JPH0446503B2 - - Google Patents

Description

[Detailed Description of the Invention] [Summary] An imaging device using a solid-state imaging device having a plurality of pixels and a means for transferring charges generated therein, in which a memory for storing correction data is configured to correspond to an accumulation time. Prepare the number of pieces, and when inputting correction data, switch the accumulation time for each frame, input the correction data corresponding to each accumulation time and store it in memory, and when changing the accumulation time, change the correction data and process. To enable imaging without deterioration in image quality.

[Industrial application field]

The present invention relates to an imaging device using a two-dimensional solid-state imaging device, and in particular, when the incident light changes from weak to large, it automatically adjusts the sensitivity and increases the sensitivity when the incident light is small. The present invention relates to an imaging device that can suppress overflow by reducing sensitivity when the incident light is large.

[Conventional technology]

FIG. 3 shows a schematic cross-section of an input section of a general charge transfer solid-state image sensor. In the figure, 3
1 is a photo diode, 30 is a semiconductor substrate, 3
2 is an input section S, 33 is an input gate I _G , 34 is a transfer gate T _G , and 34 is a transfer electrode of the CCD.

The carrier from the photo diode 31 (light receiving sensor) due to the incident light passes under the input gate I _G 33 and the transfer gate T _G 34, and is transferred to the CCD section 35.

The dynamic range of the device shown in FIG. 3 is determined by the amount of charge obtained and the size of the potential well of the CCD, that is, the brightness of the optical system, the transmittance, the size of the potential well, etc.

However, when the incident light changes from weak to large, it is limited by the charge storage capacity of the charge transfer device (CCD), making it impossible to obtain sufficient sensitivity and dynamic range. For example, if the optical system is made bright so that a large signal value can be obtained even under low illuminance, charges will overflow from the wells even if the incident light increases slightly, causing blooming. On the other hand, if the optical system is made dark so that it does not become saturated even with strong light, the detection sensitivity of the system will be reduced.

In order to overcome these drawbacks, the inventor of the present invention previously invented a method for changing the storage time for each frame in Japanese Patent Application No. 111214/1982 and Japanese Patent Application No. 279912/1982. For example, as shown in Figure 4,
The charge accumulation time Ts is switched to at least two types Ts ₁ , Ts ₂ , Ts ₁ < Ts ₂ for each operating frame _TF , and images are captured with different accumulation times alternately, and the long accumulation time Ts ₂
After removing the signals that overflow due to strong incident light among the output signals in the frame, the signals in the frame with the short accumulation time Ts ₁ are added and output. Among the image signals obtained in frames with a long accumulation time,
Overflowing signals are removed, and only the image signals obtained in the next short accumulation time frame are added and output, making it possible to improve the dynamic range without missing strong incident light points. .

However, although the above method of changing the accumulation time for each frame has the advantage of allowing the saturation signal and sensitivity to be set arbitrarily and widening the dynamic range, it encounters one difficulty when actually applying this method to an imaging device. do. This is because, in practice, when there are variations in offset and sensitivity in the output of each pixel of a solid-state image sensor, when the accumulation time is changed, this offset value may change and the image may deteriorate.

This will be explained in more detail with reference to FIGS. 5 and 6. FIG. 5 is a diagram showing the planar configuration of the element shown in FIG. 3, and the symbols for each part are the same. Similarly to the above, accumulation times Ts ₁ and Ts ₂ (Ts ₁ <
Ts ₂ ) The stored charge is transferred through I _G (input gate) and T _G (transfer gate) as shown by the broken line.
Sent to the CCD transfer section. When a pulse voltage is applied to the transfer electrode φ _o of the CCD, a depletion layer is created on the surface of the silicon substrate, forming a potential well. When signal charges are introduced by light irradiation, the signal charges are stored in this potential well. When a pulse voltage is applied to the adjacent transfer electrode φ _o+1 to form a potential well and at the same time the voltage of the first transfer electrode φ _o is removed, the signal charge under the first transfer electrode φ _o is transferred to the adjacent potential well. be done.

The signal charge generated by light irradiation is determined by the amount of incident light, the brightness (F number) of the optical system, the transmittance, the quantum efficiency in the silicon substrate, and the storage time.

