JPS60204082A - Still picture processing device - Google Patents

Abstract

PURPOSE:To obtain a picture signal of correct exposure for the whole of a picture by selecting plural still picture signals, which are picked up while varying the exposure, for individual picture elements and synthesizing them. CONSTITUTION:When two pictures different in exposure where the first and the second aperture values are so determined that they are most suitable for the dark part and the bright part of an object are taken, picture signals of individual picture elements are read out synchronously from frame memories 48 and 50. The output signal of the frame memory 48 is compared with a picture element signal level Vth by a comparator 60 with respect to picture elements having an average brightness; and if this output signal is lower than the level Vth, a gate 52 is conducted to input the output signal of the frame memory 48 to a level shifting circuit 56. If the output signal is lower than the level Vth, a gate 54 is conducted to input the output signal of the frame memory 50 to a level shifting circuit 58.

Description

DETAILED DESCRIPTION OF THE INVENTION 3.1 Technical Field The present invention relates to a still image processing device that synthesizes still images taken using solid-state imaging devices.

3.2 Prior Art In recent years, electronic cameras have been developed that use solid-state imaging devices such as COD instead of silver halide films to take still images. Electronic cameras have various advantages over conventional cameras, such as an imperfect development process, but have the disadvantage of a narrow dynamic range regarding brightness. on the other hand,
BACKGROUND ART In the medical field, particularly in endoscopic diagnosis, the inside of a body cavity is photographed and imaged using an electronic camera for comparative observation. Cancer is particularly likely to develop in both corners, and they are often photographed. Here, since the inside of the body cavity such as the corners of the body has severe irregularities, the concave portions are photographed too dark and the convex portions are photographed too bright, making it impossible to diagnose these concave and convex portions. This problem cannot be solved even if the exposure is adjusted to match the concave or convex portions and the exposure is corrected to take an image, because the convex or concave portions become too bright or too dark. In addition, there is a lot of water in the body cavity, and the light from the light source may be reflected on the water surface. In this case, halation occurs on the screen and diagnosis cannot be performed. When imaging fails in this way, it is necessary to insert the endoscope into the patient again and take photographs in order to make an accurate diagnosis, but this increases patient pain and the doctor's time. There is a drawback.

3.3 Objective The objective of the present invention is to provide a still image processing device that can obtain an image signal with appropriate brightness over the entire surface even for a subject with a large difference in brightness and darkness.

3.4 Overview According to the still image processing device of the present invention, two images can be processed by changing the exposure.
Two images are taken, and one composite image is formed using one of the two images for each pixel.

3.5 Embodiment Hereinafter, an embodiment of the still image processing device according to the present invention will be described with reference to the drawings. Here, one embodiment of a still image processing device used in an endoscopic photography system will be described. FIG. 1 is a schematic diagram thereof. The light from the lamp 12 in the light source unit 10 passes through the condensing mirror 14 and the condensing lens I6.
The light enters the light guide 20 in the endoscopic trowel Z8 through the light beam. An image inside the body cavity is formed by an objective lens 22, transmitted through an image guide 24, and incident on a camera head 26 connected to an eyepiece of an endoscope 18-. The optical image incident on the camera head 26 is incident on a solid-state image sensor 32 made of a CCD via a photographic lens 28 and an aperture 30. This solid-state image sensor 32 is, for example, 256×256
has pixels. The aperture value of aperture 30 is actuator 3.
4.

Charge storage and readout operations of the solid-state image sensor 32 are controlled by clock pulses output from a driver 36.

The output image signal of the solid-state image sensor 32 is sent to the camera control unit 3 as an output of the camera head 26 via a Noriamp.
8 is input. A magnetic disk memory 40 for image recording is connected to the camera control units 3 and 8. In this system, the shooting start button 42 is attached to the camera control unit 38 rather than to the camera controller 26.

A photographing start signal from the photographing start button 42 is input to a control signal generator 44 that generates various control signals. Control signals from control signal generator 44 are provided to rung I2, actuator 34, and driver 36. The camera head output fed into camera control unit 38 is written via selector 46 to either frame memory 48 or 50. The selector 46 supplies the camera head output to the frame memory 48 or 5o for each frame according to the control signal from the control signal generator 44. Frame memory 48.
The image signals read from 50 are respectively dart 52.5.
4 to level shift circuits 56 and 58.

The output signal of the frame memory 48 is also supplied to a comparator 60 and can be compared with a predetermined level Vth. The output signal of the comparator 60 and its inverted signal via the inverter 62 turn the gates 54 and 52 on and off, respectively. Level shift circuit 56.5
As will be described later, the No. 8 shift blocks are set in advance to different predetermined amounts. Level shift circuit 56.58
Both output terminals are commonly connected to each other, and this common connection point is connected to the input terminal of the magnetic disk memory 40. In addition,
Although not shown, writing to the frame memory 48 and 50 and bladder removal operations are also controlled by control signals from the control signal generator 44.

