JP3296837B2 - Endoscope imaging device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an endoscope image pickup apparatus capable of picking up an image of an object through an endoscope and performing monitor observation.

[0002]

2. Description of the Related Art Television cameras and the like using image pickup devices have recently become widespread. However, as image pickup devices have become smaller and more sensitive, monitor observation using these image pickup devices in the medical field has been difficult. Diagnosis and treatment can be sufficiently performed. In the case of diagnosing by observing the inside of a body cavity using an endoscope, in particular, only one surgeon could observe the affected part in the past, but using a videoscope with a built-in image sensor at the end of the endoscope. By connecting a small television camera to an ordinary fiberscope, a plurality of people can monitor and monitor at the same time. This monitor observation can reduce the fatigue of the observer's eyes as compared with the conventional endoscopic visual observation, and can be viewed by many people other than the surgeon from the viewpoint of education. Has the advantage of being very convenient.

[0003] By the way, a television camera used for endoscope observation is required to be small and lightweight. Because the surgeon must keep the endoscope with one hand during the examination and diagnosis with the television camera attached to the endoscope, and sometimes change the orientation of the endoscope,
This is because it is necessary to operate various treatment tools and the like.
In addition, the image guide fiber bundle of the endoscope is usually regularly arranged, and an image formed by the image guide fiber bundle is further discretely sampled by an image pickup device to obtain a monitor image. Significant signal generation. In order to prevent this moiré, it is necessary to arrange an optical low-pass filter in the middle of the optical path of the television camera. Usually, this optical low-pass filter is often constituted by a plurality of crystal birefringent plates. Other optical low-pass filters such as phase filters may be combined.

In order to prevent moiré, a large number of filters are required, and these filters are usually used.
It is often arranged between a lens system and an image sensor, so that the back focus of the lens system must be sufficiently long. By the way, as the density of an image sensor increases and the sensitivity increases, an optical system used for the image sensor also needs to have extremely excellent performance. However, as described above, since many filters must be arranged on the front side of the image sensor, for example, if an attempt is made to provide a space for adding one filter (about 1 to 3 mm thick) to the front side of the image sensor. Further, it is necessary to lengthen the back focus. In such a case, it is technically difficult to extend only the back focus without deteriorating the performance of the optical system. Moreover, increasing the back focus and enlarging the entire optical system must be avoided because the size and weight of the apparatus are increased.

Further, if many optical low-pass filters are used to prevent moiré, there is a problem that the resolution is reduced and the image is blurred, and more space is required. Conversely, if the optical low-pass filter is not used very much, moiré removal will be insufficient, and in either case, a problem will occur in diagnosis. As described above, in order to reduce the size and weight of the imaging apparatus and improve the optical performance, it is natural that the optical low-pass filter must be optimized and installed.

Next, the configuration of an endoscope imaging system in which an imaging device such as a television camera is attached to the endoscope will be described with reference to FIG. In the drawing, illumination light from a light source 1 is applied to an object (not shown) through a light guide fiber bundle 2 and an illumination lens 3. The light source 1 contains a halogen lamp or a xenon lamp, and usually irradiates an object with white light. The object image is formed by an objective lens 4 arranged in parallel with the illumination lens 3, transmitted through an image guide fiber bundle 5 in the case of a fiberscope, and through a relay lens in the case of a rigid scope. The light is transmitted and enlarged by the eyepiece 7 in the eyepiece 6. Eyepiece 6
Inside, a fixed stop 8 and a cover glass 9 having different aperture diameters for each scope are usually arranged behind the eyepiece 7.

The television camera 1 attached to the eyepiece 6
In 0, the light beam that has passed through the fixed stop 8 passes through the cover glass 11 and the infrared cut filter 12, and then is formed on the image sensor 15 as an endoscope image via the optical low-pass filter 14 by the imaging lens 13. It is imaged. The electric signal obtained by the image sensor 15 is subjected to signal processing by the camera control unit 16 and is displayed on the television monitor 17 as an image.

In many cases, the television camera 10 uses a so-called mosaic-type CCD (solid-state image sensor) in which a color filter is arranged in each light receiving element of the image sensor 15. In addition, the observation light is divided into three primary colors of R, G, and B using a three-color separation prism, and R, G, and B primary color images are respectively captured by three CCDs, and then combined. Three-panel television camera that obtains color images through
A plane-sequential television camera, etc., in which a rotating filter for the three primary colors G and B is provided to sequentially irradiate the illumination light of the three primary colors onto an object, sequentially capture images of different colors by a CCD, and combine them to make a color. There is.

In the endoscope imaging system shown in FIG. 46, the infrared cut filter 12 cuts light in the infrared wavelength range, since the image sensor 15 has sensitivity in the infrared range, and the image obtained is color-coded. It is provided to prevent the reproducibility from becoming poor. This infrared cut filter has an absorption type and an interference type, but the absorption type is weak in resistance because it often contains a lot of phosphoric acid in the glass component,
The surface may start to melt or become cloudy. Also, in the interference type, countless fine white spots may be generated on the surface coated with the interference film. In addition, since the light flux near the image plane becomes extremely thin, if such a filter is provided in the vicinity of the image sensor, a black spot appears on a monitor screen or an image is blurred.

Therefore, in order to improve the resistance of the infrared cut filter, conventionally, it has been practiced to sandwich the glass with strong resistance and seal the side surface with an adhesive or the like. However, this increases the thickness and outer shape of the filter, and also requires a lot of work time for joining. In addition, an infrared cut filter may be used in place of the cover glass of the image pickup device in order to shorten the back focus of the optical system as much as possible.However, since the infrared cut filter comes closer to the image plane, the beam diameter is further increased. It becomes smaller, and the change in the filter surface appears as an effect on the image. Further, when the imaging optical system is a zoom lens, the angle of incidence of light rays on the image sensor may vary greatly depending on the zooming state. In this case, since the incident angle to the infrared cut filter also changes, especially in the case of the interference type filter, the spectral transmittance characteristic changes due to the light incident angle, the color reproducibility changes, and the image is affected. There is a problem.

In such a case, as shown in FIG.
When the infrared cut filter 12 is disposed between the cover glass 11 and the imaging lens 13, at this position, the light beam diameter is substantially determined by the fixed stop 8 of the eyepiece 6, and the light beam is considerably large in the entire system. Since the filter 12 is thick, even if the filter 12 has some dust or scratches, the obtained image has almost no effect. Further, since the object point position viewed from the television camera 10 side is much larger than the stop diameter of the fixed stop 8 and the divergence angle of the light beam is small, the spectral characteristic is low like an interference filter. Even if it has an angle dependency of the incident light beam, unlike the case where it is arranged behind the imaging lens 13, it is not necessary to specifically suppress the incident angle. Further, even if the television camera has a zoom lens, the state of the luminous flux in front of the zoom lens is constant regardless of the zoom state, so that the image is not affected.

Further, in such an arrangement, the image from the endoscope eyepiece 6, that is, the object point viewed from the television camera side can be distant (although there is a difference between positive and negative diopters), For example, when the back focus is extended by one filter, it is easier to secure a space with almost no influence on the whole by disposing the filter in front of the imaging optical system.

