JP7567655B2 - Imaging device - Google Patents

Description

This disclosure relates to an imaging device used in fields such as remote sensing.

Linear image sensors have been developed for mounting on flying objects such as artificial satellites to observe the Earth's surface from the sky. For remote sensing of the Earth's surface, a linear image sensor is mounted on an artificial satellite and images are taken by aligning the vertical scanning direction (Along Track: AT direction) of the linear image sensor with the direction of the satellite's movement. The direction perpendicular to the vertical scanning direction is called the horizontal transfer direction (Cross Track: CT direction).

To obtain high-resolution imaging data, it is desirable to reduce the pixel pitch of the linear image sensor. However, as the pixel area shrinks, the amount of light incident on the pixel decreases, resulting in a problem of degradation of the signal-to-noise ratio (S/N). As a measure to improve the S/N ratio, a time-delay integration (TDI) type image sensor has been developed. A TDI type image sensor uses a two-dimensional image sensor, a full-frame transfer (FFT) type charge-coupled device (CCD), and improves the S/N ratio by synchronizing the charge transfer with the movement of the subject image. In the case of remote sensing, TDI operation can be achieved by matching the vertical charge transfer to the moving speed of the satellite. When M stages of TDI operation are performed with a vertical CCD, the effective accumulation time becomes M times, so the sensitivity is improved by M times and the S/N ratio is improved by √M times.

In addition, to improve image resolution without reducing the pixel pitch, an imaging device has been developed that performs oversampling by arranging two linear image sensors with the same pixel pitch, shifted by 1/2 pixel pitch from each other. A high-resolution image can be obtained by image processing the output signals obtained from the two image sensors. As the pixel area is maintained, there is no loss of sensitivity.

A conventional imaging device has been disclosed that has multiple image reading devices arranged in the horizontal direction, each of which performs oversampling using two TDI linear image sensors (see, for example, Patent Document 1). This expands the observation width of the TDI linear image sensor.

International Publication No. 2018/211622

However, in conventional imaging devices, of the two TDI linear image sensors arranged in the horizontal direction, the output signal wiring of the vertically upper linear image sensor must be routed around the lower linear image sensor to the pad. As the wiring length increases, the wiring capacitance increases, which causes a problem of increased noise being mixed into the signal. In addition, if a misalignment occurs between the satellite's direction of travel and the linear image sensor's vertical direction due to vibrations during imaging, a misalignment occurs in the imaging data of each linear image sensor. As a result, there is also the problem that oversampling processing becomes difficult.

This disclosure has been made to solve the problems described above, and its purpose is to obtain an imaging device that can prevent degradation of the S/N ratio and deviations in imaging data.

The imaging device according to the present disclosure includes a first linear image sensor of a TDI type that detects incident light, a second linear image sensor of a TDI type that is stacked on the first linear image sensor in the traveling direction of the incident light and detects the incident light that has passed through the first linear image sensor, and a drive pulse control unit, the first and second linear image sensors each having pixels that are arranged in a two-dimensional array in the vertical and horizontal directions and that photoelectrically convert the incident light, a vertical transfer unit that time-delay integrates charges generated by the pixels and vertically transfers them, a horizontal transfer unit that horizontally transfers the time-delay integrated charges, and an amplifier that reads out charges from the horizontal transfer unit, and the drive pulse control unit is configured to detect the second linear image sensor. a reset clock pulse in the readout operation of the amplifier of the image sensor is intermittently controlled with respect to a transfer clock pulse of the horizontal transfer section, the horizontal pixel pitch of the first linear image sensor is an integer multiple of two or more that of the second linear image sensor, the horizontal pixel pitch of the first linear image sensor is twice that of the second linear image sensor, two horizontally adjacent pixels of the second linear image sensor overlap one pixel of the first linear image sensor, and the output signal of the second linear image sensor is read out with a 1/2 pitch shift with respect to the output signal of the first linear image sensor .

In this disclosure, two TDI type linear image sensors are stacked in the direction of propagation of the incident light. This allows the signal wiring drawn from each linear image sensor to be short, preventing degradation of the S/N ratio. In addition, stacking two linear image sensors can also prevent misalignment of the imaging data.

