes the input 0VR pulse wave and φΗρ pulse wave to be reformed in the buffer circuit = 'to make (DVR! pulse wave and φΗρι pulse wave and turn out f (4) 6. Difficult to shift the register _ 6 during the low (L0) level clear (Clear) register output to make the low level. The vertical shift register (RO) 6 by φ ΗΡΙ pulse And sequentially, to select a high level) horizontal line. ^ Figure 24 is a circuit diagram of the pulse wave selector 72. Pulse wave The selector 72 is in the output & RGn level from the vertical shift |§(rq) 6 (immediately 'DREAD. '(DADRES, (10) coffee output to become 1 READn, respectively (DADRESn ' 〇) RESETn. When R〇n+i has been selected by the vertical shift register (R〇) 6 during the next horizontal active scan, the same round (!>READn+l, (|> ADRESn+l, (J) RESETn + 1. Further, the pulse wave selector 72 '(>> when the output ESn 位 south level (HI) from the vertical shift temporary storage E(ES) 5 is made. READn, (I) RESETn output The next horizontal effective % drawing period has been selected by the vertical shift register (es) 1253852 15763pif.doc 5 to output n READn+i, Cai (4). Fig. 25 shows that the above-mentioned CMOS image sensor operates the photodiode (4) to discharge the signal (10) according to the horizontal line 福~ 彻 ^ ^. Moreover, to (4) the line number 525 * read (PD read). The light diode (4) line '1 horizontal line each is changed by ad 』 electric terminal 仃 AD exchange, data data (pull out to become a 10-bit touch output. " like _ lose room between 1,2' 3 to see the storage of the diode The time of accumulation time Γ, 2, t can be understood as the same 'even if these accumulation order is offset. θ v line %·1~ each of the beginning time will be horizontally ❹Ί, pin oblique correction When the movement moves in the order, the offset occurs. For example, when shooting in the state d: an upright building, the structure will be tilted and so on, so that it does not coincide. In particular, a moving scene in motion, in Japanese Laid-Open Patent Publication No. M111900, discloses a solid-state photography device that can be used. Further, in the case of the special-type solid-state imaging device, the image-capturing device of the present invention can be satisfied. [Inventive content] The light-emitting device of the present invention is a light-emitting device of the unit 21253852 15763pif.doc polar body After the signal is discharged, the signal is accumulated by the respective photodiodes, and after the accumulation time, the signals are read by the respective photodiodes, wherein the photodiode on the semiconductor substrate and the read gate for reading the signal from the photodiode are The gate and the detection unit for detecting the read signal are arranged in a two-dimensional manner for each unit. The solid-state imaging device includes a circuit that applies a first pulse wave signal for discharging a signal of each photodiode of the horizontal line to a corresponding unit in a plurality of horizontal lines in the first horizontal scanning period. In the case of a plurality of horizontal lines in the second horizontal scanning period, the operation of applying the second pulse wave signal for reading the signals of the respective photodiodes of one horizontal line to the corresponding respective units is performed. In another embodiment of the solid-state imaging device of the present invention, after the light of each unit is dielectrically charged, a signal is accumulated by each photodiode, and after the accumulation = the signal is read from the photodiode of the upper electrode to the detecting portion, The detection is: an output signal, wherein the photodiode on the semiconductor substrate is used in the second-order square by the detecting portion for the thin light. The device has a circuit, which is performed under the condition of the solid-state photography. In the case of a plurality of horizontal lines, the signal discharge of each of the light-emitting diodes in the period of the i-field is performed under the condition of the two-thirds of the light-emitting diodes of the flat line, and the operation of each unit corresponding to the one-water is used. The second pulse wave signal is applied to the third pulse signal 10 I2538523p_ for the operation of each of the corresponding cells in the return line period of the detection unit of the one horizontal line. A form of digital camera uses the above-described solid-state imaging device. Another form of digital camera of the present invention uses the above-described solid-state photography to make the above and other objects, features and advantages of the present invention more apparent and easy to understand. Good implementation , And with the accompanying drawings, as it is detailed below. [Embodiment] Hereinafter, description will be made with reference to embodiments and drawings. Fig. 1 is a block diagram showing the configuration of a digital camera to which the solid-state imaging device according to the first embodiment is applied. The digital camera (10), the microcomputer 101 is connected to the timing generation image signal processing = the unit 103 and the DV tape portion. The Dv conversion unit 1〇3 is connected to the video tape unit 104 and the liquid crystal monitor 1〇5. Further, the optical axis of the lens 1〇6 is placed, and the solid-state imaging device (unit) s ' timing generation video signal processing unit is connected to the solid-state imaging device s. Further, the microphone is connected to the DV conversion unit 103 via the sound processing unit 108. Further, the digital photography eclipse 1 100 includes a power supply unit 109. The microcomputer 101 controls the timing generation video signal processing unit 1〇2, dv to j 103, and the DV video tape unit 1〇4. The sensor 8 picks up the image that is input through the lens 1〇4 according to the signal generated by the timing processing unit K)2, and outputs the image signal to the time-series image signal processing unit, and the microphone 1〇7 The sound is input, and the sound is output to the sound signal processing unit 108. The sound signal processed in the 11th light processing unit 108 processed by the time-series generation video signal processing unit 1〇2 is output to the DV conversion unit 103 to perform DV conversion. The image signal that has been converted = converted in the DV conversion unit 1〇3 is sent to the DV tape unit 1〇4 for recording, and the same day is exchanged for an analog signal for output to the liquid crystal monitor. In the reproduction mode, the recorded signal of the DV tape portion 104 is converted into an analog number in the - conversion portion 1A3 to be output to the liquid crystal monitor. In the second embodiment, the circuit configuration of the CM〇S image sensor (solid-state imaging device) S which has been formed on the semiconductor substrate in the second embodiment. The same part of Fig. 23 is provided with the symbol of phase (5). The image sensor, or ah to control. The image sensor in the photographic field is composed of four transistors (Ta, Tb, Tc, Td) and a photodiode. The unit is arranged in a 2-dimensional manner. The load transistor TL for the lower-middle-source circuit of the photographing area is arranged in the horizontal direction Γ =, and a 10-bit cylindrical AD conversion circuit 2 including a noise canceling circuit is present in the upper portion of the photographing area. The signal from each unit 1 is input to the AD transformation =:::=r: r° bit woven in two directions. In the AD conversion circuit 2, based on the level of the triangle s (RAMP) from i to _S 4 4 %, the money of the digit ^1 is converted into the G of the 1G bit to the 3-bit path 4, For each single-chip conversion circuit 2, triangle wave generation potential register = bit 6 register i' vertical fine register (ES) 5, vertical shift, pulse wave selector 7 action, then configuration timing generation 12 1253852 15763pif.||(, the circuit 8 and the bias generating circuit 9. The control unit c is connected to the timing generating circuit 8.