FIG. 47 shows a modification of the configuration shown in FIG.
As shown in (1), the infrared cut filter 12 may be provided on the first surface as a cover glass of the television camera 10. However, since the absorption type filter has low resistance, when a YAG cut filter or the like for removing a YAG laser beam is also used, it must be placed inside the filter or a combination of a plurality of filters. In this case, an interference-type filter such as a YAG cut filter used as a cover glass has its YAG cut coated surface turned inward in order to have resistance to disinfectants used at the time of surgery or the like. Good to do.

Also, as shown in FIG. 48, the endoscope eyepiece 6
An adapter 16 is attached between the camera and the television camera 10, and the infrared cut filter 12
May be arranged. Also in this case, if the luminous flux emitted from the adapter 16 has the above-described characteristics, the infrared cut filter 12 can be arranged in front of the lens system on the television camera side. . It is not necessary to bring all the infrared cut filters in front of the lens system, and it may be determined in consideration of the space in which the filters can be arranged, the tolerance, characteristics, and the like.
The same arrangement may be adopted not only for ordinary visible observation requiring an infrared cut filter but also for special observation and other types of filters. Instead of using an interference type filter, a lens whose surface is coated with an interference film may be used. In this case, the lens may be made of a special glass material instead of a normal glass material. Also in this case, it is necessary to arrange the light beam at a place where the incident angle of light is not so large.

Further, the present invention can be similarly applied to a case where the entire optical system is integrally provided as shown in FIG.
In this configuration, the image guide fiber bundle serving as the image transmission system is removable, and the subsequent imaging optical system is integrated, but since the internal components are the same as those described above, In the configuration, the infrared cut filter 12 is provided behind the brightness stop arranged in the middle of the lens system.

Next, an imaging optical system of an endoscope imaging system having a zoom optical system will be described with reference to FIG. In the figure, the cover glass 11 is arranged at a slight angle so that the reflected light from the CCD does not cause flare or ghost. A first optical low-pass filter 19 is provided in front of the variator lens 18 in the middle of the zoom lens system, and a second optical low-pass filter is provided behind the variator lens 18 and in front of the image sensor 15. A filter 20 is provided. These filters 19, 20
Are constituted by parallel plane plates each composed of a plurality of quartz plates. Here, by arranging the first optical low-pass filter 19 in front of the variator lens 18, its characteristics are variable with zooming, and the second optical low-pass filter 20 is located in front of the image sensor 15. Position, given certain characteristics

Although the first and second optical low-pass filters 19 and 20 may be collectively arranged on the front side of the image pickup device 15, as described above, by providing a part in the lens system, In addition to making the number of components relatively small,
There is an advantage that an optical low-pass filter having good characteristics can be configured as described later. Next, the reason for this configuration will be described.

FIG. 51 shows a two-dimensional frequency space on the image plane. The horizontal axis νx coincides with the horizontal scanning direction of the image sensor 15. In the figure, oval-shaped ellipses extending in six directions uniformly about the origin 0 indicate the possible range of image plane frequencies determined by the image guide fiber bundles of various fiberscopes. This is because there are many fiberscopes having different image guide fiber diameters and eyepiece magnifications for each application or type.
The reason that this range exists only in a specific direction is that the image guide fiber bundle forms a regular array usually called a hexagonal close-packed structure. In the figure, regularly arranged circles including those indicated by a and b indicate sampling points determined by the pixel structure of the image sensor 15.

FIG. 52 shows an arrangement of a complementary color type mosaic filter recently used as a color filter of the solid-state imaging device 15. In an image pickup system having a color filter of this type, when reading an image signal, signals of two pixels adjacent in the vertical direction are mixed. That is, in the first field (A field), horizontal scanning is performed by mixing two vertical pixels as shown on the right side of the figure, and in the second field (B field), the horizontal scanning is performed on the left side of the figure. As described above, horizontal scanning is performed by shifting one pixel in the vertical direction.

At this time, the luminance signals of the n and n + 1 lines in the A field are Y _An and Y _{An + 1} , and the luminance signals of the n and n lines in the B field are Y _Bn and Y _{Bn +1} . Assuming that the charges of the pixels due to the transmitted light are Mg, Ye, G, and Cy, respectively, Y _An = (Ye + Mg) + (Cy + G) = Y _{An + 1} = (Ye + G) + (Cy + Mg) = 2R + 3G + 2B Y _Bn = (Mg + Ye) ) + (G + Cy) = Y _{Bn + 1} = (G + Ye) + (Mg + Cy) = 2R + 3G + 2B

Since the color signal is the difference between adjacent horizontal pixels, if they are C _An , C _{An + 1} , C _Bn , and C _{Bn + 1} , C _An = (Ye + Mg)-(Cy + G) = C _Bn = (Mg + Ye) − (G + Cy) = 2R−G C _{An + 1} = (Cy + Mg) − (Ye + G) = C _{Bn + 1} = (Mg + Cy) − (G + Ye) = 2B−G R or B is obtained. In this imaging method, two vertical pixels of the solid-state imaging device correspond to one horizontal scanning line.

Therefore, the horizontal unit cell size P
When x and the unit cell size Py in the vertical direction are determined as shown, for example, the sampling point a in FIG. 51 is indicated by a horizontal frequency νx = １ ／ Px and a vertical frequency νy = 1１ ／ Py. In the figure, if the frequency spectrum of the object exists near these sampling points, the image becomes moiré and difficult to see. For this reason, an optical low-pass filter is arranged so as to drop a specific frequency spectrum of an object. In general, a quartz plate having birefringence is often used as the optical low-pass filter. The trap line indicating the cutoff frequency of each quartz plate is indicated by a straight line in the figure. FIG. 53 shows a side view of this quartz plate, and FIG. 54 shows the crystal axis direction of the quartz plate.

[0023]

By the way, as shown in FIG. 51, if the frequency spectrum of the object exists in a certain fixed area, it is not possible to configure a filter corresponding to each fiberscope. for that reason,
A number of trap lines must be formed so that a plurality of quartz plates can be properly combined and the frequency response in the region can be sufficiently reduced. Conventionally, the trap lines of the quartz plate are arranged so that the response is stopped at a higher frequency side than the sampling points such as a and b in FIG. 51 so as not to reduce the resolution in the horizontal and vertical directions. There were many. In the case of such a configuration, it is good to image a general object, but to image a special object such as the exit end face of a fiberscope, a lower frequency side than the sampling point as described above is used. In many cases, such a response has a specific frequency spectrum determined by the fiber arrangement, and such a method of responding is still insufficient.

Therefore, when the fiberscope is a subject, it is necessary to obtain a response over a wide range from the low frequency side to the high frequency side of these points with these sampling points in between.
Further, when the imaging optical system is a zoom lens, it is more complicated. For example, even if one fiber scope is taken as an example, since the image plane frequency of the fiber scope changes continuously by zooming, the frequency response of the object image should be reduced for each arbitrary zoom state. It is difficult to achieve, and it is not possible to achieve this for all types of fiberscopes.