1 is a block diagram showing an imaging device according to a first embodiment; FIG. 2 is a perspective view showing a focal plane detector according to the first embodiment; FIG. 3 is a cross-sectional view taken along line I-II of FIG. FIG. 3 is a cross-sectional view taken along line III-IV in FIG. 4 is a schematic diagram showing the TDI operation of the first and second linear image sensors according to the first embodiment; FIG. 4 is a schematic diagram showing how charges accumulated in horizontal transfer sections of the first and second linear image sensors according to the first embodiment are stored in an imaging data storage memory; FIG. 4 is a schematic diagram showing clocks for the operations of the first and second linear image sensors according to the first embodiment; FIG. 4 is an explanatory diagram showing, in time series, the driving of a transfer clock of the first linear image sensor according to the first embodiment and the associated transfer of charges; FIG. 10 is an explanatory diagram showing, in time series, the driving of a transfer clock of the second linear image sensor according to the first embodiment and the associated transfer of charges; FIG. 11 is a schematic diagram showing the TDI operation of the first and second linear image sensors according to the second embodiment. FIG. 11 is a schematic diagram showing how charges accumulated in the horizontal transfer sections of the first and second linear image sensors according to the second embodiment are stored in an imaging data storage memory; FIG. 11 is a schematic diagram showing clocks for the operations of the first and second linear image sensors according to the second embodiment. FIG. 13 is an explanatory diagram showing, in time series, the driving of a transfer clock of the second linear image sensor according to the second embodiment and the associated transfer of charges; FIG. FIG. 11 is a perspective view showing a focal plane detector according to a third embodiment. 15 is a cross-sectional view taken along line I-II of FIG. 14. FIG. 15 is a cross-sectional view taken along line III-IV in FIG. 11 is a schematic diagram showing the TDI operation of the first and second linear image sensors according to the third embodiment. FIG. FIG. 13 is a perspective view showing a focal plane detector according to a fourth embodiment. 19 is a cross-sectional view taken along line I-II of FIG. 18. 13 is a schematic diagram showing how charges accumulated in the horizontal transfer sections of the first and second linear image sensors according to the fourth embodiment are stored in an imaging data storage memory; FIG. 13 is an explanatory diagram showing, in time series, the driving of a transfer clock of a second linear image sensor according to a fourth embodiment and the associated transfer of charges; FIG.

The imaging device according to the embodiment will be described with reference to the drawings. The same or corresponding components will be given the same reference numerals, and repeated description may be omitted.

Embodiment 1.
1 is a block diagram showing an image pickup device according to embodiment 1. The image pickup device 100 is mounted on an airborne vehicle such as an artificial satellite. The image pickup device 100 includes an optical system 110, a focal plane detector 120, an attitude detection unit 130, an attitude control unit 140, and a drive pulse control unit 150.

The optical system 110 forms an image of the incident light on a focal plane, and is, for example, a telescope made up of optical elements such as lenses and mirrors. The focal plane detector 120 is an image sensor such as a TDI type CCD or CMOS that detects the image formed on the focal plane by the optical system 110. Here, the focal plane detector 120 has a TDI type linear image sensor.

The attitude detection unit 130 detects the attitude of the imaging device 100 mounted on the flying object. The attitude information detected by the attitude detection unit 130 is transmitted to the attitude control unit 140. The attitude control unit 140 transmits the attitude information detected by the attitude detection unit 130 to the drive pulse control unit 150. The attitude of the imaging device 100 is changed according to the attitude information transmitted from the ground station 200.

The drive pulse control unit 150 controls the number of TDI integration stages of the TDI linear image sensor of the focal plane detector 120 according to the attitude information of the image capture device 100 transmitted from the attitude control unit 140. The transmission/reception processing unit 160 receives the attitude control signal transmitted from the ground station 200, transmits it to the attitude control unit 140, and transmits the attitude state of the image capture device 100 to the ground station 200. The ground station 200 is a ground station that transmits and receives various control signals and status signals of the image capture device 100 to and from the image capture device 100.

Figure 2 is a perspective view showing a focal plane detector according to the first embodiment. The focal plane detector 120 has a first linear image sensor 1 of the TDI type and a second linear image sensor 2 of the TDI type stacked on the first linear image sensor 1 in the direction of propagation of the incident light.

The substrate thickness of the first linear image sensor 1 is thinner than that of the second linear image sensor 2. This allows the first linear image sensor 1 to obtain an image in the short wavelength band that can be received by a thin substrate, while the second linear image sensor 2 can obtain an image in the long wavelength band that transmits through a thin substrate. For example, the substrate thickness of the first linear image sensor 1 is made thinner than 50 um so that visible light from red to near-infrared light can be transmitted. On the other hand, the substrate thickness of the second linear image sensor 2 is made thick enough to detect red to near-infrared light.

The incident light guided by the optical system 110 is incident on the first linear image sensor 1 of the focal plane detector 120. The first linear image sensor 1 detects incident light with a wavelength shorter than that of red light. The second linear image sensor 2 detects incident light with a wavelength longer than that of red light that has passed through the first linear image sensor 1.