The horizontal line selected by the vertical shift register (ES) 5 in each cell 1 = the signal accumulated in each photodiode (PD) is discharged. Each unit! The accumulation is set by the ESDATA input by the control unit C, and the day-to-day correspondence corresponds to the vertical shift register (11〇) 6 selected horizontal line and vertical shift register (ES) 5 The number of horizontal lines between the selected horizontal lines. The vertical f-bit register (ES) 5 selects the horizontal line before being selected by the vertical shift register (R〇) 6. The vertical shift register (ES) 5 selects a horizontal line that is separated from a certain line by the horizontal line selected by the vertical shift register_)6, thereby making each photodiode (PD) The amount of accumulated signal is controlled. After the accumulation time, the signal of the photodiode (pD) in the horizontal line selected by the vertical shift register 6 is read out in each cell.

The timing generating circuit 8 reads the signal accumulated in the photodiode (4), and in order to make the gate voltage of the amplifying transistor hole, that is, the voltage 'in the detecting portion' becomes a reference voltage (resetlevd), The Dingmai ^ is turned ON (ON), and the detection unit is set to the reset level (4) t level via the pulse wave selector 72. At this time, the electric squid for the noise canceller in the level output conversion circuit 2 is reset to reset the level. Next, the timing generating circuit 8 reads the 1st 7 in the 1st milk pulse wave, and reads the chemical charge accumulated in the transistor body (PD) to the detecting portion ((10)). Moreover, the timing generation circuit 8 is effectively returned by the __丨 level 13 1253852 15763plf.doc

ΓΓΓ, (10) towel for the lion's horizontal line, the financial situation like 臓S ϊΐ m through the miscellaneous surface 117 to make the corresponding row chat transistor η amplification transistor Tb and load transistor TL: n: circuit operation. In the above manner, 'detection part_read=direct field line (VLIN), signal level + reset level". When the AD conversion buffer is activated, the signal level + reset level In addition, HL only transforms the signal to output 10 bits. In the shirt image sensor, WVR pulse wave with 30 Hz (Hz) VGa = 2, 7 s = 2 pulse 'The Doc pulse wave and the accumulation time are in the timing generation circuit 8. The time series production and production circuit 8 i吏 has been input (DVR pulse wave and φΗρ pulse wave memory) 6. Lai Zhi Shi (4) register In addition to (α (four) register output, the sequence is operated by _ pulse wave to select (4) and in the first implementation form, in order to improve the simultaneity of the dynamic 和 and to change the mussel 'Z image DU i and The mode 2 is ordered so that the triangle waveform is switched to positive (four) and Μ shape. Fig. 3 is the circuit diagram of the pulse wave material selection 7. The ΐ Λ Λ 成为 成为 成为 0 0 0 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 Bit register _6 to 14 1253852 15763pif.doc When ROn+1 is selected, the output is similar to that of Dingzhou.

Further, when the pulse wave selector 7 is input by the timing generation circuit δ (when the D VREAD pulse wave is in the south state, all the horizontal lines are selected during the vertical effective scanning period, and the full-scale output φ RESETn and φ READn··· Therefore, the full pixel can simultaneously perform the discharge of pD and the readout of PD. Fig. 4 is a driving timing diagram of the pixel unit according to the mode i (M〇de 1) operation. At this time, the image sensor It is operated in 1 field (3 Hz), and the number of horizontal lines juxtaposed in the vertical direction becomes 480 lines. First, the initial selection of the pulse wave selector 7 during the vertical return line is to make the photodiode of the whole unit 1 (PD) When the signal is discharged, all of φμαι) ι ~ 〇 READ480 must be in a high state at the same time. At this time, in order to discard the signal read by the detecting unit (DN) to the power supply (VDD)*, all of the 1 RESET1 to 〇 RESET 480 must be in the high state. Then, photoelectric conversion is performed with a photodiode (PD) to accumulate signal charges. Then, after the vertical return, the signal charge of the photodiode (PD) accumulated in all the cells 1 is read out at the same time, in order to remove the invalid signal such as the leakage current of the detecting portion (DN), all of them are Φ RESET1~φ RESET48〇 must be in the high state. Then, all of Φ ΙΙΕΑ 〇 1 φ φ 348 () are in a state of 5, and the signal charge of the photodiode (PD) accumulated in the entire cell 1 is simultaneously read out to the detecting portion (DN). Then, during each horizontal return line period in the vertical effective scanning period, a signal for signal output of one horizontal line is outputted from the detecting portion (DN). For example, when the signal of the horizontal line No. 1 (No. 1) is detected, the pulse wave selector 7 makes 1 15 125385 2 15763 pif.doc ADRES1 & , 贝d言|虎>[Standing + reset words sold to the vertical signal line (VLIN). Next, the pulse wave selector 7 resets the detection portion (DN) by resetting the φ RJESET1 to the south state, and resets the level output to the vertical signal line (VLIN), and the AD conversion circuit 2 performs the noise cancellation circuit. In the differential operation, the reset level is removed from the ''signal level + vertical level' and only the signal is output. Fig. 5 is a driving timing diagram of the pixel unit operating in mode 2 (Mode 2). The sensor also operates in one field (fieldp 〇 Hz. First, the pulse selector 7 makes OREADl during the horizontal return line during the vertical active scanning period, (DRESET1 becomes a high state, so that the horizontal line 1 light one body The signal discharge of (PD) is repeated in the above order of the horizontal line No. 1 to No. 48 (NO. 480). Then, photoelectric conversion is performed by the photodiode (ρ〇) to accumulate signal charges. The vertical charge is valid for the signal charge of the photodiode (PD). For example, when the signal of the horizontal line No. 1 is read, the pulse selector 7 is reset by MDRES1 (DRESET1 becomes high) The level is output to the vertical = line (VLIN). Secondly, the pulse selector 7 is made by ΕΑΕ)1 When the state is high, the signal level + reset level, ·, is read to (V_. Moreover, the AD conversion circuit 2 removes the weight by the noise cancellation circuit, == vertical level ' The position is set, and only the signal is AD-converted and output by the image sensor. 2 In the first embodiment, the operation of the CM0S image sensor is displayed. In Fig. 6, the accumulation time 1, 2, 3 is used to display the light two. In the present embodiment, the number of scanning lines ^ 1253852 15763pif.doc is twice the number of deleted lines. That is, the vertical return line (57〇H), the vertical effective scanning period In the case of the mode 1 (m〇m丨), the accumulation start time of all the numbers 1 to 1 is the same, and all the levels = the PD discharge and the PD read in the number are The output of the signal that has been read out in the motion detection section _) is repeated in the vertical direction. The tiger line (VUN) is not in the heart of the time = 1 time (1〇5〇) H) (or the first phase of the first field of two fields: two light-polar body (PD) to accumulate 'all photodiodes (PD) already (four) ΐ the next accumulation time 2, about 2 times the accumulation time 1 The wheel input indicates the optical input signal. In the accumulation time 2, the vertical return: AD 2 times, and ipD read = line photoelectric switch, for * to make the straight light to the light pole _) The signal charge of the known animal is read out to the detection 郜 (垂:; 丨, _ pulse wave and the _ out action is performed; when the P is terminated, the 3 READ pulse wave which has been accumulated again for reading the photodiode (4) and 2PD is performed Read operation. ^ is input i. The optical input signal twice as large as 1% can be converted into signal charge, and when the accumulation time is 3, the amount of light input signal is large. 17 1253852 15763pif.doc In this accumulation time 3, the φΚΕΑΕ) pulse wave is generated twice during the 1/2 vertical return line period, and the ipD read operation and the 2pD read operation are performed. As in the accumulation time 2, about twice the signal charge of the photodiode (pD) can be obtained. In the next frame (or field), the signal charges of the horizontal lines No. 1 to No. 480 are read out to the vertical signal line (VLIN) during the vertical effective scanning period of about 1/2 vertical period.