That is, in the case of using a zoom optical system, for example, in FIG. 51, an oval frequency range by a fiber scope is widened from a solid line to a broken line by zooming, as shown in FIG. With the configuration of the objective low-pass filter, there is a problem that moire removal becomes insufficient. In addition, in order to remove moiré in such a case, it is necessary to add more quartz plates, resulting in an increase in the size of the filter and the television camera. Further, in this case, the response of lower frequency is reduced, so that the resolution of the entire television camera is reduced. In particular, when a zoom optical system is used, moire does not occur in a high frequency region such as a wide state. However, there is a problem that an image is blurred and a resolution is deteriorated, resulting in an adverse effect.

In view of the above problems, the present invention can sufficiently suppress the occurrence of moire in a monitor image,
Moreover, it is an object of the present invention to provide a small and lightweight endoscope imaging device. Another object of the present invention is to provide an imaging apparatus for an endoscope that can sufficiently suppress the occurrence of moire even when a zoom lens is used as an imaging optical system.

[0027]

An endoscope imaging apparatus according to the present invention has a complementary color mosaic filter and obtains a luminance signal of one horizontal scanning line by mixing signals of two pixels in the vertical direction. In the endoscope imaging apparatus including the solid-state imaging device as described above and an imaging optical system having a zooming unit that forms an image of an object on the solid-state imaging device, the first before the zooming unit Along with placing an optical low-pass filter,
A second optical low-pass filter is provided between the magnification unit and the solid-state imaging device, and the unit cell sizes of the solid-state imaging device in the horizontal direction and the vertical direction are Px and Py, respectively, and eight of the two-dimensional frequency space are defined. Sampling point (-
[± 1 / 2Px], ± 1 / 4Py), (± 1 / 2Px,
± 1 / 4Py), (0, ± 1 / 4Py), (± 1 / 2Py)
x, 0) (composite same order), a polygon surrounded by a trap line on the image plane by the first optical low-pass filter is a square in the wide state of the imaging optical system. And in the telephoto state, all of the above trap lines pass through the rectangular area.

Also, a solid-state image pickup device having a complementary color mosaic filter for mixing signals of two pixels in the vertical direction to obtain a luminance signal, and an imaging optics for forming an image of an object on the solid-state image pickup device In the imaging device for an endoscope provided with a system, eight sampling points (− [± 1/1) appearing on a two-dimensional frequency space, where the unit cell sizes in the horizontal and vertical directions of the solid-state image sensor are Px and Py, respectively. 2
Px], ± 1 / 4Py), (± 1 / 2Px, ± 1 / 4P)
y), (0, ± 1 / 4Py), (± 1 / 2Px, 0)
In (composite same order), a soft focus lens for lowering the spatial frequency response at at least four points is provided on the front side of the imaging optical system or in the imaging optical system .

Also, a solid-state imaging device for obtaining a luminance signal of one horizontal scanning line for each pixel in the vertical direction, and an imaging optical system for forming an image of an object on the solid-state imaging device are provided. In the imaging device for an endoscope, eight sampling points (-[± 1 / 2Px], ± 1) appearing in a two-dimensional frequency space, where Px and Py are unit cell sizes in the horizontal and vertical directions of the solid-state imaging device, respectively. / 2Py), (±
1 / 2Px, ± 1/2 Py), (0, ± 1/2 Py),
(± 1 / 2Px, 0) (composite same order), a soft focus lens that lowers the spatial frequency response at at least four points is connected to the front side of the imaging optical system or the imaging light.
It is characterized by being provided in the academic system .

In order to remove moire for each zoom state, not only the specific frequency response of the object is reduced, but also
It is necessary to arrange an optical low-pass filter so as to reduce the response around the sampling point. By the way, it is necessary to suppress occurrence of moiré by making the trap line of the quartz plate come around the sampling point where moiré easily occurs. Also, place a trap line on the quartz plate where moiré is unlikely to occur, such as where the frequency is far from the sampling point or the positional relationship between the sampling point and the specific frequency is shifted due to the regularity of the fiber arrangement. It does not need to be done, and it does not cause much problem in practical use. Also, where the frequency is high, the frequency response characteristics of the optical system itself may be reduced, and it is not necessary to strictly reduce the response around such a point.

However, when an optical low-pass filter is to be constituted only by a quartz plate having a fixed cut-off frequency characteristic, moiré occurs in a telephoto state where the image plane frequency of each fiberscope becomes smaller among the zooming states. In order to avoid this, the response mainly in the low frequency region is degraded, so that the resolving power in the wide state is significantly reduced. Therefore, by arranging the first optical low-pass filter in front of the variator lens,
The cutoff frequency characteristic of the filter by the quartz plate can be made variable. That is, the trap line by the first optical low-pass filter is located on the high frequency side of the eight sampling points in the wide state, and the response of the image plane frequency of the fiberscope is reduced, and moves synchronously with this image plane frequency during zooming. Then, in the telephoto state, it is positioned such that the response of the image plane frequency is lowered on the low frequency side of the sampling point. Thus, if the specific image plane frequency of a certain fiberscope is kept low, the image plane frequency of that fiberscope can always be kept low regardless of the zooming state. The response can be reduced over a wide range up to the side.

When a soft focus lens is used as the optical low-pass filter, there is an area where the response decreases around the sampling point, and the response in this area can be reduced.

[0033]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in detail based on embodiments. The first embodiment of the present invention has the same basic structure as the imaging optical system shown in FIG. 50. In the present embodiment, the first optical low-pass filter disposed in front of the variator lens 18 is denoted by reference numeral 22. , And the second optical low-pass filter disposed behind the variator lens 18 is denoted by reference numeral 23. FIG. 1 shows trap lines on a two-dimensional frequency space by the first and second optical low-pass filters 22 and 23.
(A) shows a wide (wide-angle) state, and (B) shows a tele (telephoto) state. In the figure, in the wide state shown in FIG. 3A, the oval image plane frequency of the fiberscope is located on the high frequency side with respect to the sampling point, but in the telephoto state, the image plane frequency of the fiberscope is reversed to the sampling point. On the other hand, it is located on the low frequency side.

Here, no matter how high the resolution of the fiberscope is, it is not possible to resolve a frequency higher than the sampling frequency of the image pickup device when taking an image with a television camera.
Therefore, since the sampling frequency becomes the resolution limit in the wide state, there is no problem even if the response at a high frequency higher than the sampling point is suppressed. Also, in the telephoto state, the image plane frequency of the fiberscope becomes smaller than the sampling point, so that the resolution limit is determined by the image plane frequency of the fiberscope. Therefore,
If the response is reduced to about the image plane frequency of the fiberscope, the resolution of the fiberscope hardly decreases.

In order to make full use of this feature, the first optical low-pass filter 22 is arranged in front of the variator lens 18 as described above, so that its trap line is indicated by a broken line on the two-dimensional frequency space in FIG. The quartz plate is constructed so as to appear at the position shown. This trap line is moved by zooming and is a movable trap line. In addition, by disposing the second optical low-pass filter 23 behind the variator lens 18 as described above, the trap line of the second optical low-pass filter 23 is located on the two-dimensional frequency space of FIG. The crystal plate is configured to appear at a nearby position. This trap line does not move even by zooming, and is a fixed trap line.