In the first and second linear image sensors 1 and 2, pixels 3 that photoelectrically convert incident light are arranged in a two-dimensional array in the vertical and horizontal directions. The vertical transfer unit 4 performs time delay integration of the charges generated by the pixels 3 and transfers them vertically. The vertical transfer unit 4 is a CCD (Charge Coupled Device) having vertical transfer gate electrodes 4a and 4b and a storage gate electrode 4c. The horizontal transfer unit 5 horizontally transfers the time delay integrated charges. The FDA 6 (floating diffusion amplifier) reads out the horizontally transferred signal from the horizontal transfer unit 5. The drive pulse generation unit 7 generates clocks for each operation in synchronization with the drive pulse signal input from the drive pulse control unit 150.

The horizontal pixel pitch of the first linear image sensor 1 is designed to be twice the horizontal pixel pitch of the second linear image sensor 2. The pitch of the horizontal transfer section 5 of the first linear image sensor 1 is also designed to be twice the pitch of the horizontal transfer section 5 of the second linear image sensor 2. The first linear image sensor 1 and the second linear image sensor 2 are stacked so that two horizontally adjacent pixels of the second linear image sensor 2 exactly overlap one pixel of the first linear image sensor 1.

The charge generated by the first linear image sensor 1 upon receiving light is transferred vertically by a vertical transfer clock generated by the drive pulse generating unit 7, and sent to the horizontal transfer unit 5. The horizontal transfer unit 5 transfers the charge horizontally by a horizontal transfer clock. The charge transferred to the final stage of the horizontal transfer unit 5 is read out as an output signal by the FDA 6. The output signal is converted into digital data by the A/D converter 8. The digital data is stored in the image data storage memory 9. The second linear image sensor 2 also performs transfer and read operations by the same means as the first linear image sensor 1. The output signal of the second linear image sensor 2 is also A/D converted and stored in the image data storage memory 9.

The horizontal pixel pitch of the first linear image sensor 1 is not limited to twice the horizontal pixel pitch of the second linear image sensor 2, and may be an integer multiple of 2 or more. The pitch of the horizontal transfer section 5 of the first linear image sensor 1 is also not limited to twice the horizontal pixel pitch of the second linear image sensor 2, and may be an integer multiple of 2 or more. However, the multiple ratio of the horizontal pixel pitch and the pitch of the horizontal transfer section 5 of the first linear image sensor 1 and the second linear image sensor 2 is the same. In other words, if the multiple of the horizontal pixel pitch is 3, the multiple of the pitch of the horizontal transfer section 5 is also 3.

Figure 3 is a cross-sectional view taken along line I-II in Figure 2. This cross section corresponds to the horizontal transfer section 5 of the first and second linear image sensors 1 and 2. The horizontal transfer section 5 is a CCD having horizontal transfer gate electrodes 5a and 5b. A drive pulse signal for driving the sensor is given from the bottom surface of the second linear image sensor 2 and is transmitted to the first linear image sensor 1 via the through electrode 10. The charge transferred vertically from the pixel 3 is transferred from right to left in the drawing in the buried channel 11 and output to the outside of the sensor via the through electrode 12 by the FDA 6. Two CCDs of the second linear image sensor 2 are arranged directly below one CCD of the first linear image sensor 1. This allows the image of the second linear image sensor 2 to have twice the resolution of that of the first linear image sensor 1.

Figure 4 is a cross-sectional view taken along III-IV in Figure 2. This cross section corresponds to the pixel section of the first and second linear image sensors 1 and 2. A photodiode 13 is provided in each pixel 3. Charges generated by the photodiode 13 in response to incident light are transferred from right to left in the drawing within the buried channel 11 and sent to the horizontal transfer section 5. One pixel 3 of the second linear image sensor 2 is disposed directly below one pixel 3 of the first linear image sensor 1.

Figure 5 is a schematic diagram showing the TDI operation of the first and second linear image sensors according to embodiment 1. For a pixel of interest in the first linear image sensor 1, the pixel of interest in the second linear image sensor 2 corresponds to the pixel directly below and to the right of the first linear image sensor 1 and the pixel adjacent to that pixel to the right. The pixel of interest in the first linear image sensor 1 and the pixel of interest in the second linear image sensor 2 are shifted in position by half a pitch.

Figure 6 is a schematic diagram showing the process in which the charges stored in the horizontal transfer section of the first and second linear image sensors according to embodiment 1 are stored in the imaging data storage memory. Charges <1> to <5> are stored in the horizontal transfer section 5 of the first linear image sensor 1. These are transferred horizontally, read out by the FDA 6, converted to digital, and then stored in a specified address in the imaging data storage memory 9. Meanwhile, charges (1) to (10) are stored in the horizontal transfer section 5 of the second linear image sensor 2. These are transferred horizontally, read out by the FDA 6, converted to digital, and then stored in a specified address in the imaging data storage memory 9.

