JPH0530424A - Picture input device - Google Patents

Abstract

PURPOSE:To realize the picture input device with a large dynamic range. CONSTITUTION:A CCD area sensor 1 uses an electronic shutter function to vary a shutter time for image pickup, the result is amplified at an amplifier 2 and an A/D converter 3 converts the signal into a digital signal and stored in frame memories 51,52 respectively via a distributer 4. Then picture signals in the frame memories 51,52 are synthesized by a signal synthesizer 6 to synthesize one picture signal.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image input device using an area sensor, and more particularly to a technique for inputting a high quality image with a wide dynamic range.

[0002]

2. Description of the Related Art An area sensor is a device for converting a subject image formed by a lens or the like into a time-series electric signal by means of a two-dimensional light receiving element array equipped with an appropriate scanning means. In an image input device using such an area sensor, one of the important conditions for inputting a high quality image is
The first is to increase the range between the minimum level and the maximum level of the illuminance of the subject that can be extracted as a signal, that is, to increase the dynamic range. FIG. 6 is an explanatory diagram of the dynamic range. The characteristic shown in FIG. 6 is the relationship between the illuminance of the subject (or the illuminance of the sensor surface) and the output signal level of the area sensor. The output signal of the area sensor, regardless of type, contains not only signal components but also noise components such as noise generated by the switching element used in the scanning unit and the first stage amplifier, or dark current noise generated in the light receiving element. ing. If the signal component becomes smaller than these noise components as the subject illuminance decreases, the signal cannot be extracted. This is a factor that determines the lowest level of illuminance of the subject that can be input (indicated by L _MIN in FIG. 6). On the other hand, as the illuminance of the subject increases, the signal also increases in proportion, but from a certain point, the signal does not increase even if the illuminance of the subject increases. That is, the output signal exhibits a saturation characteristic. Illuminance at which this saturation begins (L in Fig. 6
₍ Indicated by _MAX ) is the highest level that can be extracted as a signal. Therefore, the dynamic range of the illuminance of the subject is represented by the difference between L _MAX and L _MIN .

A normal area sensor has a dynamic range of about three digits. When the illuminance difference between the dark and bright areas of the subject exceeds the dynamic range, it is impossible to reproduce the gradation of the image in the entire illuminance range. For example, when capturing a shadow of a person in a place with a bright background, if the lens aperture is adjusted to the brightness of the background, the face of the person becomes black and the shape of the face cannot be recognized well. Occurs. Also, the background becomes pure white when adjusted to the brightness of the shadowed person.

To solve the above problem, it is necessary to expand the dynamic range. As described above, the dynamic range is determined by the noise level and the saturation characteristic. Of these, the saturation characteristics are mostly due to the structure of the light receiving element, and it is not easy to increase the saturation value. Therefore, in the conventional method, the noise component is reduced by a dedicated circuit provided after the first-stage amplifier, and thereby the minimum level (L _MIN ) is lowered to expand the dynamic range.

[0005]

SUMMARY OF THE INVENTION In the above conventional method,
Although the dynamic range is expanded by reducing the noise in the circuit, the noise reduction that can be realized by the circuit means is at most about one digit. Therefore, the expansion of the dynamic range is limited to about one digit, and there is a problem that the dynamic range is still insufficient for inputting a high quality image.

An object of the present invention is to solve the above problems and provide an image input device having a large dynamic range.

[0007]

An image input device according to the present invention is a device that uses one of a plurality of image signals picked up with different accumulation periods of photogenerated charges in photoelectric conversion in an area sensor.
The image signals of one sheet are combined.

An area sensor with an electronic shutter function is used as a method of varying the accumulation period of the photo-generated charges. In a normal area sensor, photo-generated electric charges are accumulated in a pixel for one frame period and are taken out as a signal electric charge at the end of the frame. Only the accumulated charges can be taken out as signal charges, and the accumulation period (hereinafter, referred to as shutter time) can be set freely.