On the other hand, the size of the charge storage capacity of a charge transfer device (CCD) is determined by the SiO ₂ film thickness, electrode area, and substrate carrier concentration, and has an upper limit. Therefore, when strong light is irradiated, the potential well exceeds its capacity and overflows. As a result, a so-called blooming phenomenon occurs, which appears as a white line in the transfer direction, as shown in FIG.

In order to prevent this overflow and obtain a good picture, if the accumulation time is changed for each frame as described above, the offset value will change as described above and the image will deteriorate.

Figure 7A shows the signal level of the CCD output in the case of one frame scanning time and the normal amount of incident light (accumulation time Ts ₂ ), and Figure B shows the signal level of the CCD output when the amount of incident light is large and overflow occurs. It shows the relationship between CCD output signal levels when switching the accumulation time. Within the scanning time of one frame (in this case, each pixel of the two-dimensional solid-state image pickup device), there are offsets shown by diagonal lines and variations in sensitivity shown by white bars. On the other hand, in FIG. 7B, if the accumulation time is shortened to prevent overflow,
As shown in the figure, while the signal level decreases, the offset also changes, and the variation in offset for each pixel also changes.

In order to detect the true signal level, a correction value is obtained by subtracting the offset and dividing it by the sensitivity (gain). However, if the offset changes depending on the accumulation time as shown in Figure 7A and B, proper correction processing may not be possible. It becomes difficult and the image deteriorates.

[Problem that the invention seeks to solve]

The present invention provides an image pickup device that has a function of changing the accumulation time for each frame according to the amount of incident light and prevents image deterioration such as blooming due to overflow, by adjusting the offset variation appropriately by changing the accumulation time. This enables accurate correction.

[Means for solving problems]

As the concept of the present invention is shown in FIG. 1, a solid-state image sensing device 1 having a plurality of pixels and a means for transferring electric charge generated in each pixel is used, and the solid-state image sensing device 1 is used to adapt to changes in the amount of incident light during imaging. Therefore, an imaging apparatus including means 21 for switching the accumulation time for each frame includes the following (a) to (d).

(a) A memory 25 prepared for the number of storage times (n) for storing correction data of the output offset and sensitivity for each pixel of the solid-state image sensor 1.

(b) Means 22 for storing correction data for each accumulation time in the memory 25 by sequentially switching the accumulation time when inputting the correction data.

(c) means 2 for reading out the correction data stored in the memory 25 in accordance with switching of the means 21 for switching the accumulation time for each frame during imaging;
4.

(d) Means 23 for correcting the imaging data using the read correction data.

[Effect]

As a result of the above, each memory has a different accumulation time.
Since correction data corresponding to each of Ts ₁ to _{Ts o} is stored, if the amount of incident light changes, the data used for correction can be changed at the same time as the accumulation time according to the amount of incident light, and the appropriate amount can be obtained. The correction process can be performed by the correction means 23, and it becomes possible to take an image without deteriorating the image quality.

〔Example〕

FIG. 2 shows a circuit diagram of an embodiment of the present invention. In the figure, 1 is a two-dimensional solid-state image sensor as previously shown in FIGS. 3 and 5, and its output is sent to an amplifier 2. The signal is then input to the A/D converter 3 and becomes digital data. This digital data includes offsets and variations in sensitivity (gain) of each pixel of the two-dimensional solid-state image sensor.

Therefore, a subtracter 4 removes the offset, and a divider 5 corrects variations in sensitivity to make it uniform.

The offset correction data in this case is stored in the offset memory 6 and offset memory 7, while the sensitivity correction data is stored in the gain memory 8 and gain memory 9. Therefore, multi-
It is selected and read by the plexer (MPX) 10 or 11.

The data whose offset and sensitivity have been corrected are input alternately to the frame memory 12 and the frame memory 13, and are alternately read out to the D/A converter 14 in accordance with the TV display timing and displayed on the CRT 15.