In the second linear image sensor 2, the charges of two adjacent horizontal transfer sections 5 are integrated in the FD to perform a readout operation. This integration operation will be described later. The read-out output signal of the second linear image sensor 2 is shifted by half a pitch compared to the output signal of the first linear image sensor 1 due to the integration operation. Note that, as an address designation method for storing the output signal of the second linear image sensor 2 in the imaging data storage memory 9, the address is shifted in the drawing to make it easier to distinguish that the data is shifted by half a pitch, but the address designation method is not limited to this.

Figure 7 is a schematic diagram showing the clocks for each operation of the first and second linear image sensors according to the first embodiment. The charges stored in the horizontal transfer section 5 of the first linear image sensor 1 are transferred to the FD in response to the transfer clock CLK1. The charges transferred to the FD are converted to a voltage by the FDA 6 and read out. This read operation is triggered by the sample and hold clock CLK2. After the read operation, the potential of the FD is reset before the next charge is transferred. This operation is triggered by the reset clock CLK3. Note that the read to reset operation of the horizontal transfer section 5 of the first linear image sensor 1 is a typical CCD operation.

The electric charge accumulated in the horizontal transfer section 5 of the second linear image sensor 2 is transferred to the FD in response to the transfer clock CLK4. The electric charge transferred to the FD is converted to a voltage by the FDA6 and read out. This read operation is triggered by the sample and hold clock CLK5. After the read operation, the electric potential of the FD is reset before the next electric charge is transferred. This operation is triggered by the reset clock CLK6.

In the read and reset operations of the second linear image sensor 2, the sample and hold clock CLK5 and the reset clock CLK6 are driven to generate a pulse every two clocks of the transfer clock CLK4. According to this operation, the charges (2) and (3) transferred by the transfer clock CLK4 are added together in the FD.

Figure 8 is an explanatory diagram showing the driving of the transfer clock of the first linear image sensor according to the first embodiment and the accompanying transfer of charges in a time series. At time t1, the reset clock φR makes the potential of the FD equal to the reset potential VR. Next, at time t2, the transfer clocks φH1 and φH2 operate to transfer the accumulated charge to the adjacent charge pocket on the right. Next, the transfer clock operation at time t3 transfers the accumulated charge further to the adjacent charge pocket on the right, and the charge is transported to the FD. FDA6 reads out the charge.

Next, at time t4, a reset operation resets the potential of the FD to the same potential as the reset potential VR. Next, at time t5, similar to time t2, the accumulated charge is transferred to the adjacent charge pocket on the right by the operation of the transfer clock φH1 and clock φH2. From time t6 onwards, the charge transfer and readout operation from time t3 onwards continues. As described above, the first linear image sensor 1 sequentially transfers and reads out the accumulated charges in the horizontal transfer section 5.

Figure 9 is an explanatory diagram showing the driving of the transfer clock of the second linear image sensor according to embodiment 1 and the accompanying transfer of charges in a time series. At time t1, the reset clock φR makes the potential of the FD equal to the reset potential VR. Next, at time t2, the transfer clocks φH1 and φH2 operate to transfer the accumulated charge to the adjacent charge pocket on the right. Next, the transfer clock operation at time t3 transfers the accumulated charge further to the adjacent charge pocket on the right, and the charge is transported to the FD. FDA6 reads out the charge.

Next, at time t4, the potential of the FD is reset to the same potential as the reset potential VR by a reset operation. Next, at time t5, the accumulated charge is transferred to the charge pocket on the right by the transfer clock φH1 and clock φH2 operation, similar to time t2. Next, at time t6, charge transfer is performed similar to time t3, but the charge is not read out in the FD. Next, at time t7, the accumulated charge is transferred to the charge pocket on the right by the transfer clock φH1 and clock φH2 operation without generating a reset clock φR pulse. Therefore, the potential of the FD is not reset to the reset potential VR. Next, the accumulated charge is transported to the FD by a transfer operation at time t8, and the FD reads the charge. This operation allows the sum of the two adjacent accumulated charges to be read out at time t8. Next, at time t9, the potential of the FD is reset. After time t10, the same operation as after time t2 is repeated.

As described above, in this embodiment, the first and second linear image sensors 1 and 2 of the TDI method are stacked in the direction in which the incident light travels. Therefore, the signal wiring drawn from the first and second linear image sensors 1 and 2 can be shortened, thereby preventing deterioration of the S/N ratio. In addition, by stacking the first and second linear image sensors 1 and 2, it is also possible to prevent deviations in the imaging data.

The horizontal pixel pitch of the first linear image sensor 1 is an integer multiple of 2 or more that of the second linear image sensor 2. The reset clock pulse in the readout operation of the FDA 6 of the second linear image sensor 2 is intermittently controlled with respect to the transfer clock pulse of the horizontal transfer section 5. This makes it possible to obtain, for one subject image, reference imaging data and imaging data with a pixel pitch that is 1/integer multiple of that.