Setting the shutter time to different times for inputting corresponds to taking an image with different accumulation periods of photogenerated charges. Therefore, in the present invention, by utilizing the electronic shutter function, one frame (image) is composed from a plurality of images input at different photo-generated charge accumulation times.
Hereinafter, for simplicity, a method of synthesizing images having a wider dynamic range than the two input images with different shutter times will be described. Here, in the input of two images, the shutter time is set to be sufficiently shorter than the input of the second image in the input of the first image.

[0010]

FIG. 5 is a diagram showing the principle by which the dynamic range of an input image can be expanded according to the present invention. FIG. 5A shows the relationship between the object illuminance and the output signal level at two inputs with different shutter times. In FIG. 5A, C1 is the characteristic when the shutter time is long, and C2 is the characteristic when the shutter is short. Also, L1 _MAX , L
1 _MIN and L2 _MAX , L2 _MIN are the maximum subject illuminance and the minimum subject illuminance that can be input when the shutter time is long and short, respectively. Here, L1
_Suppose that the shutter time for each input is set so that _MAX and L2 _MIN are the same.

As described above, the shutter time is proportional to the accumulation time of the photo-generated charges in each pixel. That is, if the shutter time is long, the exposure time becomes long, and it is possible to input with high sensitivity, and as shown in FIG. 5A, it is possible to input in the low illuminance region of the subject. However, in the high illuminance region of L1 _MAX or more, the signal charge is saturated, and the input image is uniformly white. On the other hand, if the shutter time is short, the exposure time is short, and the sensitivity is low.
As shown in, the shutter is shifted to the right side (higher illuminance side) than when the shutter time is long. As a result, the gradation cannot be reproduced in the dark part of the subject and the area is painted black, but saturation is not generated in the high illuminance region, and gradation expression with good reproducibility is possible.

Therefore, in the present invention, the signal at the first input with a long shutter time is used in the low illuminance region, while the signal at the second input with a short shutter time is used in the high illuminance region. FIG. 5B shows a principle diagram in which one image signal having a wide dynamic range can be combined from two signals input with different shutter times as described above. Here, the second input signal is amplified by the ratio of the shutter times at the two inputs. Second by this amplification
The characteristic C2 at the input of is shifted to the point C2, the two characteristics can be continuously connected at the point P shown in FIG. 5B, and can be regarded as one characteristic. Then, the dynamic range of the synthesized characteristic becomes equivalent to L2 _MAX -L1 _MIN. As a result, for example, when the dynamic range in the conventional method is set to 3 digits, the dynamic range in the present invention becomes 6 digits, which is a dramatic improvement. Therefore, good gradation reproduction is possible in all the brightness regions from the bright part to the dark part of the subject, and a high quality input device can be realized as compared with the conventional method.

[0013]

FIG. 1 shows a first embodiment of the present invention. In FIG. 1, 1 is a CCD area sensor, 2 is an amplifier, and 3 is A.
A / D converter, 4 is a distributor, 51 and 52 are frame memories, 6 is a signal synthesizer, 7 is a control signal generator, and 8 is a lens. In this embodiment, in order to make the minimum subject illuminance of the composite image as small as possible, the first input is made without using the electronic shutter function. Therefore, the accumulation time of the signal charges is equal to the frame period. On the other hand, in the second input, the shutter time is set to 1/1000 of the frame period and the input is performed. In this embodiment, in order to reproduce the gradation of a wide range of illuminance of about 6 digits, after the analog output signals of the CCD area sensor 1 at the first and second inputs are both amplified,
LOG conversion, A / D conversion of this with 8 bits, and accumulation in frame memories 51 and 52, respectively.