As a guide, the fixed trap line is located on the higher frequency side than the sampling point. The movable trap line is located on the high frequency side of the sampling point in the wide state, is located on the low frequency side of the sampling point in the telescopic state, and is in synchronization with the image plane frequency of the fiberscope during zooming. It moves between low frequencies. Therefore, when the response of the image plane frequency larger than the sampling point is taken, the second optical low-pass filter 23 is used. When the response of the image plane frequency smaller than the sampling point is taken, the first optical low-pass filter 22 is used. To do.

In the two-dimensional frequency space, at least four sampling points a (1 / 2Px, 1 / 4P
y), c (-1 / 2Px,-／ Py), d (1/2
Px,-／ Py), e (-／ Px, ４P)
With respect to the square formed by y), the polygon surrounded by the movable trap line by the first optical low-pass filter 22 is located outside the area of the square in the wide state. Thus, the resolution of the television camera in the wide state is not reduced. Further, in the telephoto state, the movable trap lines are positioned on the image plane frequency in the rectangular area. As a result, without reducing the resolution of the fiberscope in the telephoto state,
Moire can be prevented from occurring.

Moreover, each sampling point a, c,
In each neighborhood of d and e, at least 3
It is preferable to have a zooming state in which more than two trap lines exist. Further, other sampling points f (0, 1 / 2Py) and g (0, -1 / 2P
y) It is good to keep the response in the vicinity. By sandwiching these sampling points a, c, d, e, or f, g by three or more trap lines in this way, the response in the vicinity of these points can be reduced more reliably over a wider range. be able to. Further, the sampling points b (1 / 2Px, 0) and h (-1 / 2Px,
Regarding 0), it is more preferable that a state in which three or more trap lines are interposed similarly exists so that the frequency response in these regions is sufficiently reduced.

FIGS. 2 and 3 show the quartz plates constituting the first and second optical low-pass filters 22 and 23 capable of realizing such a trap line. That is,
FIG. 2 shows a side view of the quartz plate. The first optical low-pass filter 22 includes three quartz plates 22a and 22a.
b, 22c, and the second optical low-pass filter 23 is composed of two quartz plates 23a, 23b. FIG. 3 shows each of the quartz plates 22a, 22a.
It is a figure which shows the crystal axis direction of b, 22c, 23a, 23b, respectively, and is comprised so that the angle which each crystal plate makes may be shifted sequentially by 45 degrees.

Since the present embodiment is configured as described above, for example, if the imaging optical system of the endoscope imaging apparatus according to the present embodiment is in the wide state shown in FIG.
5 (see FIG. 50), six sampling points a,
In the rectangular area formed by b, c, d, e, and h, the dashed movable trap line formed by the first optical low-pass filter 22 is located outside the rectangular area, that is, on the high-frequency side. Does not deteriorate. In this state, the response of the image plane frequency is reduced by the solid fixed trap line formed by the second optical low-pass filter 23 located near the high-frequency side of these sampling points. Resolution can be realized.

Next, when zooming is performed by moving the variator lens 18, the frequency of the movable trap line changes in synchronization with the movement of the image plane frequency of the fiberscope during zooming. A response of about the frequency can be achieved. Further, since the response of the image plane frequency near the sampling point is reduced by the fixed trap line, the response can be reduced over a wide range near the sampling point. In the tele state shown in FIG. 1 (B), it is necessary to take a response at a frequency lower than the sampling point, but this effect is obtained by a movable trap line which moves in synchronization with the image plane frequency of the fiberscope. Can be relatively prevented from lowering the resolution.
Moreover, the response can be reduced over a wide range near the sampling point.

Here, if the response at a frequency lower than the sampling point is reduced by a fixed trap line, the response in the low frequency region is reduced irrespective of the zooming state such as the wide state or the tele state. The decrease is remarkable, and the image quality deteriorates.
It becomes unbearable for observation. In addition, since moire is most likely to occur near the sampling point, in order to prevent moire from occurring even when zooming, it is preferable to reduce the frequency near these points by a fixed trap line. . This is because, when the movable trap line is moved by zooming, if all of the optical low-pass filters are arranged in front of the variator lens 18, depending on the zoom state, the manner in which the optical low-pass filter is degraded becomes insufficient, and moire may occur. .

As described above, according to the present embodiment, in the endoscope image pickup apparatus using the zoom optical system, the occurrence of moire can be reliably prevented regardless of the zoom state, and the optical low-pass filters 22 and 23 can be used. Since the configuration can be optimized, the device can be reduced in size and weight.

In the first embodiment described above, the arrangement of the image guide fiber bundles of the fiberscope has a hexagonal close-packed structure, but the fiber bundle is formed of a random image guide fiber bundle in which quartz fibers are randomly arranged. Even when used in combination with a fiberscope having such a configuration, a sufficient moiré removing effect can be exhibited. This fiberscope basically has no directional characteristics because the image guide fiber bundles are arranged irregularly, and the spacing between the fibers of the image guide also varies. The upper spatial frequency characteristic has a donut shape as shown in FIG. 4 (A) or (B). Therefore, it is necessary to ensure that the frequency response is reduced in all the sampling point areas as compared with the above-described embodiment.

The spatial frequency characteristic shown in FIG. 4 is an example, and if another fiberscope having different distances between the fibers of the image guide and eyepiece magnification of the fiberscope is used, a donut-shaped image plane frequency region is obtained. It goes without saying that it will also be different. However, even in this case, as in another embodiment described below, if the vicinity of each sampling point in the horizontal direction, the vertical direction, and the like is made as uniformly and efficiently as possible, the basic idea of the present invention can be maintained as it is. The present invention can be applied to the case of a random image guide fiber bundle.

Next, specific examples having the same basic configuration as the first embodiment will be sequentially described as Embodiments 1 to 6. In the first embodiment, FIG. 5 is a side view of the quartz plate, and the first optical low-pass filter 25 is constituted by three quartz plates 25a, 25b, and 25c as in the first embodiment. Also, the second optical low-pass filter 26
Is composed of two quartz plates 26a and 26b. FIG. 6 shows each of the quartz plates 25a, 25b, 25c, 26
It is a figure which shows the crystal axis direction of 26a and 26b, respectively, and is comprised so that the angle which each crystal plate makes may be shifted sequentially by 45 degrees. Further, when the angle between the optical axis and the crystal axis is set to about ± 45 °, there is an advantage that the quartz plate can be made thinnest for the same amount of separation. Further, the trap lines in the two-dimensional frequency space of the first embodiment appear as shown in FIG. 7A in the wide state and as shown in FIG. 7B in the tele state.

FIGS. 8 to 10 show a second embodiment. In FIG. 8, a first optical low-pass filter 2 is shown.
7 comprises three quartz plates 27a, 27b and 27c, and the second optical low-pass filter 28 comprises two quartz plates 28a, 27b and 27c.
a, 28b. The crystal axis direction of each quartz plate is shown in FIG.
Is shown in The trap line in the two-dimensional frequency space of the second embodiment appears as shown in FIG. 10A in the wide state and as shown in FIG.