In addition, the drive frequency of the horizontal transfer section 5 of the second linear image sensor 2 is twice that of the first linear image sensor 1. This allows the output signals of the first and second linear image sensors 1 and 2 to be acquired in synchronization.

The output signal of the second linear image sensor 2 is read out with a 1/2 pitch shift relative to the output signal of the first linear image sensor 1. By oversampling the read-out output signals of the first and second linear image sensors 1 and 2, a high-resolution image can be obtained without reducing the pixel pitch and decreasing the S/N ratio.

In addition, in order to drive the stacked first and second linear image sensors 1 and 2 in accurate synchronization, it is required to input the same drive signal. For this purpose, it is necessary to electrically connect the stacked first and second linear image sensors 1 and 2 with short wiring. Therefore, the first and second linear image sensors 1 and 2 are electrically connected via the through electrodes 10 and 12 that penetrate the first linear image sensor 1 or the second linear image sensor 2. The first and second linear image sensors 1 and 2 are driven in synchronization by the same drive pulse signal via the through electrode 10. In addition, since the input terminals of the drive pulse signals of the first and second linear image sensors 1 and 2 can be shared, the number of terminals can be reduced. In addition, it is possible to simplify the drive pulse control unit 150 that supplies the drive clock pulse signals of the first and second linear image sensors 1 and 2.

In addition, the first and second linear image sensors 1 and 2 are preferably back-illuminated sensors that receive light from the back surface of the substrate. Since there is no electrode on the back surface of the substrate that blocks light, the back-illuminated sensor can receive all light that enters the light receiving surface. However, only one of the first and second linear image sensors 1 and 2 may be a back-illuminated sensor. In addition, the back-illuminated sensor is most effective when receiving light that is photoelectrically converted in a very shallow part of the silicon substrate, such as the short wavelength band of visible light. On the other hand, near-infrared light penetrates deep into the silicon substrate and is photoelectrically converted. For this reason, it is preferable to make the first linear image sensor 1 thin and back-illuminated, and the second linear image sensor 2 a front-illuminated type. As a result, the first linear image sensor 1 receives all light in the short wavelength band, while the second linear image sensor 2 receives the light that has passed through the first linear image sensor 1, thereby receiving all light in a wide wavelength band.

Embodiment 2.
10 is a schematic diagram showing the TDI operation of the first and second linear image sensors according to embodiment 2. For a pixel of interest in the first linear image sensor 1, the pixel of interest in the second linear image sensor 2 corresponds to two adjacent pixels located immediately below the left and right of the first linear image sensor 1. Therefore, the pixel of interest in the second linear image sensor 2 and the pixel of interest in the first linear image sensor 1 are in a vertically overlapping positional relationship.

Figure 11 is a schematic diagram showing the process in which the charges stored in the horizontal transfer section of the first and second linear image sensors according to embodiment 2 are stored in the imaging data storage memory. Charges <1> to <5> are stored in the horizontal transfer section 5 of the first linear image sensor 1. These are transferred horizontally, read out by the FDA 6, A/D converted, and then stored in a specified address in the imaging data storage memory 9. Meanwhile, charges (1) to (10) are stored in the horizontal transfer section 5 of the second linear image sensor 2. These are transferred horizontally, read out by the FDA 6, A/D converted, and then stored in a specified address in the imaging data storage memory 9.

In the second linear image sensor 2, the charges of two adjacent horizontal transfer sections 5 are integrated by the FD to perform a readout operation. The read-out output signal of the second linear image sensor 2 is not misaligned with the output signal of the first linear image sensor 1 due to the integration operation. Note that, as an address designation method for storing the output signal of the second linear image sensor 2 in the imaging data storage memory 9, the address is shown to match the data of the first linear image sensor 1 in the drawing to make it easier to determine that there is no misalignment, but the address designation method is not limited to this.

Figure 12 is a schematic diagram showing the clocks for each operation of the first and second linear image sensors according to embodiment 2. The transfer clock CLK1, sample-and-hold clock CLK2, and reset clock CLK3 of the first linear image sensor 1 are the same as those in Figure 7, so a description thereof will be omitted.

The electric charge accumulated in the horizontal transfer section 5 of the second linear image sensor 2 is transferred to the FD in response to the transfer clock CLK4. The electric charge transferred to the FD is converted to a voltage by the FDA6 and read out. This read operation is triggered by the sample and hold clock CLK5. After the read operation, the electric potential of the FD is reset before the next electric charge is transferred. This operation is triggered by the reset clock CLK6.