Next, the operation of this embodiment will be explained. Here, it is assumed that the subject has extremely uneven surfaces such as the corners, and the screen has a large difference in brightness and darkness within the screen.

If you shoot this subject with an average metered exposure,
Both the concave portion and the convex portion result in an inappropriate scene. However,
If you shoot the entire screen with an exposure that is determined by partially metering only the concave areas (darkest areas) or convex areas (brightest areas), correct gradation (incident light intensity and image signal An image signal with a proportional value is obtained. Then, by selecting an image signal with correct gradation for each pixel and mixing the two image signals, a composite image signal that accurately expresses the difference in brightness and darkness across the entire screen can be obtained.

The specific operation will be explained below. Figure 2 (,) ~ (,)
takes two still images with different exposures and saves them as frame memo I.
It is a time chart when recording on J48 and J50, respectively. The control signal generator 44 is connected to the lamp Z2, the actuator 34, and the driver 36 as shown in FIG. 2(b), respectively.
(e) Supply the control signal shown in e (d). Before starting photography, the lamp 12 emits light with a small amount of light. Figure 2 (
When an instruction to start photographing is given as shown in a), the level of the control signal to the lamp 12 is increased as shown in FIG. 2(b), and the amount of light emitted from the lamp 12 increases accordingly. In synchronization with the shooting start operation, the level of the control signal to the actuator 34 becomes the second level B as shown in FIG. At the same time, the control signal generator 44 switches the selector 46 to the frame memory 48 side, as shown in FIG. 2(e). When the control signal generator 44 starts photographing, the control signal generator 44 performs
), a control signal is supplied to the driver 36 to set the solid-state image sensor 32 first in the storage mode and then in the readout mode. With this, the first aperture υ value is 1
A still image is taken, and the image signal read from the solid-state image sensor 32 is recorded in the frame memory 48. When the reading of the first image signal is completed, the control signal generator 4
4 supplies the second level B signal to the actuator 34 as shown in FIG.
In addition to setting it to a second value different from the value of
As shown in e), switch the selector 46 to the frame memory 50 side. In this state, the control signal generator 44 is activated as shown in FIG.
d) As shown in Fc, a control signal is supplied to the driver 36 to first set the solid-state image sensor 32 to the storage mode and then to the readout mode. As a result, even at the second aperture value, one still i! The +i image is photographed, and the image signal is written into the frame memory 50 via the selector 46. Here, the first. The second aperture values are determined to be optimal for dark and bright areas of the subject, respectively. Therefore, an image signal is recorded in the frame memory 48 in which the gradation is correct for pixels of average brightness from dark areas, but the gradation is not displayed correctly for pixels in bright areas, and the frame memory 50 is Conversely, an image signal is recorded in which the gradation is correct for pixels of bright to average brightness, but the gradation is not expressed correctly for pixels in the dark area.

In this way, when the continuous shooting of two images with different exposures is completed, the image signals for each pixel are synchronously read out from the frame memory 48.50, and only one of the pixel signals is read out from the dart 52.54. is supplied to level shift circuit 56 or 58. The output signal of the frame memory 48 is compared with a pixel signal level Vth for a pixel of average brightness by a converter 60. frame memory 4
When the output signal of gate 8 is below this level Vth, gate 5
2 becomes conductive, and the output signal of the frame memory 48 is input to the level shift circuit 56. Conversely, the frame memory 48
When the output signal of is higher than this level Vth, dart 5
4 becomes conductive, and the output signal of the frame memory 50 is input to the level shift circuit 58. Accordingly, the output signal of the frame memory 48 is valid for pixels from the dark area to the average brightness area, and the output signal of the frame memory 50 is valid for the pixels of the bright area. Here, since each image signal is obtained by being photographed with different exposures, specifically overexposure and underfog, the level shift circuit 56.5
8, the level is uniformly increased and decreased, respectively. In this way, for each pixel, the correct output of the frame memory 48.50 is selected, level-shifted,
By writing the image signal on the magnetic disk memo+) 40 as one image signal, an image signal having correct gradation for the entire screen can be obtained. According to this embodiment, the dynamic range of the solid-state image sensor 32 can be substantially expanded by taking two images and combining them into one image. In addition, since two images are automatically taken in succession with a single shooting instruction operation, operability is good.

Next, another embodiment of the invention will be described. The embodiment described below is a modification regarding the shooting of two images with different exposures. 3(11) to (,) are time charts showing the photographing operation of the second embodiment, and correspond to FIG. 2(,) to (e), respectively. When an instruction to start shooting is given (Fig. 3(a)), the control No. 16 for the lamp I2 is set to the second level D, and the light emission amount of the rung 12 is set to the first high light amount as shown in Fig. 3(b). shall be. In the second embodiment, as shown in FIG. 3(C), since the control signal to the actuator 34 is constant, the aperture value of the aperture 30 is constant. When the lamp 12 emits the first amount of light, the control signal generator 44 outputs the signal as shown in FIG. 3(d).
The first still image is captured by controlling the driver 36 as shown in FIG. When the first photograph is taken, the control signal generator 44 sets the control signal to the lamp 12 to the second level D, as shown in FIG.
A second large amount of light is set, which is different from the large amount of light. When the lamp 12 emits the second amount of light, the control signal generator 44 outputs the signal as shown in FIG.
), the driver 36 is controlled to take a second still image.

