JP2014097124A - Imaging system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an imaging system capable of accommodating to a case when observation is performed by switching illumination light, and forming an image of the quality easy to observe and no conspicuous noise.SOLUTION: An imaging system that switches between first illumination light by a narrow band filter group 6 and second illumination light by a broad band filter group 7 to illuminate the body cavity of a patient 9, includes an AGC circuit 47 for amplifying an imaging signal imaged with a CCD 35, and an emphasizing circuit 55 for performing contour emphasizing processing. According to the content of the contour emphasizing processing, and the information on the combination of the first illumination light by the narrow band filter group and the second illumination light by the broad band filter group, the imaging system performs a control to suppress the maximum value of a gain to amplify the imaging signal by the AGC circuit 47.

Description

  The present invention relates to an imaging system for imaging a subject.

In recent years, an imaging system including an imaging unit such as an endoscope inserted into a body cavity has been widely used.
In addition to the normal light observation mode or the white light observation (WLI observation) mode for imaging under the illumination of the normal visible region illumination light, the narrow-band illumination light is different from the visible region illumination light. In addition, an imaging system capable of performing special light observation such as a narrow-band light observation (NBI observation) mode in which imaging is performed is also put into practical use.
For example, a first conventional example of WO2010 / 131620 discloses a frame sequential imaging device for performing narrowband light observation. Narrowband light observation focuses on the use of light that is strongly absorbed by blood and strongly reflected / scattered by the mucosal surface layer in order to observe blood vessels with high contrast. By irradiating the biological tissue sequentially with the narrow-band light and imaging the reflected light, the contrast between the capillary blood vessel in the mucosal surface layer and the deep blood vessel in the deep part is highlighted.

In addition, since the blue narrow band light tends to be insufficient in the amount of illumination light compared with the green narrow band light, in the first conventional example of the above publication, the green narrow band light G and the blue narrow band light Two narrowband lights B1 and B2 are sequentially irradiated, and narrowband light observation is performed using a narrowband light observation image created from reflected light images (narrowband images) corresponding to the narrowband lights G, B1 and B2. I have to. In addition, the imaging device of the first conventional example discloses a configuration for performing brightness control.
In the second conventional example of Japanese Patent Application Laid-Open No. 2001-8097, an image sensor that images the inside of an observation object based on light irradiation from a light source, and exposure information is input to variably control light output from the light source. A light amount control circuit, a gain control circuit that variably controls the gain of the image signal output from the image sensor, a detail circuit that performs edge enhancement on the image signal, and the exposure amount information results in insufficient image brightness. When it is detected that the light output by the light amount control circuit is at the upper limit value, the gain control circuit increases the gain of the image signal and decreases the detail circuit detail amount in proportion to the increase of the gain value. An endoscope system provided with a control circuit for controlling in this way is disclosed.

WO 2010/131620 JP 2001-8097 A

In the first conventional example, the observation light can be observed in the normal light observation mode and the narrow band light observation mode by switching between the illumination light in the visible region and the illumination light in the narrow band. In the case of performing observation with narrow-band light where the amount of light tends to be insufficient, there is a drawback that noise becomes conspicuous for brightness control.
In the second conventional example, when the exposure amount is insufficient, the gain of the image signal is increased by the gain variable means so that an image having a predetermined brightness can be obtained, and is proportional to the increase of the gain value. The detail circuit is controlled to reduce the amount of detail, but it is assumed to be applied to the case of normal illumination light.Therefore, noise is generated in the case of narrowband light that is much smaller than normal illumination light. Has the disadvantage of becoming prominent.
The present invention has been made in view of the above-described points, and provides an imaging system that can cope with a case where observation is performed by switching illumination light, and that can generate an image having an image quality that is easy to observe without noticeable noise. For the purpose.