In the read and reset operations of the second linear image sensor 2, the sample and hold clock CLK5 and the reset clock CLK6 are driven to generate a pulse every two clocks of the transfer clock CLK4. According to this operation, the charges (1) and (2) transferred by the transfer clock CLK4 are added together in the FD.

Figure 13 is an explanatory diagram showing the driving of the transfer clock of the second linear image sensor according to the second embodiment and the accompanying transfer of charges in a time series. At time t1, the reset clock φR makes the potential of the FD equal to the reset potential VR. Next, at time t2, the transfer clocks φH1 and φH2 operate to transfer the accumulated charge to the adjacent charge pocket on the right. Next, the transfer clock operation at time t3 transfers the accumulated charge further to the adjacent charge pocket on the right, and the charge is carried to the FD. Here, no charge is read out from the FD.

Next, at time t4, the accumulated charge is transferred to the adjacent charge pocket on the right by the operation of the transfer clocks φH1 and φH2, without generating a pulse of the reset clock φR. At time t5, similar to time t3, the accumulated charge is transported to the FD by the operation of the transfer clocks φH1 and φH2. Here, FDA6 executes a read operation. With this operation, the combined value of two adjacent accumulated charges can be read out at time t5. Next, at time t6, the potential of the FD is reset. From time t7 onwards, the same operation as from time t2 onwards is repeated.

In this embodiment, clock generation is controlled to synchronize the reset clock pulses of the readout circuits of the first and second linear image sensors 1 and 2, and the output signals of the first and second linear image sensors 1 and 2 are read out synchronously without any shift. By adding up the readout signals, it is possible to expand the dynamic range and obtain a high-sensitivity image.

In addition, by switching the reset clock pulse of FDA6, which is the readout circuit of the first and second linear image sensors 1 and 2, it is possible to selectively read out the output signal of the second linear image sensor 2 with a 1/2 pitch shift from the output signal of the first linear image sensor 1, or to read out the output signals of the first and second linear image sensors 1 and 2 synchronously. This makes it possible to selectively obtain a high-definition image by oversampling in embodiment 1 and an image with a wide dynamic range by combining the output signals in embodiment 2 with a single imaging device.

Embodiment 3.
14 is a perspective view showing a focal plane detector according to embodiment 3. In this embodiment, not only the horizontal pixel pitch of the first linear image sensor 1 but also the vertical pixel pitch is designed to be twice that of the second linear image sensor 2. The first linear image sensor 1 and the second linear image sensor 2 are stacked so that two horizontal pixels by two vertical pixels of the second linear image sensor 2 exactly overlap one pixel of the first linear image sensor 1. The other configurations are the same as those of embodiment 1.

The horizontal pixel pitch of the first linear image sensor 1 is not limited to twice the horizontal pixel pitch of the second linear image sensor 2, and may be an integer multiple of 2 or more. The vertical pixel pitch of the first linear image sensor 1 is not limited to twice the vertical pixel pitch of the second linear image sensor 2, and may be an integer multiple of 2 or more. The pitch of the horizontal transfer section 5 of the first linear image sensor 1 is also not limited to twice the pitch of the horizontal transfer section 5 of the second linear image sensor 2, and may be an integer multiple of 2 or more. However, the multiple ratios of the horizontal pixel pitches of the first linear image sensor 1 and the second linear image sensor 2, the vertical pixel pitches of the first linear image sensor 1 and the second linear image sensor 2, and the pitch of the horizontal transfer section 5 are the same. In other words, if the multiple of the horizontal pixel pitch is 3, the multiples of the vertical pixel pitch and the pitch of the horizontal transfer section 5 are also 3.

Figure 15 is a cross-sectional view taken along line I-II in Figure 14. This cross section corresponds to the horizontal transfer section 5 of the first and second linear image sensors 1 and 2. Optical waveguides 14 and light reflecting layers 15 are formed alternately side-by-side between the first and second linear image sensors 1 and 2. Two CCDs of the second linear image sensor 2 are arranged directly below one CCD of the first linear image sensor 1. This allows the image of the second linear image sensor 2 to have twice the resolution of that of the first linear image sensor 1.

Figure 16 is a cross-sectional view taken along III-IV in Figure 14. This cross section corresponds to the pixel portion of the first and second linear image sensors 1 and 2. Two pixels 3 of the second linear image sensor 2 are arranged directly below one pixel 3 of the first linear image sensor 1.

Figure 17 is a schematic diagram showing the TDI operation of the first and second linear image sensors according to embodiment 3. For the pixel of interest of the first linear image sensor 1, the pixel of interest of the second linear image sensor 2 corresponds to the four pixels located directly below the first linear image sensor 1. Therefore, the pixel of interest of the second linear image sensor 2 and the pixel of interest of the first linear image sensor 1 are in a vertically overlapping positional relationship.