According to the above operation, the accumulation time can be continuously controlled from one line (1H) to one frame period (or one field, between). Further, when the OREAD is sold by u to generate a pulse wave twice or more during the frame period (or the field period), a signal amount of twice or more the photodiode is obtained. Further, when the vertical effective scanning period is 1/2 frame (or field) (1/2 V), the leakage current of the detecting portion can be made as low as 1/2. Therefore, the signal noise can be improved by more than 2 times than the S/N. In the present embodiment, although OREAD is read to generate a pulse wave twice or more, it may be generated three times or more when the saturation of the output portion is excessive. When compared with the accumulation _ from the start of the PD discharge to the 1 PD read operation, the two-times start from the start of the 1PD|buy operation to the read operation: the inter-discipline is long. In this operation, the 2PD readout signal can be added to the portion where the lpD is read without the fuller day unevenness f. When the pD is saturated at 2pD reading, there is no problem caused by the saturation of the pD portion when the 1JPD is mastered. The accumulation time can be set to the full _ of the frame or any period during the return line period. ΓΡΤΤι I 7A is a 矽1 sectional view and a potential operation diagram of the photodiode for explaining the operation of the accumulation time 2 in Fig. 6 . The low-leakage type 18 1253852 15763pif.doc photodiode (PD), the detection part (〇1^) and the surface of the well formed by the well (wdI) or the ruthenium substrate are formed by a p-type seal.汲极〇/〇〇). The signal read gate (Φ READ) and the reset gate (φ RESET) are formed by polycrystalline spine. The detection unit (DN) is connected to the amp transistor (Tb) for rotation. Fig. 7B is a potential diagram before 1 pD reading =. In the photodiode (PD), the optical signal is converted into a wide charge to accumulate approximately the approximately saturated signal charge Q1 of the photodiode (PD). Fig. 7C is a potential map after 1 PD readout. The signal charge Q1 is read out from the photodiode (fD) into the detecting portion (DN). The applied pulse of the read pulse 〇 READ is read at Vh which is deeper than the potential of the photodiode (PD). Since the surface of the photodiode (PD) portion is sealed by p-type, the electric potential of the photodiode (pD) is not deeper than the potential formed by a certain read voltage. Figure ^2 shows the potential map before PD reading. The optical signal is then charged in the photodiode (pD) to accumulate approximately the saturated signal turtle Q2 of the photodiode (PD). Fig. 7E疋2PD shows the potential map after it is taken out. The jg electric charge Q1+Q2 which is added by the photodiode (pD) = the read signal charge Q2 and the signal charge Q1 accumulated by the detecting portion (DN) towel is accumulated in the detecting portion (DN). , 丄 can get about 2 times the signal charge of the photodiode (PD). Similarly, in the electromagnetization time 3, a signal of about twice the (four) diode (PD) can be obtained. In the first embodiment, since the vertical signal line readout period is first, 1/2' The output of the image sensing gain becomes twice the previous chirp = 4 MHz signal. Further, the output period of the signal is compressed to 1/2 of the period of the vertical effective T field. Therefore, in order to output to a normal monitor, it is necessary to perform speed conversion using an external frame memory. In the first embodiment, "mode" and mode 2 are switched. At 19 1253852 15763pif.c|〇c

= f f, the fineness of the KTC enamel shown in Fig. 2G (described later) is deteriorated. - In terms of the picture, Bai Congcheng's: The temple did not improve, but the KTC noise could be smashed by the noise of ^. The native drawing diagram 8 shows mode 1 (M〇de ^ and mode Na can be set up by people (4). 1 = The action is performed by the control of the control unit C. ^, ^ In step S1, if the time is used by In the image sense (4), the j machine (for example, 'digital camera') is manually set. In step S2, C sets the mode i for improving the simultaneity to the sport mode, and the mode 2 for the enamel mode is set to the standard mode. In step S1, if the user uses the camera to determine whether the color signal processing D:p 2 is used in a 1 U gain mode or in a gain boosting manner, the gain of the shirt image sensor is used. When the gain is increased, the standard mode obtained by the high image quality in step S4 is dark. The gain is not the gain ς ^日. The semaphore is judged in step S5, in step time (TS). The accumulation time (TS) is deleted in the S8 center; in the step S7, the accumulation time (TS) is (10) (10), and the mode 2 is set in S9. The mode 1 and the mode 2 are set. Then, in step S1G, the day-to-day generation circuit 8 ^ READ, mread and ORESET pulse waves are switched and output. Step % gain The determination or the reference value of the accumulation time determination in step S7 can be arbitrarily changed. 20 125385i^63pif.doc However, in the first embodiment, the read frequency becomes faster. The number of horizontal lines of the video film is twice as high.