In this way, in the second embodiment, the aperture value υ of the aperture lens 30 is constant, but by changing the amount of light emitted from the lamp 12 in the light source unit 10, two still images can be produced by changing the exposure. Shoot continuously.

Next, the operation of the third embodiment will be explained with reference to the time charts shown in FIGS. 4(8) to 4(d). In the third embodiment, the lamp 12,
The control signal supplied to actuator 34 is set at a predetermined level. Unlike the above embodiment, this level remains constant during the shooting of the two still images. The control signal generator 44 causes the solid-state image sensor 32 to accumulate the first image via the driver 36, as shown in FIG. 4(d).

Readout, storage of the second image, and readout mode are sequentially set. The readout period is unique to the solid-state image sensor 32, and 1
The same is true for the second and second images. However, the accumulation period is the same as the exposure in general photography, and is variable depending on the brightness of the subject. In this embodiment, the accumulation periods for the first and second shots are different.

In this way, according to the third embodiment, exposure is changed by changing the storage period of the solid-state image sensor. This produces the same effect as the embodiment described above.

In the second embodiment, the amount of light emitted from the lamp Q12 is changed by the control signal, but it is also possible to provide separate lamps with different amounts of light and make the light incident on the light guide by switching with a mirror or the like. Alternatively, a discharge tube may be used as the light source lamp, and the discharge light may be caused by capacitors having different capacities. The configuration of the light source unit IO in this case is shown in FIG.

Capacitors 62 and 64 with different capacities are installed at one end of the discharge tube 60.
One end of the capacitor 62 and the other end of the capacitor 64 are connected to the other end of the discharge tube 60 via the changeover switch 56. A trigger signal to the discharge tube 60 and a switching control signal to the switch 66 are generated by a control signal generator (fifth
(not shown in the figure). Since the amount of light emitted from the discharge tube 60 depends on the discharge current, capacitors with different capacities PL?
By switching the shutters 62 and 64 to light the discharge tube 60 and taking pictures twice, it is possible to take two still images with different exposures, as in the above embodiment.

This invention is not limited to the embodiments described above and can be modified in various ways. For example, the number of images to be combined is not limited to two, but if the difference in brightness of the subject is large compared to the dynamic range of the solid-state image sensor, it may be necessary to combine three or more images. Alternatively, the output signal of the frame memory 50 may be input to the comparator 60 and compared with the reference level Vth. Furthermore, the camera head is provided at the tip of the inner hand, making it easy to use.

3.6 Effects of the Invention The present invention provides still image processing in which an image signal with correct exposure can be obtained for the entire screen by selecting and compositing a plurality of still image signals shot with different exposures for each pixel. Equipment is provided.

[Brief explanation of the drawing]

FIG. 1 is a schematic diagram of an embodiment of a still image processing device according to the present invention, FIGS. 2(,) to (,) are time charts showing its operation, and FIGS. figure(,)~(,
) are m2. A time chart showing the operation of the m3 embodiment, FIG. 5 is a diagram showing the first four components of a modification of the present invention. 12...Lamp, 20...Light guide, 22...
- Image guide, 30... Aperture, 32... Solid-state image sensor, 40... Magnetic disk memory, 44... Control signal generator, 4, 8. 50... Frame memory, 5
2.54...Dart, 56.58...Level shift circuit. Applicant's agent Patent attorney Junseki Tsuboi? Document of amendment to the proceedings Kazuo Wakasugi, Commissioner of the Patent Office 1, Indication of the case Patent Application No. 1982-60219 2, Name of the invention Still image processing device 3, Person making the amendment Relationship to the case Patent applicant (037) Olympus Kogaku Kogyo Co., Ltd. 4, Agent address: 17 Mori Building 7, 1-26-5 Toranomon, Minato-ku, Tokyo, "Memory 40 is connected," as stated on page 4, line 19 of the statement of contents of the amendment. ” to “memory 40 and monitor (not shown)
By the way, in general, an image recording medium such as a magnetic disk has a dynamic range (wider 1/X, corrected as J) compared to a solid-state image sensor.

Claims (1)

[Claims] The first photo taken with different exposures. means for recording a second image signal; and a first image signal read from the recording means. means for comparing one of the second image signals with a predetermined level for each pixel; The second image signals are each corrected according to the exposure at the time of shooting. means for converting the first correction signal into a second correction signal; A still image processing device comprising means for forming a composite image using either one of the second correction signals.