The vertical transfer unit 4 integrates the charge by TDI operation. However, when capturing an image of a very bright subject, the amount of accumulated charge due to TDI operation may become too large, resulting in overflow of excess charge such as blooming or smear. Therefore, in this embodiment, the vertical pixel pitch of the first linear image sensor 1 is doubled that of the second linear image sensor 2, thereby reducing the amount of accumulated charge per pixel. The drive frequency of the vertical transfer unit 4 of the second linear image sensor 2 is also double that of the first linear image sensor 1. This makes it possible to prevent the accumulation of excess charge.

Embodiment 4.
FIG. 18 is a perspective view showing a focal plane detector according to the fourth embodiment. FIG. 19 is a cross-sectional view taken along line I-II in FIG. 18. In the second linear image sensor 2, two horizontal transfer units 5-1, 5-2 are arranged in parallel. Charges transferred from the storage gate electrode 4c are transferred to the horizontal transfer unit 5-1, and then transferred to the horizontal transfer unit 5-2 via the inter-horizontal CCD transfer electrode 5-3. The end of the first linear image sensor 1 is covered with a light-shielding film 16. The other configurations are the same as those in the third embodiment.

Figure 20 is a schematic diagram showing the process in which the charges stored in the horizontal transfer sections of the first and second linear image sensors according to embodiment 4 are stored in the imaging data storage memory. Charges (1) to (10) are stored in the two horizontal transfer sections 5-1 and 5-2 of the second linear image sensor 2. These are transferred horizontally, read out by the FDA 6, and after A/D conversion, are stored in a specified address in the imaging data storage memory 9. The read output signal of the second linear image sensor 2 is shifted by half a pitch from the output signal of the first linear image sensor 1 due to an integration operation. Note that, as an address designation method for storing the output signal of the second linear image sensor 2 in the imaging data storage memory 9, the address is shifted in the drawing to make it easier to distinguish that the data is shifted by half a pitch, but the address designation method is not limited to this.

FIG. 21 is an explanatory diagram showing the driving of the transfer clock of the second linear image sensor according to the fourth embodiment and the associated charge transfer in a time series. The charge accumulated in the pixel 3 is transferred to the bottom of the storage gate electrode 4c by the operation from time t1 to t4. At time t5, the transfer clock φH1 is set to H, and the accumulated charge is transferred to the horizontal transfer unit 5-1. At time t6, the clock φSC is set to L, and the charge transfer to the horizontal transfer unit 5-1 is completed. At time t7, the charge accumulated in the pixel 3 is transferred to just before the storage gate electrode 4c. At this time, the charge accumulated in the horizontal transfer unit 5-1 is transferred in the horizontal direction sequentially, and all the charge in the horizontal transfer unit 5-1 is discharged. At time t8, the clock φSC is set to H, and the charge is transferred to the bottom of the storage gate electrode 4c. At time t9, the clock φV2 is set to L, and the charge transfer to the bottom of the storage gate electrode 4c is completed. At time t10, the transfer clock φH1 is set to H, and the accumulated charge is transferred to the horizontal transfer unit 5-1. At time t11, the clock φSC goes low, completing the charge transfer to the horizontal transfer unit 5-1. At time t12, the potential φTr of the horizontal inter-CCD transfer electrode 5-3 goes high, transferring the charge to below the horizontal inter-CCD transfer electrode 5-3. At time t13, the potential φTr of the horizontal inter-CCD transfer electrode 5-3 goes low, completing the charge transfer to below the horizontal inter-CCD transfer electrode 5-3. At time t14, the clock φH2 goes high, transferring the accumulated charge to the horizontal transfer unit 5-2. At time t14, the clock φH2 goes low, completing the charge transfer to the horizontal transfer unit 5-2.

In the first to third embodiments, the amount of accumulated charge of the pixel 3 located at the left end of the pixels 3 arranged in a two-dimensional array of the first and second linear image sensors 1 and 2 may be A/D converted and stored in memory, and the two may be compared and determined in a later process. This makes it possible to detect the presence or absence of a misalignment in the imaging data of the first and second linear image sensors 1 and 2.