The signal is twice as fast as 74 MHz. V, Z Fig. 9 shows the countermeasure of the i-th embodiment. The luminous cycle of Rong New is matched. For example, the vertical return line period is set to 8 3 milliseconds (ms) in the frequency range of 60 Hz. The frequency of the system is set to 1 G sec (4) in the area of 5 GHz. Figure 9 shows the action of 8 3 sec (4) during the vertical return line of the 60 Hz region.枳 , the 30 Hz VGA action is set to 45 h of 8 8 ms during the vertical return line. In the example of Fig. 9, the vertical return line period is operated by the number of horizontal lines of 16 640h milliseconds '1 frame 640H. The action solution became the previous 640/525 spit 22 times faster. The 2 million-pixel high-definition movie adds only a little bit of frequency at 45 MHz. Mode 2 is performed at an accumulation time longer than 83 milliseconds of the minimum illumination period of the fluorescent lamp. Set mode 1 when the accumulation time is below 8.3 milliohms. By such an operation, the simultaneity of the accumulation time of 8.3 milliseconds or less can be improved when the operating frequency is increased a little. When the camera of the lens aperture F2.8 of the 330,000 element CMOS image sensor of the first embodiment is used, an accumulation time of 10 milliseconds or more is obtained when shooting is performed in a normal office. Since the motion or the like is performed at a brighter %, the accumulation time of 8.3 milliseconds (10 milliseconds or less) is an effective method for improving the simultaneous generation. When the accumulation time is 8.3 milliseconds (1 milliseconds) or less, if the image is captured in mode 2, the amount of light emitted from the fluorescent lamp will vary depending on the accumulation start time of each horizontal line, and the monitor image will be reproduced. A horizontal stripe corresponding to the amount of luminescence of the fluorescent lamp occurs. Therefore, when 21 1253852 15763plf.doc becomes 1, the horizontal stripes in which the luminescence amount of the luminescence does not occur in the first frame due to the same accumulation start time. However, although the difference between the frames will be ίί::, but with the CCD camera component - each frame is below seconds), switching to mode i becomes an effective means. (The basis of the switch between the hair = die ^1 and the mode 2 is 1/2 of the time even if it is 2 times the milliseconds or 16 6 milliseconds of the fluorescent light _ 2 times. In the 'i-th embodiment, the number of vertical effective scanning lines is not changed, and the stitch return line _ is increased. Fine, the number of vertical effective scanning lines can also be reduced, but the vertical return line _ must be added at this time. VGA = valid 昼The level is _ 昼 ,, 教 昼 且 且 且 且 且 且 且 且 且 且 A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A A During the effective drawing period 360H, the vertical return line period is 165H for 1〇5 milliseconds. In addition, the VGA's 4 times the prime number of 丨230,000 昼 ,, the level 128 〇昼 ,, vertical 960 昼 中 成为 成为 成为 HD HD HD HD HD HD HD · At 9 o'clock, it becomes vertical wo = prime, so it can be combined with HDTV's level of 1280 昼 ,, vertical 72 〇昼 = = 柁 。 。 HD HD HD HD HD HD HD HD HD HD HD HD HD HD HD HD HD HD HD HD HD HD HD HD HD HD HD HD HD HD HD HD HD · When it is 3 milliseconds (1 〇 milliseconds), it is implemented in Equation 2. When it is 1/2 of 4.15 milliseconds or less, it switches to mode 丨. In the method of effectively scanning the line, the frame memory that can be realized at the same frequency as before is also unnecessary. Moreover, in the case of the model, the PD reading can be performed twice or more as shown in FIG. /N is improved to obtain a saturation signal twice as large as that of the photodiode 22 1253852 15763 pif. doc. Fig. 10 shows a configuration of a path for the countermeasure relating to the first embodiment. By adding a photodiode (PD) to each unit 1 Signal discharge band = transistor Te and wiring ΦESREAD (for electronic shutter readout pulse "%. In the vertical signal line (milk IN), perform a readout action two >: action., 匕' can As in the previous case, the accumulation time of the 1 frame coffee is used: the result is that the output of the image sensor can also be like the signal of the first j. ^ t becomes the above-mentioned digital camera for moving photography. The solid-state imaging device to which the first embodiment is applied is described as an example, and the solid-state imaging device can also be applied to a digital camera for stationary imaging, and the solid state of the first embodiment of the present invention is shown. A block diagram of the configuration of a camera suitable for a photographing device. The digital camera is provided with a micro-computer 01, and the microcomputer 201 is connected to the time-series video signal processing unit, such as the bevel compression unit 203 and the video memory 204. The video data is transmitted through the image memory 204 and the liquid crystal monitor 205. The solid-state imaging device (sensor) S is disposed on the second axis 21, and the time-series generation image signal processing unit 2〇2 is connected in a solid state. Further, the digital unit 〇〇 has a power supply unit 207. The data 2Q1 butterfly timing is generated. The video signal processing unit 2Q2 generates a scene in the video sequence. ^ and the shirt image memory 204. The sensor s performs the imaging via the lens 206 based on the signal incident from the second and the processing unit 2〇2, and outputs the image signal to the timing generation 23 1253852 15763pif.doc image ## processing Part 202. The video signal processed by the time-series generated video signal processing unit 2〇2 is output to the video data compressing unit 1〇3, converted into an analogy number, and displayed on the liquid crystal monitor 205. If the photographing is reversed, 'then=image k is compressed in the image data compressing unit 203 and recorded in the image memory 204. In the reproduction mode, the lean material from the image memory 2〇4 is read into the image data compressing unit 103 and converted into an analog signal to be displayed on the liquid crystal monitor. And, even if the graffiti imaging device s is applied to the digital camera 2 (8), the operation of the solid-state imaging device S is still _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ The circuit configuration of the large cm〇s image sensor S is placed. In Figure 12, the lang symbol is used to indicate. In this configuration, three verticals are provided: although the full-quality simultaneous reading (4) is not provided (DREAD driving state, (DREAD pulse wave is input to the pulse wave selector 71. Further, Sa is made by the bitter circuit 81) Control the vertical shift register (4) 5" with: i =) p6 2 3 and pulse wave generating circuit 82 of mode 2. The mode 1 of the wave cutting circuit 81 and the pulse of the mode 2 are changed X四 (4) The circuit diagram of the output signal buffer 71 of the slit generation circuit 82. The rotation of the three vertical shifts of the wise son Yi X <, , (ADR) 1 〇 R 〇 n, ESn ( Electric (AND^VI) (2) is supplied to the pulse selector 7 and 1READn, 1RESETn, 1 24 Ι 2538ι· ADRESn is rotated to the unit i via the 〆() loop. The timing generation circuit 8 is used. φ READ '(J)ADRES, φΚΕδΕΤ is supplied to each AND circuit. Figure 14 is the driving timing diagram of the pixel unit operating in mode 1 (Mode 1). (1) eld) operates in the manner of Hz. In Fig. 14, in the horizontal effective scanning period corresponding to the accumulation time, the signals