In the figure, a comparator 17 and a counter 18 exist to detect the saturation of the output of the two-dimensional solid-state image sensor 1. For example, when the value of the A/D converter 3 exceeds a certain value, the comparator 17 generates one pulse. By counting these pulses, the counter 18 counts how many pixels exceed a certain value. When the count value exceeds a certain value, the microprocessor 16 determines that an overflow has occurred, and sends a switching signal to the MPXs 10 and 11, and also sends a signal to the timing generator 19 to change the state of the two-dimensional solid-state image sensor 1. Ts ₁ that generates drive pulses and has a shorter accumulation time than before (Ts ₁ < Ts ₂ )
Switch to.

Note that the microprocessor 16 performs other overall control.

Next, the operation of capturing correction data and switching the accumulation time, which are the core of the present invention, will be described.

[Input of correction data]

When inputting correction data, the microprocessor 16 first sets a predetermined accumulation time _Ts1 , and the two-dimensional solid-state image sensor 1 operates during the accumulation time _Ts1 .

Next, low and high brightness reference light sources (not shown)
The bistandard data is stored in the frame memory,
The microprocessor 16 reads the data, calculates offset correction data and sensitivity correction data, and inputs the results to the offset memory and gain memory.

After the correction data for the accumulation time Ts ₁ is obtained in this way, the same process is repeated with the accumulation time Ts ₂ (Ts ₂ > Ts ₁ ), and the obtained correction data is stored in the offset memory 7 and the gain memory 7. Input into memory 9.

[Switching the accumulation time during imaging]

Operation at low illuminance is performed with an accumulation time Ts ₂ (>Ts ₁ ). In this case, an offset memory 7 and a gain memory 9 are used.

The data A/D converted by the A/D converter 3 is input to a comparator 17, which determines whether the output is above a certain value, and when the output is above a certain value, the comparator 17 outputs to the counter 18. counter 1
The output from 8 indicates the number of pixels in which the output of the two-dimensional solid-state image sensor is greater than a certain value, and if this output is greater than a certain value, it can be determined that the output of the two-dimensional solid-state image sensor is saturated. .

Now, when the incident light increases rapidly and the counter output reaches a certain value, the microprocessor 16 switches the accumulation time to Ts ₁ , and at the same time switches the multiplexer 1
0 and 11 to select the offset memory 7 and gain memory 9. Thereby, it is possible to obtain an imaging operation in which the output does not become saturated even in response to excessive input, and the image quality does not deteriorate.

As is clear from the above, according to this embodiment, even when there are variations in offset and sensitivity for each pixel in the output of the solid-state image sensor,
Sensitivity can be adjusted so that the signal value always remains within a certain range in response to large changes in incident light, making it possible to achieve high performance in the imaging device.

〔Effect of the invention〕

According to the present invention, even if the output of the solid-state image sensor has variations in offset and sensitivity for each pixel,
The light-receiving sensitivity can be switched according to changes in the amount of incident light without deteriorating the image quality, and imaging operation with high sensitivity and high dynamic range can be obtained, which is extremely effective in improving the performance of the imaging device.

[Brief explanation of drawings]

FIG. 1 is a conceptual diagram of the present invention, FIG. 2 is a circuit diagram of an embodiment of the present invention, FIG. 3 is a sectional view showing the input section of a solid-state image sensor, FIG. 4 is a waveform diagram of a conventional example, and FIG. 6 is a plan view of the input section of a solid-state image sensor, FIG. 6 is an explanatory diagram of the blooming phenomenon, and FIGS. 7A and 7B are diagrams illustrating the influence of offset variations. Main codes, 1... Two-dimensional solid-state image sensor, 4... Subtractor, 5... Divider, 6... Offset memory,
7...Offset memory, 8...Gain memory, 9...Gain memory.

Claims (1)

[Scope of Claims] 1. A solid-state image sensor 1 having a plurality of pixels and means for transferring charges generated in each pixel is used, and means for switching the accumulation time for each frame according to changes in the amount of incident light during imaging. 21, (a) a memory 25 prepared for the number of storage times and for storing correction data of the output offset and sensitivity for each pixel of the solid-state image sensor 1; and (b) a memory 25 for storing the correction data. (c) means 22 for storing correction data for each accumulation time in the memory 25 by sequentially switching the accumulation time at the time of input; (c) means 21 for switching the accumulation time for each frame at the time of imaging; means 2 for reading correction data stored in the memory 25;
4; and (d) means 23 for correcting imaging data using the read correction data.