1 First linear image sensor, 2 Second linear image sensor, 3 Pixel, 4 Vertical transfer section, 5 Horizontal transfer section, 6 FDA, 8 A/D converter, 9 Image data storage memory, 10, 12 Through electrode, 150 Drive pulse control section

Claims (11)

a first linear image sensor of a TDI type for detecting incident light;
a second linear image sensor of a TDI type that is stacked on the first linear image sensor in a traveling direction of the incident light and detects the incident light that has passed through the first linear image sensor;
A drive pulse control unit;
The first and second linear image sensors each include pixels arranged in a two-dimensional array in a vertical direction and a horizontal direction and performing photoelectric conversion on the incident light, a vertical transfer section which performs time-delay integration on charges generated by the pixels and vertical transfers the charges, a horizontal transfer section which horizontally transfers the time-delay integrated charges, and an amplifier which reads out the charges from the horizontal transfer section;
the drive pulse control unit intermittently controls a reset clock pulse in a read operation of the amplifier of the second linear image sensor with respect to a transfer clock pulse of the horizontal transfer unit;
a pixel pitch in the horizontal direction of the first linear image sensor is an integer multiple of that of the second linear image sensor, and
the pixel pitch of the first linear image sensor in the horizontal direction is twice that of the second linear image sensor;
two horizontally adjacent pixels of the second linear image sensor overlap one pixel of the first linear image sensor;
an output signal from said second linear image sensor being shifted by 1/2 pitch from an output signal from said first linear image sensor, said image pickup apparatus comprising: a first linear image sensor of a TDI type for detecting incident light;
a second linear image sensor of a TDI type that is stacked on the first linear image sensor in a traveling direction of the incident light and detects the incident light that has passed through the first linear image sensor;
A drive pulse control unit;
The first and second linear image sensors each include pixels arranged in a two-dimensional array in a vertical direction and a horizontal direction and performing photoelectric conversion on the incident light, a vertical transfer section which performs time-delay integration on charges generated by the pixels and vertical transfers the charges, a horizontal transfer section which horizontally transfers the time-delay integrated charges, and an amplifier which reads out the charges from the horizontal transfer section;
the drive pulse control unit intermittently controls a reset clock pulse in a read operation of the amplifier of the second linear image sensor with respect to a transfer clock pulse of the horizontal transfer unit;
a pixel pitch in the horizontal direction of the first linear image sensor is an integer multiple of that of the second linear image sensor, and
the pixel pitch of the first linear image sensor in the horizontal direction is twice that of the second linear image sensor;
two horizontally adjacent pixels of the second linear image sensor overlap one pixel of the first linear image sensor;
an image pickup apparatus comprising: a first linear image sensor and a second linear image sensor, the output signals of which are read out in synchronization with each other; a first linear image sensor of a TDI type for detecting incident light;
a second linear image sensor of a TDI type that is stacked on the first linear image sensor in a traveling direction of the incident light and detects the incident light that has passed through the first linear image sensor;
A drive pulse control unit;
The first and second linear image sensors each include pixels arranged in a two-dimensional array in a vertical direction and a horizontal direction and performing photoelectric conversion on the incident light, a vertical transfer section which performs time-delay integration on charges generated by the pixels and vertical transfers the charges, a horizontal transfer section which horizontally transfers the time-delay integrated charges, and an amplifier which reads out the charges from the horizontal transfer section;
the drive pulse control unit intermittently controls a reset clock pulse in a read operation of the amplifier of the second linear image sensor with respect to a transfer clock pulse of the horizontal transfer unit;
a pixel pitch in the horizontal direction of the first linear image sensor is an integer multiple of that of the second linear image sensor, and
the pixel pitch of the first linear image sensor in the horizontal direction is twice that of the second linear image sensor;
two horizontally adjacent pixels of the second linear image sensor overlap one pixel of the first linear image sensor;
An imaging device characterized by selectively reading out the output signal of the second linear image sensor by shifting it by 1/2 pitch relative to the output signal of the first linear image sensor, or reading out the output signals of the first and second linear image sensors in synchronization. 4. The imaging device according to claim 1, wherein a drive frequency of the horizontal transfer section of the second linear image sensor is an integer multiple of that of the first linear image sensor. the first and second linear image sensors are electrically connected via a through electrode that passes through the first or second linear image sensor;
5. The imaging device according to claim 1 , wherein the first and second linear image sensors are driven in synchronization with each other by a same driving pulse signal via the through electrode. 6. The imaging device according to claim 1, wherein a substrate thickness of the first linear image sensor is thinner than a substrate thickness of the second linear image sensor. 7. The imaging device according to claim 1, further comprising an A/D converter for converting output signals of the first and second linear image sensors into digital data. 8. The imaging apparatus according to claim 7 , further comprising an imaging data storage memory for storing the digital data. 9. The imaging device according to claim 1, wherein at least one of the first and second linear image sensors is a back-illuminated sensor. a pixel pitch in a vertical direction of the first linear image sensor is twice that of the second linear image sensor;
10. The imaging device according to claim 1, wherein a driving frequency of the vertical transfer section of the second linear image sensor is twice that of the first linear image sensor. 11. The imaging device according to claim 1, wherein the second linear image sensor has two of the horizontal transfer units arranged in parallel.

Images