of the photodiode (pD) where the horizontal line is located, for example, are sequentially placed. On this day, the pulse wave selector 71 sequentially applies 1 RESET1 to 〇reSET4 to each detection unit (DN) during the horizontal effective scanning to reset each detection unit (DN). Moreover, the pulse wave selection The device force sequentially applies φκΕΑΕ)1 to ΦΙΙΕΑΓ)4 to each photodiode (pD) during the next horizontal effective scanning period to perform a discharge operation of the signal caused by the t-electron shutter. Similarly, the pulse wave selector 71 is under During each horizontal effective scanning period, φ RESET5 〇 RESET8 is sequentially applied to each detection portion (DN) where the next four lines are located, so that each detection portion (dn) is reset. Moreover, the pulse wave selector π is effective at the next level. During the scanning period, φ READ5~1read8 is sequentially applied to each photo-polar body (PD) to perform a discharge operation of the signal caused by the electron fast (10). The above operations are sequentially repeated to make all horizontal lines of photodiodes (PD) The signal is discharged. Further, after the accumulation time set by ESDATA, the signal reading operation by the photodiode (PD) is performed. At this time, the pulse wave selector 71 sequentially applies φ RESET during the horizontal effective scanning.丨~φ reset: to each check Part (DN), resetting each detecting unit (DN). Moreover, the pulse wave selector 71 sequentially applies the brush ~d> READ4 to each photodiode (4) during the next horizontal effective scanning. The action of each photodiode (pD) 25 1253852 15763pif.doc eucalyptus is performed. Similarly, the pulse wave selector 71 applies φ RESET5~φ to the next four horizontal effective scanning periods. Each detection unit (DN) resets each detection unit_. Further, the pulse wave k is 71, and the read operation by the respective photodiodes (4) = is performed by sequentially applying 1 ft. to the respective photodiodes (pD) ′ during the next horizontal effective scanning. The above operations are repeated in order to read out the signals of the photodiodes (PD) of all the water lines. After crying, the vertical signal and line VUN are read. First, the pulse wave selection 3 ADRES1 becomes the ❺ state ‘the detection unit of the horizontal line No. 1 搵1 level: reset level, and read out to the vertical signal line. Secondly, the pulse selection method is such that ΦΙΙ·Τ1 becomes high “so that the detection unit resets two sets” 'to output the reset level to the vertical signal line. The ad conversion circuit 2 ^ is composed of the noise circuit The differential action causes the reset level to be removed by the "money level + weight" and only the money is output. This crane is carried out in sequence on the parental level 1 horizontal line during the horizontal return line. Figure 15 is a vertical shift register (Es) 5, _) 6, (ADR) 1Q (10). In Fig. 15, the output waveform of the object 14 can be compared with the output of the electronic shutter during the horizontal effective scanning period by the vertical shift register (10) 5: tES1 sequentially becomes the high state. In the '1 horizontal effective scanning period, s# "buy the day" from the vertical shift register (4)) 6 * (10) ~ the sequence becomes the high state. When the vertical signal line is output, it is shifted vertically. The ADR of the ADR is a high state in the same period of time during the blood level effective scanning. The above actions are performed sequentially for the horizontal line No. 1 to No. 480. 26 1253852 15763pif· 2 In the implementation form, although the electronic shutter of the photodiode (pD), ^ signal readout is implemented in a horizontal line during the horizontal effective scanning, but only Md) READ pulse or 〇 RESET pulse The number of lines in the vicinity of the horizontal effective scanning period can also be increased in the range of the responsiveness. Further, the generation of the pulse wave is not limited to the one-level effective scanning period, and the pulse wave can be generated in one horizontal scanning period including the horizontal return line period. The 16 series is based on the operation timing chart of the element unit of the M〇de u operation, and its simultaneity is improved as compared with the case of Fig. 14. "In Fig. 16, the electronic shutter is discharged or the signal is read. · X flat has a period of time during the sweeping of the heart The flat line is one set and these ^ flat lines are selected at the same time, and each of the selected horizontal lines is sequentially discharged or read by the photodiode (PD). For example, the pulse wave selector 71 applies the same phase 峨_~ Secondly, the pulse of the same phase is added (DREAD5~(DREAD8.) The pulse of the same phase is applied to the i-group with 4 horizontal lines, and then the pulse wave of her offset is applied to the 4 horizontal lines in turn. In comparison, four times of simultaneity can be obtained. Although the * horizontal line is made into the same phase, it can be increased from two horizontal lines until the number of available lines. Figure π is the vertical shift register in the second embodiment ( ES), (R〇) the output waveform. In Figure 17, the output waveform of Figure 16 can be compared with the output of the vertical shift register (ES) 5 and the output of the vertical shift register (R〇) 6. The two horizontal lines are respectively one set and become high. At the same time, the vertical shift register (ADR) is like the one in Fig. 5, and the horizontal line becomes high. Figure 18 shows the action of the CMOS image sensor. Timing diagram. Figure 27 1253852 lsveapif.ijoc ρ^^ change in accumulation time to accumulate Π1 Ί and water Each line of the line number; two ^:, horizontal line! No. ^ϊί^Γρ?^, 2 ρ ^ 1 fφ water thousand during the effective scanning period in the horizontal line of the horizontal release of the horizontal ^ PD discharge and PD reading, The vertical detection part of the signal accumulated in the 丨 2 water + line on the line _ , the shutter mechanism of the monocular reflected light camera of the money, the shutter speed of 1 / 125 seconds ~ 1 / 250 seconds The focus plane operation is performed. In the case of performing a 30 Hz (1/30 sec.) motion photography operation, the operation of reading a 4-horizontal line or an 8-horizon line by a horizontally-perceived period towel can be 4 The shutter speed of 1/120 second or 8x speed of 1/240 second is the same as that of the single-eye, light-emitting camera. Moreover, by increasing the number of horizontal lines, the simultaneity can be improved when compared with a single-lens reflex camera. As described above, in the present embodiment, the mode 1 and the mode 2 can be performed. When the mode 1 is 1,1253852, the 15763 pif.doc mode 1 can be (4) the time is changed, and the L is caused by the granules of the granules. The offset produces a pull-like Wei, and the enamel deteriorates. On the one hand, the mold, but because it can eliminate the KTC noise, the granular ^ can be the same to the southern tannin. The timing chart of mode 2 is shown in Fig. 5 and Fig. 19 as mode 1 (M〇de丨) and =, and is manually set or controlled by the control unit C. The mode 2 of the camera # is in pure mode. The timing is as follows: === Sensing 2 = Lifting mode is used to set the subject in the configuration mode ^=14 obtained by using the dark enamel of the image. In step S15, it is judged that the formula is 2. Not gain boost time CTS). In step S17, the accumulation is changed in η%,: and mode 1 is set in (10). ;; = = (IS) is shorter than (4), and mode 2 is set in (10). Dream 7 is set by the temple _), then the pulse value of the casting method is changed and the wheel is switched in the step (4) 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日 日. In the READ pulse wave generating circuit in the timing generating circuit 8, 1H(10) (DREAD pulse wave generation number and (10)^ generation line number (horizontal horizontal line) are changed in accordance with the accumulation time. 