FIGS. 11 to 13 show a third embodiment. In FIG. 11, a first optical low-pass filter 29 is composed of three quartz plates 29a, 29b and 29c and a second optical low-pass filter. The filter 30 also includes three quartz plates 30a, 30b, and 30c. The crystal axis direction of each quartz plate is shown in FIG. Further, the trap line in the two-dimensional frequency space of the third embodiment appears as shown in FIG. 13A in the wide state and as shown in FIG. 13B in the tele state.

FIGS. 14 to 16 show a fourth embodiment. In FIG. 14, the first optical low-pass filter 31 comprises three quartz plates 31a, 31b and 31c, and the second optical low-pass filter The filter 32 includes two quartz plates 32a and 32b. The crystal axis direction of each quartz plate is shown in FIG.
It is shifted by 5 °. The trap line in the two-dimensional frequency space according to the fourth embodiment appears as shown in FIG. 16A in the wide state, and as shown in FIG. 16B in the telescopic state.

FIGS. 17 to 19 show a fifth embodiment. In FIG. 17, a first optical low-pass filter 33 comprises three quartz plates 33a, 33b, 33c and a second optical low-pass filter. The filter 34 also includes three quartz plates 34a, 34b, 34c. The crystal axis direction of each quartz plate is shown in FIG. Further, the trap line in the two-dimensional frequency space according to the fifth embodiment appears as shown in FIG. 19A in the wide state and as shown in FIG.

FIGS. 20 to 22 show a sixth embodiment. In FIG. 20, a first optical low-pass filter 35 is composed of two quartz plates 35a and 35b, and a second optical low-pass filter 36 is provided. Is three quartz plates 36a,
36b and 36c. The crystal axis direction of each quartz plate is shown in FIG. Further, when the crystal axes of adjacent crystal plates are shifted by 90 ° like the crystal plates 35a and 35b and the crystal plates 36b and 36c, the λ / 4 plate 3
The polarization state is eliminated by sandwiching 7, 38.

In the configuration sandwiching the λ / 4 plate, the constituent angles of the respective quartz plates can be arbitrarily set, so that the angles of the quartz plates can be freely determined according to, for example, the arrangement of sampling points. Further, a thin quartz plate or other depolarizing means may be used instead of the λ / 4 plate. Further, these λ / 4, thin quartz plate, other depolarizing means, etc. are used for the first optical low-pass filter 35 or the second optical low-pass filter 3.
6 may be provided, or may be provided at the rearmost side of the first optical low-pass filter 35 or at the frontmost side of the second optical low-pass filter 36 as necessary. The trap line in the two-dimensional frequency space of the sixth embodiment appears as shown in FIG. 22A in the wide state and as shown in FIG. 22B in the telescopic state.

Next, an optical low-pass filter using a soft focus lens instead of a quartz plate will be described as a second embodiment of the present invention. The soft focus lens can arbitrarily give the characteristics of the frequency response of the image by controlling the point image intensity distribution. By using a soft focus lens to generate, for example, spherical aberration or other aberrations to give a certain amount of blur, or by splitting a light beam to form an image, the image is superimposed on the image plane and becomes a kind of blur. Can be given. These blur effects are intended to prevent the occurrence of moiré.

As a specific configuration of such a soft focus lens, it is preferable to use an aspherical lens. Since it is an aspherical surface, the point image intensity distribution can be arbitrarily controlled. Instead of providing an independent lens, a cover glass or an imaging lens of an adapter, a television camera, or the like may be used as an aspherical lens to have a function of a soft focus lens. With such a configuration, there is an advantage that no space is required for the soft focus lens, the overall length can be reduced, and the imaging optical system can be made compact. If a part of the optical low-pass filter, for example, the second optical low-pass filter located on the rear side of the variator lens 18 is constituted by a soft focus lens, the entire optical low-pass filter is formed using only the quartz plate. There is also an advantage that the number of quartz plates can be reduced as compared with the case of (1).

The frequency characteristic when a quartz plate is used as an optical low-pass filter is shown in FIG.
Since it changes functionally, as the frequency increases, MT
The response rises again after the F characteristic becomes 0. On the other hand, when the soft focus lens is used, as shown in FIG.
However, since the MTF characteristic gradually decreases, the frequency range where the MTF characteristic is small is relatively wide. Therefore, it has a characteristic that is convenient for lowering the frequency response near the sampling point.

For this reason, eight sampling points a (1 / 2Px, 1 / 4Py), c on the two-dimensional frequency space where moiré appears strongly by the soft focus lens
(-1 / 2Px,-／ Py), d (1 / 2Px,-
１ ／ Py), e (-／ Px,-／ Py), i
(0, 1/4 Py), j (0,-／ Py), b (1
/ 2Px, 0) and e (-1 / 2Px, 0), it is preferable to reduce the frequency domain response at at least four points. Because, depending on the size of the unit cells Px and Py of the image sensor and the size of the frequency region where the response can be cut by the soft focus lens, one soft focus lens can suppress more than four sampling points at a time. This is because they cannot. However, in such a case, 2
More than one soft focus lens may be used, or an optical low-pass filter of a quartz plate may be added at points that cannot be pressed.

The soft focus lens includes an axisymmetric aspherical lens and an aspherical lens having a certain directivity such as a cylindrical lens. FIG. 25A shows a cross section and a front view of an example of the soft focus lens 38 of the axially symmetric type. The periphery of the axis is raised in a ring shape to form an aspherical surface. Since this lens 38 is axially symmetric, when it is used as an optical low-pass filter, the point image is blurred in a ring shape, so that the MTF characteristic of the region 39 that is axially symmetric in the two-dimensional frequency space is reduced as shown in FIG. Lower. FIG. 26A shows the side surface and the front surface of the soft focus lens 40 having directivity. In the drawing, the soft focus lens 40 has a pair of raised portions in the horizontal direction. The lens 40 has MTF characteristics in a pair of regions 40a and 40b parallel to the horizontal direction as shown in FIG.

In the case of the axially symmetric soft focus lens 38 shown in FIG. 25, four sampling points a (1/1) on the two-dimensional frequency space where moiré appears strongly in FIG.
2Px, 1 / 4Py), c (-1 / 2Px,-／ P)
y), d (1 / 2Px,-／ Py), e (-／)
Px, 1 / 4Py). Also, in the vertical direction, the sampling point i
(0, 1/4 Py), between f (0, 1/2 Py), and j
The setting is such that a response in an area between (0,-／ Py) and g (0,-／ Py) is taken. Also, the sampling points f (0, 1 / 2Py) and g (0,-
Moiré also occurs in the vicinity of (ＰPy), so this region is also suppressed.

In the directional soft focus lens 40 shown in FIG. 26, a (1 / 2Px, 1 / 4Py), i
(0, 1 / 4Py), e (-1 / 2Px, 1 / 4Py)
And c (− ／ Px, − ／ Py), j (0, −1
/ 4Py) and d (1 / 2Px,-／ Py) at the respective sampling points in the vertical direction.