An imaging system of one embodiment of the present invention includes illumination light in a first illumination band having a first narrow band and a second narrow band, and illumination light in a second illumination band having a plurality of wide bands. Illuminating means for switching and illuminating; imaging means for imaging the subject illuminated by the illuminating means;
A signal level variable means for changing an imaging signal level of an imaging signal output by the imaging means, an image signal is generated from the signal output from the signal level variable means, and an edge enhancement process is performed on the generated image signal. A combination of the first narrow band and the second narrow band forming the first illumination band in the illumination unit, and the second Signal level variable control means for controlling the imaging signal level by the signal level variable means in accordance with the information of the combination of the plurality of wide bands forming the illumination band.

  ADVANTAGE OF THE INVENTION According to this invention, it can respond to the case where it observes by switching illumination light, and can produce | generate the image of the image quality which is easy to observe, without noise conspicuous.

FIG. 1 is a diagram illustrating an overall configuration of an imaging system according to a first embodiment of the present invention. FIG. 2 is a diagram showing the transmission characteristics with respect to the wavelength of the narrow band filter constituting the first filter group provided on the inner peripheral side of the rotary filter. FIG. 3 is a diagram illustrating transmission characteristics with respect to wavelengths of a wideband filter constituting a second filter group provided on the outer peripheral side of the rotary filter. FIG. 4 is a block diagram illustrating a configuration of a brightness calculation unit. FIG. 5 is a block diagram illustrating a configuration of a synthesis processing unit. FIG. 6A is a diagram illustrating a characteristic example in which the noise reduction intensity is changed stepwise according to the synthesis rate. FIG. 6B is a diagram illustrating a characteristic example in which the intensity of noise reduction is continuously changed according to the synthesis rate. FIG. 7A is a diagram showing control contents for suppressing the maximum gain value of the AGC circuit in accordance with the enhancement level in the narrow-band light observation mode. FIG. 7B is a diagram showing the control contents for suppressing the maximum gain value of the AGC circuit according to the enhancement level in the normal light observation mode. FIG. 8 is a diagram showing control contents for suppressing the maximum gain value of the AGC circuit according to the type of emphasis performed by the emphasis circuit. FIG. 9 is a flowchart showing the processing contents in the first embodiment. FIG. 10 is a diagram illustrating an overall configuration of an imaging system according to a modification of the first embodiment. FIG. 11 is a diagram showing the contents for limiting the maximum gain value in the NBI observation mode in the modification of the first embodiment. FIG. 12 is a diagram illustrating a state in which a displayed image is divided into a plurality of regions. FIG. 13A is a diagram showing the content of changing frequency characteristics for performing structure enhancement according to a bright region and a dark region. FIG. 13B is a diagram showing the contents of changing the intensity and the like for structure enhancement according to the presence or absence of the electronic zoom according to the presence or absence of the electronic zoom. FIG. 14 is a diagram showing a relationship between a defective pixel and its surrounding eight pixels when the image sensor is read out using two lines of an A channel and a B channel. FIG. 15 shows the timing of an image input to an image memory of a 65 conversion circuit that converts an image of 60 frames / second into an image of 50 frames / second, the timing of an image converted from 65 and output from the image memory, and the like. Timing diagram. FIG. 16 is a timing chart showing the timing at which 65-converted image signals are written into the synchronization memory and read out as synchronized RGB signals.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
(First embodiment)
An imaging system 1 according to the first embodiment of the present invention shown in FIG. 1 includes an endoscope 2 that is inserted into a body cavity of a patient 9 and performs an endoscopic examination using an affected part in the body cavity as a subject, and an endoscope. 2 is a light source device 3 that selectively supplies illumination light composed of a plurality of narrow-band lights and illumination light composed of a plurality of broadband lights, and signal processing that performs signal processing on the imaging means mounted on the endoscope 2. A video processor 4 as a means and a color monitor 5 as a color display means for color-displaying an image signal output from the video processor 4 are provided.
The endoscope 2 includes an elongated and flexible insertion portion 11 so that it can be easily inserted into a body cavity, an operation portion 12 provided at the rear end (base end) of the insertion portion 11, And a universal cable 13 extending from a side portion or the like. A light guide 14 for transmitting illumination light is inserted into the insertion portion 11, the operation portion 12, and the universal cable 13, and the light source connector 15 provided at the end of the light guide 14 includes the first narrow band and the second narrow band. The illumination light of the first illumination band having a narrow band and the illumination light of the second illumination band having a plurality of broadbands are detachably connected to the light source device 3 constituting the illumination means for performing illumination. Is done.