〇rEad pulse wave generation in 1H The number can be: CDREAD pulse wave generation (four) + (525_ accumulation line number) / accumulation line number) OREAD pulse wave simultaneous generation line number can be: OREAD generation line number = 〇> READ pulse wave generation number / maximum 1 ^ The number of calculations is slightly omitted. The maximum number of generations is lower than that of Kazaki. (The DREAD pulse generation number is expressed by how many skins can be generated during the m period. Currently in the L98 microsecond period during the period of 1/3 〇 (sec) / 525 (line) = 63.5 microseconds (us) The wave produces a phantom. The simultaneity (twist line) is the number of squall lines recorded in the table, so that only a small degree of human correction value == gain determination or the accumulation time of step S17 is determined. Most of the input is in the face-to-face shutter action. The focus plane shutter makes the gap in front of the film in the upper: p action... The current monocular reflected light is usually made into a full opening and stored in 1/125~ seconds. In the actual camera, the second time is two: when S 6H '1/250 seconds, the accumulation time is 63H, and the same simultaneity of a single-eye reflected light camera using the upper and lower slits can be obtained. The short accumulation time of the sea can also improve the simultaneity compared to the single-lens reflex camera. Gas 1253852 15763pif.doc M jf is the second embodiment of the KTC noise when the low illumination is displayed. The subject is very noticeable and the electronic shutter is actuated. The influence of C noise is small. Compared with the operation of accumulating time 1 at low illumination, it is not possible to obtain the signal in the mode 2 'detection unit (DN), and the signal is in the photodiode (4). Medium. i horizontal return line is output by the photodiode item and output to the vertical signal line ^^ very). In the f mode 1, first, in order to reset the black signal of the leakage current of the detecting portion (DN), the 〇RESE pulse wave must be applied. Next, in order to read the signal from the photodiode (?0), it is necessary to apply (DREAD pulse wave to keep the read signal in the detection portion (DN). Moreover, if 〇ADRES is applied during the horizontal return line The pulse wave is held in the detection unit (DN), and the signal is level + reset level, and is output to the vertical signal line (VLIN). Moreover, the oreset pulse wave is applied to output the reset level to Vertical signal line (VLIN). Then 'reset the level sum in the AD conversion circuit 2, the signal level + reset level differential action, by the "signal level + reset level," remove the weight The noise component is left and only the signal is left. However, there is KTC noise mixed in the signal due to the difference between the reset level of the initial detection unit (DN) and the reset level during the horizontal return line. In mode 2, first, in order to reset the invalid signal of the leaking flow of the detecting portion (DN), the ORESET pulse must be applied. At this time, the reset level is output to the vertical signal line ( VLIN). Secondly, in order to read the signal from the photodiode (PD), an OREAD pulse must be applied. 31 I25385263p_ ΐίΓLt out In the detection part (DN), the signal level + reset bit Ϊ ^ 重 Hf# (VUN) °, in the AD conversion circuit 2 signal level material weight + money, (4) minute action, by " In this kind of moving h 4, the resetting of the noise component is removed and only the signal is left. In the case of 槿 Γ Γ 重置 重置 重置 重置 重置 重置 kt kt kt kt kt kt kt kt kt kt kt kt kt kt kt kt kt kt kt kt kt kt kt kt kt Output. On the other hand, in the mode I, the output is reversed. In this case, the signal of the mode 2 is reversed and the signal becomes positive. In the first formula, first, the detection unit ( DN) is reset, then the check after reset == by, the KTC noise of different levels caused by the reset action. The other side resets the position ' first, after the checkout _ reset, due to Output the input surface, and the KTC noise mixed in the reset level. Then, the signal is read out from the completed photodiode (PD). The detection unit _ becomes, ^. KTC noise reset level, and. Moreover, AD conversion power inherently contains signal level + ΚΤ (: the weight of noise

KTC = quasi-removed, only signals that do not contain KTC noise. When the mode 丨 and mode 2 described above are switched in correspondence with the storage of two, the 照S/N of low illuminance can be realized. Further, in the above-described form of application, although the configuration of the unit of the pixel unit composed of one photodiode (PD) and four transistors is described, one photodiode shown in FIG. The configuration of the i-cell 1 unit composed of (PD) and five transistors can also be reduced by switching between mode 1 and mode 2 to reduce the KTC noise at a low illumination of 32 1253852 15763 pif.doc degrees. Also, the polarity of the signal after AD conversion in mode 1 and mode 2 is reversed. In Fig. 12, the output after AD conversion is inverted in mode 1 and mode 2. Further, in FIG. 2, the polarity of the triangular waveform is reversed in mode 1 and mode 2. In other cases, the input polarity of the operational amplifier type comparator of the AD conversion circuit has a switching method in mode 1 and mode 2. At this time, the signal is input to the positive side of the comparator in mode 1, and the triangular waveform is input to the negative side. On the one hand, in mode 2, the signal is reversely input to the negative side of the comparator, and the waveform is input to the positive side. Also, an input of the comparator is provided with an inverting amplifier. The mode 1 and mode 2 can also be used with or without the inverting amplifier. Moreover, in the above-described embodiment, although the image sensor of the 330,000-pixel VGA method is taken as an example, there are many more pixels, and the number of CMOs image sensors that are also available in the first place are usually The accumulative start time of the action frequency dipole with the frame is accumulating when it is offset at every horizontal line: the first horizontal line of the door t starts to the last horizontal line, and the different 'moving' When the subject is photographed, there is an unsuitability for the 1st, and there is no simultaneity. Shutter movement:: In the form of 2, the electrons in the CM〇S image sensor are all carried out on the Μ/inside 2 horizontal line at the time of splitting at each level. This electronic shutter action has a short accumulation time and is added. The number of lines read by the photodiode during such a t-throw period is 増 /, which can correspond to the unchanging electric power during the vertical return line. 33 1253852 15763p|f· The sub-shutter accumulation time improves the simultaneity. Moreover, in another embodiment, when the electronic gate of the CMOS image sensor is operated, the vertical return line period of one frame (or one field) becomes 4, 15 milliseconds of 1/2 of the minimum period of the fluorescent light. (mS) or more than 5 sec., I, and rabbit are about 1/2 or less during the frame period (or during the field). When the signals arranged in the second-order photodiode are simultaneously discharged, the signals are accumulated in the respective photodiodes, and the signals of all the photo-integrators are simultaneously read after the accumulation time. _ If the vertical return line period is set to 1/1 2. The amount of light can be controlled in the 1/2 step during the human time during the frame. That is, the accumulation time can be made by ^1 frame' 1/2 frame, 1M frame.... Further, when the accumulation time is 1/2 or less, the accumulation time can be controlled in the following time in the horizontal scanning period. In one aspect, by making the vertical return line period 4.15 milliseconds (mS) or 5 milliseconds of 1/2 of the minimum period of illumination of the fluorescent lamp, or setting it to 8. 