With the above method, the two frame memories 51,
The two image signals input to 52 are combined into one image signal by the following method. First, the magnitudes of the image signals stored in the frame memory 51 are checked in the order of addresses. If the magnitude of the signal is less than the maximum gradation of 255, the frame memory 51
The signal from is used as the output signal as it is. If the signal from the frame memory 51 has a maximum gradation of 255, the signal of the same address in the frame memory 52 will be 255.
Is added to the output signal. Here, adding 255 is a process corresponding to the amplification of the second input described in the above principle. Although this output signal is LOG-converted, appropriate γ-conversion may be performed according to the type of image output device such as display or recording or the type of processing such as color conversion.

ON / OFF of the electronic shutter function for each frame of the area sensor is performed by a control signal from the control signal generator 7, and the image signal at the first input and the second input is further used by using this control signal. Are distributed to the two frame memories 51 and 52 in the distributor 4. With the above-described configuration and signal processing, in the present embodiment, it is possible to input in a dynamic range that is twice the number of digits as compared with the image input device according to the conventional method.

A second embodiment of the present invention is shown in FIG. Figure 2
In, 21 and 22 are amplifiers, and 31 and 32 are A / D converters. Others are the same as those in FIG. The CCD area sensor 1 used in this embodiment is shown in FIG. In FIG. 3, 100 is a photosensitive section, 110 and 120 are signal storage sections, 130 and 140 are horizontal transfer paths, and 150 and 16
0 is an output terminal. In this embodiment, an FT type CCD area sensor is used.

A normal FT type CCD area sensor has a structure as shown in FIG. 7, and the electronic shutter function is realized by the following method. That is, after the frame (or field) starts, the charge accumulated in the photosensitive section 100 is discharged to an outlet on the opposite side of the accumulating section for T ₀ seconds, and then the photosensitive section 100 reaches the end of the frame (or field) again. The signal charge is stored in and is sent to the signal storage unit 110 as the signal charge. Therefore, assuming that the frame period is T, the effective signal storage time (shutter time) is (T-T ₀ ), and the shutter time is controlled by changing T ₀ .

On the other hand, in this embodiment, the FT type CCD area sensor having the signal accumulating sections 110 and 120 on both sides of the photosensitive section 100 is used. Then, after the frame starts, T ₀
The signal charges accumulated in the photosensitive unit 100 during the period up to are accumulated in the signal accumulation unit 110, and the signal charges accumulated in the remaining (T−T ₀ ) are accumulated in the signal accumulation unit 120.

The signal charges provided in the two signal storage sections 110 and 120 of the FT-type CCD area sensor are read out in parallel from the two output terminals 150 and 160 as shown in FIG. 2, and further amplified by amplifiers 21 and 22, respectively. L, A /
A / D conversion is performed by the D converters 31 and 32, and the frame memory 5
1 and 52 respectively. The images stored in the two frame memories 51 and 52 have a shutter time ratio (T-
Two images input by T ₀ ) / T _0. For example, if the ratio is increased to 1: 1000, it is possible to synthesize an image with a wide dynamic range by the same method as in the first embodiment.

In this embodiment, two inputs having different shutter times can be input within one frame, and an image having a wide dynamic range can be input at the same speed as the conventional method. Moreover, since the two image signals are read in parallel and processed, the present invention can be performed without increasing the drive frequency.

A third embodiment of the present invention is shown in FIG. Figure 4
In the figure, 11 and 12 are CCD area sensors, and 9 is a half mirror having the same transmittance and reflectance. In this embodiment, two identical CCD area sensors 11 and 12 are used. Although the two CCD area sensors 11 and 12 both have an electronic shutter function, the shutter time of the CCD area sensor 12 is 1/100 of that of the CCD area sensor 11.
Set to 0. In order to input the same subject image to the two CCD area sensors 11 and 12, the optical path after passing through the lens 8 is divided into two by the half mirror 9, and the respective C
The same subject image is formed on the CD area sensors 11 and 12. The analog signals from the CCD area sensors 11 and 12 are amplified and further converted into LOG-converted 8-bit digital signals, which are then sent to the frame memories 51 and 52, respectively.