The above description is applicable to a case where a complementary color mosaic filter as shown in FIG. 52 is used for an image sensor to generate color signals for four pixels in the vertical direction. , B, and B illuminating the subject with the illumination light of each color to capture and combine the images of each color to obtain a color image. It cannot be directly applied to a three-plate type imaging device that combines three images obtained by the elements. That is,
Since the color filter shown in FIG. 52 is not used in the image pickup system of the frame sequential color system or the three-plate color image pickup system, the horizontal line of the solid-state image pickup device coincides with the horizontal scanning line. Therefore, when applying the description of the method of FIG. 52 to an image pickup system of a frame sequential color method or a three-plate color image pickup system, if 2Py is replaced with Py considering the vertical direction in half, the above-described description can be obtained. Will be accepted.

Next, an example of the configuration of an image pickup optical system when a soft focus lens is used as an optical low-pass filter will be described with reference to FIGS. FIG. 1A shows a cover glass of a television camera as a soft focus lens 42, one or both sides of which are aspherical. Further, the soft focus lens 42 may be coated with an interference film on the surface so as to have a role as an infrared cut filter. Behind it, a zoom lens 41, an optical low-pass filter 23 composed of a quartz plate, and an image sensor 15
Are sequentially arranged. In FIG. 6B, the infrared cut filter 43 and the soft focus lens 44 are separately provided, and the two are arranged upside down in FIG.

FIG. 4D shows the infrared cut filter 43.
Entrance-side lens surface 4 of the imaging optical system 45 located behind
5a is an aspherical surface and is made into a soft focus lens. In FIG. 7E, the cover glass 46 and the imaging optical system 45 are formed as soft focus lenses, and the number of components of the ordinary optical low-pass filter 23 made of a quartz plate or the like can be reduced by employing a plurality of soft focus lenses. There are advantages that can be reduced.

In these arrangements, a soft focus lens is provided on the front side of the lens system. By disposing the soft focus lens in the vicinity of the entrance pupil position, the soft focus lens can be uniformly distributed over the center and periphery of the screen. A soft focus effect can be achieved. When the cover glass 42 is a soft focus lens as shown in FIG. 3A, the cover glass 42 may be detached and replaced with another cover glass. For example, by using a normal cover glass when using a rigid scope, and using a cover glass 42 as a soft focus lens when using a fiber scope, it is possible to prevent occurrence of moiré of the fiber scope without lowering the resolution of the rigid scope. Can be.

As shown in FIGS. 28A and 28B,
An adapter 47 that can be attached to a rigid scope and has a built-in parallel flat cover glass 46 and an adapter 49 that can be attached to a fiberscope and has a built-in soft focus lens 48 may be provided. Thus, the soft focus lens 48 can be selectively used by using either the rigid endoscope or the fiberscope. Alternatively, these adapters 47 and 49 may be provided in one camera control unit and switched and used as needed.

FIG. 29 shows the observation optical system 50 and the photographing optical system 5.
1 shows an imaging device including a half prism 5;
A soft focus lens 48 is provided in the optical path on the imaging optical system 51 side after the optical path is divided by 2. According to this configuration, since the soft focus lens 48 is provided after splitting the optical path, there is an advantage that unnecessary reduction in resolution during visual observation can be prevented. Further, in this case, instead of the soft focus lens 48, the lens surface of the photographing optical system 53 may be made aspherical to form a soft focus lens.

FIG. 30 shows a main part of a three-chip type image pickup apparatus, in which a soft focus lens 55 is provided in front of a photographing optical system 54. Instead of this configuration, a soft focus lens 55 may be provided in the photographing optical system 54. Also in this example, as an optical low-pass filter,
The soft focus lens 55 and the quartz plate may be used together.

If the characteristics of the soft focus lens are made variable, the soft focus lens can be optimized according to various fiberscopes. Examples of a method of changing the characteristics include changing the shape of the lens surface by changing the shape, changing the refractive index inside the lens, and the like. One example is shown in FIG. 31, in which a liquid crystal (twisted nematic liquid crystal) lens 57 is joined to a soft focus lens 56. When the state is changed from a state in which a voltage is applied to the liquid crystal lens 57 as shown in FIG. 7A to a state in which no voltage is applied as shown in FIG. You can change the rate. Thus, the characteristics of the soft focus lens 56 can be changed. Such a lens may be used in combination with a quartz plate or another soft focus lens.

When a soft focus lens having a directivity as shown in FIG. 26A is used by attaching it to an image pickup apparatus, the positional relationship with the image pickup element 15 of the television camera must be strictly adapted. . Unless the positions of the soft focus lens and the CCD are accurately determined, a good image cannot be obtained. As a means for this, it is necessary to form a positioning portion for front / back discrimination and direction discrimination in the soft focus lens itself.
This will be described below with reference to FIGS. In FIG. 32, the soft focus lens 58 is Y
It is formed in an aspherical shape along the axial direction, and a lower portion 58a of the outer periphery is cut in parallel with the aspherical shape direction. Thereby, the arrangement direction can be determined. Further, by performing chamfering 58b on the outer edge of one surface, the front and back sides can be distinguished. Note that the size of the chamfer may be changed between the front and back surfaces.

In FIG. 33, small projections 58c and 5 are provided at two positions above and below one surface of the soft focus lens 58.
8d are provided. Since the soft focus lens 58 can be manufactured by molding, it can be easily formed by forming a concave portion in the mold. The main body to which the soft focus lens 58 is attached is connected to the projections 58c and 5c.
With a frame structure that fits with 8d, it is possible to assemble with high accuracy.

The lower part 58 as shown in FIG.
If a cannot be successfully formed by molding, as shown in FIG.
It is also possible to form a linear minute protrusion 58e in which the direction can be determined, and to additionally fabricate a flat lower portion 58a as shown in FIG. 32 based on the minute protrusion 58e. FIG.
The soft focus lens 58 shown in FIG. 5 has a relatively large projection 58f protrudingly formed on the peripheral edge of one surface, and if the frame of the main body that fits the projection 58f is manufactured, the direction can be accurately set.

However, in each case, it is necessary to provide the above-described positioning portions outside the effective diameter of the lens. In addition, these positioning portions may be formed in any direction as long as the direction of the aspheric surface is clear. Further, these positioning portions may be provided not only at one position but also at two or more positions, or a plurality of positioning portions may be combined. Of course, the unevenness of the positioning portion may be formed in reverse.

The soft focus lens may be directly mounted on the television camera, or may be mounted in an adapter, and the adapter may be mounted on the television camera. In such a case, it is necessary to precisely determine the positional relationship between the soft focus lens in the adapter and the CCD of the television camera. To do so, use a one-touch type attachment means that does not allow the adapter to be attached unless the adapter is positioned in a fixed direction in accordance with the vertical direction of the CCD, or if it is a screw-in type, if both are fixed, It is only necessary to cut off certain screws so that they are fixed in relation.

FIG. 36 is a cross-sectional view showing an example of a state in which a soft focus lens 58 having a positioning portion as shown in FIG. FIG. 37 is an enlarged view of the main part, and a pair of protrusions 58c and 58d
Nine recesses 59a and 59b are fitted and assembled. In this state, the direction is precisely set. If such a positioning means is not provided, it is necessary to accurately determine the direction by using a microscope during assembly.