3 milliseconds or 10 milliseconds of the minimum period of illumination of the fluorescent lamp. , continuous photographic shooting such as MpE^ mode or Motion JPEG mode can be realized. In particular, the aspect ratio of the HDTV method is 9 to 16 hours, and the operation can be directly issued to the monitor without using the memory. Further, the number of scanning lines of the frame does not increase, and the operation can be performed in the same manner as the driving frequency of the first move. This embodiment of the towel, 1 _ i昼_ obtained progressive (pr〇greSS1Ve) way to illustrate, but also applies to 2 games! In the written (interlace) way. - Fig. 21 is a diagram showing a 30,000-pixel CMOS image sensor (solid-state imaging device) s (four) path formed on a semiconductor substrate in the third embodiment. The same portions in Fig. 21 as those in Fig. 2 are provided with the same symbols. In Fig. 21, the frame 1253852 15763pif.doc memory 11 and the subtraction circuit 12 are connected to the AD conversion circuit 2, and the memory 11 is connected to the subtraction circuit 12. The power supply to the z-fork Figure 22 is the implementation of the open circuit; + θ one time sequence flow chart. In the figure, the motion of the wide CM0S image sensor! Shown when. The 3rd light: the accumulation time of the polar body is 蓄 Η Η 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 、 From Fig. 22, it can be seen that the accumulation time of the 梃4峨 accumulation time; the f shed horizontal line No. 1 to the medium print discharge and the PD 4 day horizontal line No. 1 to _ (DN) have been read? Lu II: Acting at the same timing. During the inspection period, during the period; in: the secret of the material and (four) the straight signal line f Lu's work A will not be carried out in abundance. In the horizontal line 1 to 4 8 〇 =] =;;:: rN) _ leakage =;:, = fr test becomes a high state, the detection part is _ ,, read by the light diode (PD) In the 2/3 period of the 1 frame (1575H) (or 1 field) when the signal is output to the time of the detection, all the light and polar bodies (PD) are accumulated, and all the photodiodes (PD) (4) The signal charge of /Λ is simultaneously read out to the detection unit (DN). Then, the signal charges of the horizontal lines No. 1 to No. 480 are sequentially read out to the old line (VUN). The vertical signal line π·) is rejected until it is sequentially converted into a digital signal by the AD conversion circuit 2 and stored in and 11 . At this time, the photodiode (PD) is memorized in the frame memory 11, and the signal of the leakage current of the detecting portion (DN) is also stored. 35 1253852 15763pif.doc Then, in the vertical effective scanning period (after the reset of all the detection parts in the 1/s period of the 1 frame, the charges accumulated in the leakage current of the _AD pulse detection unit (DN) are not sequentially applied. Read out to the number line (VLIN). The signal that has been read out to the vertical signal line (service) is converted into a digital signal by the AD conversion circuit 2 to be input to the subtraction power f =

% /7IL. At this time, only the signal of the detecting unit (DN) is input to the subtraction circuit 12. : The signal of the photodiode (PD) of the cell in which the signal of the leakage current input from the subtraction circuit 12 is rotated and the signal of the leakage current of the detecting portion (DN) are read out from the frame memory 11 to the subtraction In circuit 12. The subtraction circuit & subtracts the leakage current of the detection portion (DN) input by the ad conversion circuit 2 from the signal of the photodiode (PD) and the detected leakage current signal read by the frame memory 11. ^吕号 and output the result. Further, the leakage current of the detecting portion can be about 100 times the leakage current of the photodiode (PD). In the above example, the signal of the light-pole (PD) signal and the leakage current of the detecting portion (〇ν) are stored in the frame memory 11, and then the subtraction circuit 12 inputs only the detecting portion. (DN) leakage current signal, but the signal of the photodiode (PD) and the leakage current of the detection unit (DN) may be input to the subtraction circuit 12 to cause the detection unit (DN). The letter of leakage current of the letter of the tiger has been recalled in the box § 丨 丨 11 11 11 to carry out the same reduction process. According to the third embodiment, the leakage current of the detecting portion (DN) is removed, and an image with improved enamel can be obtained. As described above, the solid-state imaging device according to the embodiment of the present invention can improve the enamel. That is, it can improve the above-mentioned "CMOS image sensor 36 1253852 15763plf.doc does not have the simultaneity, the shortcomings. The use of simultaneity. Also, dry 兀 wood ^ and the mouth of the Japanese The low-light photography of the KTC noise is the same as the image of the leakage current of the detection unit (DN). Also, the form of the solid-state imaging device is applied to a digital camera and a digital camera (four) to a portable telephone having a camera. According to the solid-state imaging device of the embodiment of the present invention, a solid-state imaging device with improved enamel, a digital camera and a digital camera can be provided. Although the present invention has been disclosed in the above preferred embodiments, the present invention is not intended to be limited thereto, and the present invention may be modified and modified in the spirit and scope of the present invention. The scope of protection is subject to the definition of the scope of the patent application. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a block diagram showing the configuration of a digital camera to which a solid-state imaging device according to a first embodiment is applied. Fig. 2 is a circuit configuration of a CMOS image sensor in the first embodiment. Fig. 3 is a circuit diagram of a pulse wave selector in the first embodiment. Fig. 4 is a timing chart showing the operation of the pixel unit operating in the mode RMode 1) in the first embodiment. Fig. 5 is a timing chart showing the operation of the pixel unit operating in mode 2 (Mode 2) in the first embodiment. Fig. 6 is a first timing chart for showing the operation of the CMOS image sensor in the first embodiment. 7A to 7E are cross-sectional views and potential diagrams showing the CMOS image sensing 37 1253852 15763plf.d〇c accumulation, time 2 in the first embodiment. The flow chart of the set order. Fig. 9 is a second timing chart for cM in the first embodiment. Fig. 10 shows a circuit configuration related to the third embodiment.