The two frame memories 51,
The method of synthesizing the two image signals input to 52 into one image signal is the same as in the first embodiment. In this embodiment, 2
By using one CCD area sensor 11 and 12, it is possible to input an image with a wide dynamic range at the same speed as the conventional method.

Of the three embodiments described above, the second and third embodiments are suitable for capturing moving images because high-speed input is possible. However, it is of course also applicable for inputting a still image. Further, although the first embodiment is suitable for a still image such as an electronic still camera, it can be used for the same image by increasing the driving frequency of the CCD area sensor 1.

Needless to say, various modifications can be made without departing from the spirit of the present invention. For example, although the CCD type is used as the area sensor in the above embodiment, it is clear from the principle of the present invention that the present invention can be implemented by using a MOS type area sensor with an electronic shutter function. Further, in the first embodiment, one image is synthesized from two image signals input with different shutter times, but an image with a wider dynamic range can be obtained by synthesizing from three or more image signals with different shutter times. Can be input. According to the same principle, in the third embodiment, the dynamic range becomes wider by increasing the number of area sensors used. In the above embodiment, the shutter time ratio for inputting two images was selected to be 1: 1000, but this ratio can be arbitrarily selected as long as it is equal to or less than the dynamic range of one area sensor used. Further, in the above embodiment, the LOG conversion is performed before the A / D conversion, but this process can be omitted if necessary.

Further, in the above-mentioned embodiment, two image signals having different storage times are generated in a time division manner.
Although the frame memories 51 and 52 are used in both cases, the present invention can be carried out by omitting the frame memories 51 and 52 by using appropriate signal delay circuits and comparators. In this case, in the third embodiment, since the two image signals are generated in parallel, the signal delay circuit is also unnecessary.

[0026]

As described above, according to the present invention, in the image input device using the area sensor, the area sensor in which the accumulation period of the photo-generated charge generated in photoelectric conversion is variable, and the photo-generated charge generated by the area sensor. Since it has a signal synthesizer that synthesizes one image signal from a plurality of image signals captured with different accumulations, the dynamic range can be expanded by the number of inputs or a multiple of the number of area sensors used. The gradation can be reproduced over a wide range of illuminance, and a high quality image input device can be realized. Further, since the accumulation time of the photo-generated charges generated in the photoelectric conversion can be made different by using the electronic shutter function, it is possible to freely change the dynamic range by controlling the shutter time. In addition, the electronic shutter function allows images with different shutter times to be captured continuously at high speed, and thus can be used for moving images.

[Brief description of drawings]

FIG. 1 is a block diagram showing a schematic configuration of a first embodiment of the present invention.

FIG. 2 is a block diagram showing a schematic configuration of a second embodiment of the present invention.

FIG. 3 is a schematic configuration diagram of a CCD area sensor used in a second embodiment.

FIG. 4 is a block diagram showing a schematic configuration of a third embodiment of the present invention.

FIG. 5 is a diagram for explaining the principle of the present invention.

FIG. 6 is an explanatory diagram of a dynamic range of a conventional image input device.

FIG. 7 is a schematic configuration diagram of a normal FT type CCD area sensor.

[Explanation of symbols]

 1 CCD area sensor 2 amplifier 3 A / D converter 4 distributor 6 signal combiner 7 control signal generator 8 lens 9 half mirror 11 CCD area sensor 12 CCD area sensor 21 amplifier 22 amplifier 31 A / D converter 32 A / D converter 51 frame memory 52 frame memory 100 CCD area sensor photosensitive area 110 CCD area sensor signal storage area 120 CCD area sensor signal storage area 130 CCD area sensor horizontal transfer path 140 CCD area sensor horizontal transfer path 150 CCD Area sensor output terminal 160 CCD area sensor output terminal

Claims (1)

Claim: What is claimed is: 1. In an image input device using an area sensor, if the accumulation period of the photo-generated charges generated by photoelectric conversion is variable and that of the area sensor is different. 1 from multiple image signals captured by
An image input device comprising: a signal synthesizer for synthesizing one image signal.