Next, specific examples having the same basic configuration as the second embodiment will be described below as Embodiments 7 to 9.
The configuration other than the optical low-pass filter is omitted here. In the seventh embodiment shown in FIGS. 38 and 39, FIG. 38 is a side view of the soft focus lenses 60 and 61 having directivity. One soft focus lens 60 has one side surface and the side surface orthogonal thereto. And the side as viewed from the direction of movement. As can be seen, the soft focus lenses 60 and 61 used in this embodiment each have an aspheric surface having a pair of raised portions in one direction. One soft focus lens 60 has another one so that a pair of raised portions are arranged in a horizontal direction.
Two soft focus lenses 61 having the same shape are used in combination such that a pair of ridges are arranged in a direction orthogonal to the soft focus lens 61.

In this configuration, a double image blurred in each of the horizontal and vertical directions is formed. Therefore, in the two-dimensional frequency space of the seventh embodiment shown in FIG. Are regions 60a and 60b due to the soft focus lens 60, and regions 61a and 61b due to the soft focus lens 61. Therefore, the frequency response around the sampling points a, i, e, d, j, c, b, h, etc. in the horizontal and vertical directions in this area is reduced. Note that the regions 60a, 60
The angle between b and the regions 61a and 61b is 90 °,
This angle may be set as many times as necessary. Further, in this example, the soft focus lens is constituted by both having directivity, but an axially symmetric type and a lens having directivity may be combined.

Incidentally, the MTF characteristic of the soft focus lens does not always become zero as shown in FIG. 24, and if an optical low-pass filter is constituted by this lens alone, the moire removal may be insufficient. In that case, an optical low-pass filter may be configured in combination with a quartz plate. Since the quartz plate can set the MTF characteristic to 0 by the trap line, it is preferable to arrange the quartz plate so as to reduce the moire level. for that reason,
For example, b (1 / 2Px, 0), h (-1 / 2Px,
0), a (1 / 2Px, 1 / 4Py), c (-1 / 2P
x,-／ Py), d (１ ／ Px,-／ P)
It is preferable to reduce the frequency response of each sampling point in the horizontal direction of y) and e (− ／ Px, ４Py). Examples thereof will be described below as Examples 8 and 9.

FIGS. 40 to 42 show an eighth embodiment. In FIG. 40, an optical low-pass filter includes a soft focus lens 60 having a pair of raised portions arranged in a vertical direction, and a λ / 4 plate. It consists of two quartz plates 62a and 62b sandwiching 63. Also, each of the quartz plates 62a, 62
The crystal axis direction of b is shown in FIG.
Is slightly thicker. Example 8 shown in FIG.
In the two-dimensional frequency space, the trap lines formed by the quartz crystals 62a and 62b have horizontal sampling points a and
It appears to decrease the frequency response around e, b, h, d, and c.

FIGS. 43 to 45 show the ninth embodiment. In FIG. 43, the optical low-pass filter comprises the same soft focus lens 60 as in the eighth embodiment and two quartz plates 64a and 64b. Also, each crystal plate 64a,
The crystal axis directions of 64b are shown in FIG.
The crystal plates are arranged so as to be shifted by 5 ° in the crystal axis direction, and the thicknesses of the two quartz plates are the same. The trap line formed by the quartz crystals 62a and 62b in the two-dimensional frequency space of the ninth embodiment is shown in FIG.
It appears as shown in FIG.

In the eighth and ninth embodiments, the optical low-pass filter is formed by combining one directional soft focus lens and two quartz plates. The number of components and how to combine are not limited to these cases,
It can be selected as appropriate. Also, i (0, 1) in the vertical direction
/ 4Py) and j (0,-／ Py), when the influence of moire at the sampling points is weak, it is not necessary to provide a soft focus lens in which an area where the response is reduced particularly in the vertical direction appears. When two soft focus lenses are used, the front and back surfaces of one lens may be aspherical.

By the way, even when a quartz plate or a soft focus lens is used as the optical low-pass filter,
Some moire may occur in combination with a fiberscope using an image guide fiber having a hexagonal close-packed structure or a random image guide fiber. Needless to say, one of the causes is related to the component processing accuracy and the assembly accuracy. In such a case, if the fiber scope or the television camera is slightly rotated, the moire may disappear or its level may be weakened. It is appropriate that the rotation angle is within a range of about ± 30 ° (the sign is used to distinguish the rotation direction). Even when it is rotated further, moire often occurs again due to the problem of the direction in which the fibers are stacked.

In addition, moire can be removed by defocusing. Specifically, it is conceivable to move a part of the lens of the imaging optical system, move the entire lens, or move the image sensor. This makes it possible to reduce only the moiré level without substantially reducing the resolving power when trying to improve the level at which moiré is slightly anxious by the fiber scope.

[0082]

As described above, the imaging device for an endoscope according to the present invention is optically controlled in a range from the wide state to the telephoto state so that the occurrence of moiré can be sufficiently suppressed during monitor observation of an endoscope image. The low-pass filter can be optimized, and the device can be reduced in size and weight.

[Brief description of the drawings]

FIG. 1 is a diagram showing a trap line on a two-dimensional frequency space, an image plane frequency of a fiberscope, and a sampling point of an image sensor according to a first embodiment of the present invention, and FIG.
Indicates a wide state, and (B) indicates a telephoto state.

FIG. 2 is a side view of each quartz plate constituting the optical low-pass filter of the first embodiment.

FIG. 3 is a view showing a crystal axis direction of each quartz plate of FIG. 2;

FIG. 4 is a diagram showing a two-dimensional frequency space when the fiberscope is a random image guide fiber, where (A) shows a wide state and (B) shows a tele state.

FIG. 5 is a side view of each quartz plate constituting the optical low-pass filter in the first embodiment of the first embodiment.

FIG. 6 is a view showing a crystal axis direction of each crystal plate of FIG. 5;

FIGS. 7A and 7B are diagrams showing trap lines and the like in a two-dimensional frequency space similar to FIG. 1 according to the first embodiment, where FIG. 7A shows a wide state and FIG.

FIG. 8 is a side view of each quartz plate constituting the optical low-pass filter in the second embodiment.

FIG. 9 is a view showing a crystal axis direction of each quartz plate of FIG. 8;

10A and 10B are diagrams illustrating trap lines and the like in a two-dimensional frequency space similar to FIG. 1 according to the second embodiment, where FIG. 10A illustrates a wide state and FIG.

FIG. 11 is a side view of each quartz plate constituting an optical low-pass filter according to a third embodiment.

FIG. 12 is a view showing a crystal axis direction of each quartz plate of FIG. 11;

13A and 13B are diagrams showing trap lines and the like in a two-dimensional frequency space similar to FIG. 1 according to the third embodiment, where FIG. 13A shows a wide state and FIG.

FIG. 14 is a side view of each quartz plate constituting an optical low-pass filter according to a fourth embodiment.

FIG. 15 is a view showing a crystal axis direction of each quartz plate of FIG. 14;

FIGS. 16A and 16B are diagrams showing trap lines and the like in a two-dimensional frequency space similar to FIG. 1 according to the fourth embodiment, where FIG. 16A shows a wide state and FIG.

FIG. 17 is a side view of each of the quartz plates constituting the optical low-pass filter according to the fifth embodiment.

18 is a view showing a crystal axis direction of each quartz plate of FIG.