Cx» 〇 ^ is used in the form of digital === solid state. Fig. 12 is a circuit diagram of a pulse wave selector in the circuit configuration of the CM〇S image sensor in the embodiment of the second embodiment. In the form of the L-day application of the Department of Alien, the mode (ES), (R〇) of the model Mei, and the pattern of the operation of the shifting temporary storage unit in the == implementation mode The 17th implementation of the second system is based on the timing chart. In the 乂 type, the vertical shift register (ESX (RO) is the timing diagram of the second embodiment. The 图 颂 CMOS CMOS image sensor's moving picture 19 shows the second real miscellaneous Flowchart of the setting procedure of 彡, "Mode 1 (Mode υ and Mode 2 I253852763p_ Figure 20 is a timing chart showing KTC noise countermeasures in the case of low illumination in the second embodiment. Fig. 21 is a semiconductor in the third embodiment. The circuit configuration of a 330,000 pixel CMOS image sensor formed on a substrate. Fig. 22 is a timing chart showing the operation of the CMOS image sensor in the third embodiment. Fig. 23 is a previous CMOS image sensor. Fig. 24 is a circuit diagram of the previous pulse wave selector. Fig. 25 is a timing chart showing the operation of the previous CMOS image sensor. [Main component symbol description] 1 port - early 2 AD conversion circuit 3 level Shift register 4 Triangle wave generation circuit 5 Vertical shift register (ES) 6 Vertical shift register (RO) 7 , 71 , 72 Pulse selector 8 Timing generation circuit 9 Bias generation circuit 11 Frame memory Body 12 subtraction circuit 81 accumulation time control circuit 82 Pulse wave generating circuit 39 I253852763pif.doc 100 Digital camera 101 ^ 201 Microcomputer 102, 202 Time-series generated video signal processing unit 103 DV converting unit 104 DV video tape unit 105, 205 Liquid crystal monitor 106, 206 Lens 107 Microphone 108 Acoustic signal processing unit 109 , 207 Power Supply Unit 200 Digital Camera 201 Microcomputer 203 Image Data Compression Unit 204 Image Memory PD Light Diode DN Detection Unit S Sensor / Solid State Photography Device Ta, Tb, Tc, Td Transistor

40

Claims (1)

1253852 is the Chinese patent scope of No. 93139773 without a slash correction Year of the month | Amendment date: 2005.12.15 X. Patent application scope: 1. A solid-state photographic device, on which a photodiode on a semiconductor substrate is sold by a haiguang one item "is 5 (gate) and a detection unit for detecting the read signal are arranged in a two-dimensional space as each unit, and the signal is discharged by the photodiode of each unit, and then the four Lu numbers are accumulated by the respective photodiodes. After the accumulated time required for the accumulation, the light-receiving device is read by the above-mentioned photodiodes, and the sturdy and sturdy photographic device has a single electric 峪, and a plurality of horizontal sizings are performed during the second horizontal scanning period. The third pulse of the signal discharge of each photodiode is applied to the relatively long-range single-action, and simultaneously, for each of the plurality of waters in the second horizontal scanning period, each photodiode of the one horizontal line is made. The signal reading and the daily wave signal are applied to the corresponding units. , 11 signal movement: the number of times in the accumulation time becomes shorter =: and brother 2 pulse wave 3 · As described in the scope of patent application, the solid-state fresh-keeping department has a way to make the first action mode + heart In the first embodiment, the switching electric detection unit for switching the light-emitting mode in the first operation mode detects the signal accumulated in the detection j and reads the signal to detect the detection unit. After the reset = level 1 , the difference between the check level and the reset level is diverged and then the signal is read out from the photodiode by the photodiode. The second operation mode detection unit reset level, after the fall ° 2 丄 5 hai detection part reset, to detect the 4 accumulated signal in the money diode read 15763pif2.doc 41 12: Race · 尸 j 94; 12. Year == the signal level of the quasi-portion, and then output the signal that has been divided into the reference circuit, and switch to the first product time when the action mode 4θ L is less than 1 When the time is 1 or more, the signal is switched to the second state. For example, the solid-state imaging device described in the patent shaft 1 is used, and the signal of the pole body of 1 is discharged and simultaneously read. Xiangshou (4) Dou: Buy the pulse wave in the 1 frame or 1 field during the η field... ♦ 2 times or more, add the signal continued by the photodiode and turn it out. (4) The light diode of each unit of coughing out causes the signal to be placed: *Bu, 10, 21 " an accumulation nickname 'The accumulated time required for the accumulation is read by the above-mentioned respective diodes to - after the detecting portion, the portion, the 'number', the photodiode on the t semiconductor substrate, the gate for the tiger to be bought by the photodiode 5, and the signal for reading out The detecting unit is configured as a unit in a two-dimensional manner, and the solid-state imaging device includes a circuit that performs a plurality of horizontal lines during the horizontal scanning period—recording a level_signal discharge for each photodiode (4) 1 When the pulse wave is applied to the operation of each unit of the subtraction and simultaneously performed in the case of a plurality of horizontal lines in the second horizontal scanning period, the second pulse signal for reading the signals of the respective photodiodes of the one horizontal line is performed. The action applied to each of the corresponding units is performed every time the horizontal return line of the shirt is performed. The signal output of the detection unit of the horizontal line is applied to the corresponding pulse signal of 15763 pi0.doc 42 The action of the unit. 7. The action of solid-state photography as described in item 6 of the patent application. The signal of each of the first polar bodies of the J^ line is simultaneously read by the solid-state photography skirt according to item 6 of the range of the electric field, wherein the electric r is placed in the red eye, wherein each of the groups is carried out In the second horizontal line, the action of sequentially discharging the m-line in the second level is performed. 'The signals of the respective photodiodes are grouped in the flat line. 10. The operation is sequentially read out as the application axis r _ line. With - kind of make the first! The solid-state imaging device according to the second aspect of the invention, wherein the path "the first operation mode in the middle of the eclipse": the switching electrical detection portion for switching the upper operation mode to detect the detection portion; Go to the set to detect the reset two level of the detection part and then the difference between the check level and the reset level, and then make the detected signal The signal of the light i is rotated; the reset level of the second operation mode detecting unit is reset, and the detecting unit is reset to detect the output to the detecting unit, so that “the accumulated in the second detecting unit The 邛δ level of the signal is read, and then it is cut by 15763pif2.doc 43; the reference signal is output, I action mode, and the mode is switched to the first action mode when the first time is less than the first time. When the time is longer than the time, the light is switched to the second type, and the light period of the one horizontal line is performed - the flat return line period is pushed to the electric operation, and the second water is discharged. > The tribute of each horizontal diode of the horizontal line The signal is read by each H Γ Γ 第 第 第 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 The emitter body f ί signal to the detection of the neighboring party 罟 ^ ^ l5 tiger, and the signal output of the brother 1 signal is not read by the photodiode after the accumulation time ^ signal and money, off The solid-state photographic device described in the above-mentioned accumulation item, wherein the solid-state photographic device described in item 6 of the private full-period or vertical return line, wherein the solid-state photographic device described in claim 6 ;ϊ=, the device of the pulse wave in 1 frame or 1 field 1 The patent application mentioned in the first paragraph of the patent range 15763pif2.doc 44 I_3S meal 2 exchange 1 R 94 12, 1 i year month 曰 ^mmmtmyww____ι_ιιτι- :·*__*, _>, t 16. A digital camera using the solid-state imaging device according to claim 6 of the patent application. 17. A solid-state imaging device as claimed in claim 1 Digital camera 18. A digital camera using a solid-state imaging device as described in claim 6. 45 15763pif2.doc