19A and 19B are diagrams illustrating trap lines and the like in a two-dimensional frequency space similar to FIG. 1 according to the fifth embodiment, where FIG. 19A shows a wide state and FIG.

FIG. 20 is a side view of each of the quartz plates constituting the optical low-pass filter according to the sixth embodiment.

21 is a view showing a crystal axis direction of each quartz plate of FIG. 20.

22A and 22B are diagrams illustrating trap lines and the like in a two-dimensional frequency space similar to FIG. 1 according to the sixth embodiment, where FIG. 22A illustrates a wide state and FIG.

FIG. 23 is a diagram showing MTF characteristics of an optical low-pass filter using a quartz plate.

FIG. 24 is a diagram showing MTF characteristics of an optical low-pass filter using a soft focus lens.

FIG. 25 is a diagram illustrating a blur region in a two-dimensional frequency space of an optical low-pass filter using an axially symmetric soft focus lens.

FIG. 26 is a diagram illustrating a blur region in a two-dimensional frequency space of an optical low-pass filter using a directional type soft focus lens.

FIGS. 27 (A), (B), (C), (D), and (E) are diagrams each showing an imaging optical system using a soft focus lens as an optical low-pass filter.

28 (A) is an explanatory view in which an adapter incorporating a parallel plane plate is attached to a rigid endoscope, and FIG. 28 (B) is an explanatory view in which an adapter incorporating a soft focus lens is attached to a fiberscope.

FIG. 29 is a configuration diagram of an imaging optical system including an observation optical system and a photographing optical system.

FIG. 30 is a configuration diagram of a three-plate imaging optical system having a soft focus lens.

FIGS. 31A and 31B are configuration diagrams in which a liquid crystal lens and a soft focus lens are combined, wherein FIG. 31A shows a state when power is supplied and FIG.

32A and 32B are a side view and a front view showing a positioning portion of a soft focus lens having directivity.

FIGS. 33A and 33B are a side view and a front view showing a positioning portion of a soft focus lens having directivity. FIGS.

34A and 34B are a side view and a front view showing a positioning portion of a soft focus lens having directivity.

FIGS. 35A and 35B are a side view and a front view showing a positioning portion of a soft focus lens having directivity. FIGS.

FIG. 36 is a cross-sectional view showing a state where the soft focus lens of FIG. 33 is attached to an imaging device main body.

FIG. 37 is an enlarged view of a main part of FIG. 36.

FIG. 38 is a side view of a soft focus lens and a quartz plate constituting an optical low-pass filter according to a seventh embodiment of the second embodiment.

FIGS. 39A and 39B are diagrams illustrating a region where a response in a two-dimensional frequency space is reduced in the seventh embodiment, where FIG. 39A illustrates a wide state and FIG.

FIG. 40 is a side view of a soft focus lens and a quartz plate constituting an optical low-pass filter according to an eighth embodiment.

FIG. 41 is a diagram showing a crystal axis direction of each quartz plate of FIG. 40.

FIG. 42 is a diagram illustrating a region where a response is reduced in a two-dimensional frequency space and a trap line according to the eighth embodiment.

FIG. 43 is a side view of a soft focus lens and a quartz plate constituting an optical low-pass filter according to a ninth embodiment.

FIG. 44 is a view showing a crystal axis direction of each quartz plate of FIG. 43.

FIG. 45 is a diagram illustrating a region where a response is reduced in a two-dimensional frequency space and a trap line according to the ninth embodiment.

FIG. 46 is a schematic explanatory view of a conventional endoscope imaging apparatus.

FIG. 47 is an explanatory diagram of a main part showing another configuration example of a conventional endoscope imaging apparatus.

FIG. 48 is an explanatory view of a principal part showing another configuration example of a conventional endoscope imaging apparatus.

FIG. 49 is an explanatory view of a principal part showing another configuration example of a conventional endoscope imaging apparatus.

FIG. 50 is a configuration diagram of an imaging optical system in which optical low-pass filters are arranged before and after an imaging optical system, respectively.

FIG. 51 is a diagram illustrating trap lines and the like of the optical low-pass filter in FIG. 50 in a two-dimensional frequency space.

FIG. 52 is a diagram showing an arrangement of complementary color type mosaic filters.

FIG. 53 is a side view of the quartz plate of the optical low-pass filter of FIG. 50.

54 is a view showing a crystal axis direction of the quartz plate of FIG. 53.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 15 Image sensor 18 Variator lens 22 First optical low-pass filter 23 Second optical low-pass filter 38, 40 Soft focus lens

──────────────────────────────────────────────────続 き Continued on the front page (58) Field surveyed (Int.Cl. ⁷ , DB name) H04N ^9/ 04-9/11

Claims (3)

(57) [Claims] 1. A solid-state imaging device having a complementary color mosaic filter and mixing signals of two pixels in a vertical direction to obtain a luminance signal of one horizontal scanning line, and an image of an object on the solid-state imaging device. An imaging device for an endoscope having an imaging optical system having a variable power section that forms a first optical low-pass filter before the variable power section, and the variable power section and the solid A second optical low-pass filter is provided between the solid-state imaging device and the imaging device, and the horizontal and vertical unit cell sizes of the solid-state imaging device are Px and Py, respectively, and eight sampling points (− [ ± 1 / 2Px],
± 1 / 4Py), (± 1 / 2Px, ± 1 / 4Py),
A rectangular area formed by (0, ± 1/4 Py), (± 1/2 Px, 0) (composite same order) is surrounded by a trap line on the image plane by the first optical low-pass filter. An imaging device for an endoscope, wherein a polygon surrounds the rectangular area when the imaging optical system is in a wide state, and the trap lines all pass through the rectangular area when in a telescopic state. 2. A solid-state image pickup device having a complementary color mosaic filter for obtaining a luminance signal by mixing signals of two pixels in a vertical direction, and an imaging optics for forming an image of an object on the solid-state image pickup device. In the imaging device for an endoscope provided with a system, the unit cell size in the horizontal direction and the vertical direction of the solid-state imaging device is Px and Py, respectively, and eight sampling points (− [± 1 / 2P
x], ± 1 / 4Py), (± 1 / 2Px, ± 1 / 4P)
y), (0, ± 1 / 4Py), (± 1 / 2Px, 0)
In the (composite order), a soft focus lens for lowering a spatial frequency response at at least four points is provided on the front side of the imaging optical system or in the imaging optical system . Imaging device. 3. A solid-state imaging device for obtaining a luminance signal of one horizontal scanning line for each pixel in a vertical direction, and an imaging optical system for forming an image of an object on the solid-state imaging device. In the imaging device for an endoscope, eight sampling points (− [± 1 / 2P) appearing in a two-dimensional frequency space, where the unit cell sizes of the solid-state imaging device in the horizontal direction and the vertical direction are Px and Py, respectively.
x], ± 1/2 Py), (± 1 / 2Px, ± 1 / 2P
y), (0, ± 1/2 Py), (± 1/2 Px, 0)
In the (composite order), a soft focus lens for lowering a spatial frequency response at at least four points is provided on the front side of the imaging optical system or in the imaging optical system . Imaging device.