The light source device 3 has a lamp 21 that emits light by a lamp lighting power source by a lamp power circuit 20. The lamp 21 is constituted by a xenon lamp, for example, and generates broadband light covering the visible region.
Then, as will be described below, the light source device 3 includes the broadband R, G, B plane sequential illumination light corresponding to the normal light observation mode or the white light observation mode (WLI observation mode or WLI mode), and narrowband light observation. One surface sequential illumination light of narrow band G and B surface sequential illumination light corresponding to the mode (NBI observation mode or NMI mode) is supplied to the light guide 14.
Note that the narrow band G and B plane sequential light tends to be short of the amount of illumination light compared to the broadband R, G and B plane sequential illumination light, and is narrower than the narrow band G plane sequential light. Since the amount of illumination of the B-band sequential light in the band is likely to be small due to transmission loss in the light guide 14 and the like, in the present embodiment, as described below, one illumination of the narrow-band G-plane sequential light is performed. The narrow-band B-plane sequential light illumination is performed twice. One of the two times is B1 illumination light and the other is B2 illumination light.

When the main narrow-band B-plane sequential light is the narrow-band B1 plane-sequential light and the sub (slave) narrow-band B-plane sequential light is the narrow-band B2 plane-sequential light, in this embodiment, the narrow-band G , B1 and B2 are sequentially illuminated, or narrow-band B2, G, and B1 are sequentially used for illumination. Then, by combining the captured images captured with the same narrowband B1 illumination light and B2 illumination light, respectively, the B1 captured images (or B2 capture images) respectively captured with the narrowband B1 or B2 illumination light. An image brighter than (image) can be generated.
Further, when the amount of illumination light is insufficient, the imaging signal level is increased by the gain of the AGC circuit 47 as signal level changing means for changing the imaging signal level so that a bright image can be generated. However, due to frame sequential illumination, when a moving subject is imaged, a contour shift or the like occurs in the composite image. Therefore, as described later, in this embodiment, the maximum value of the gain by the AGC circuit 47 is limited according to the observation mode, the contour enhancement level, and the synthesis coefficient (or synthesis ratio) when two images are synthesized, and noise is reduced. Try to get an image that doesn't stand out.

The light of the lamp 21 is adjusted by the diaphragm 22A arranged in the illumination optical path, and after passing through the rotating color filter 24 rotated by the motor 23, is condensed by the condenser lens 25 and is condensed into the light guide. 14 is incident on the incident end face. The aperture 22A varies the size of the aperture that determines the amount of light passing through the aperture 22A by the aperture drive circuit 22B. The diaphragm 22A has a sensor (not shown) that detects the opening amount of the diaphragm 22A. In the open state where the maximum opening amount is reached, the maximum opening amount or a signal that detects the open state is driven by the diaphragm. It outputs to the control circuit 32 mentioned later via the circuit 22B.
The sensor may output directly to the control circuit 32 without going through the aperture drive circuit 22B.
The rotating color filter 24 includes a plurality of fan-shaped filters respectively disposed along the inner circumferential side circumferential direction and the outer circumferential side circumferential direction, and includes illumination light in the first illumination band and illumination light in the second illumination band. Has a first filter group 6 and a second filter group 7.

G filters 6G, B filters 6B1 and 6B2 that respectively transmit narrowband light in the green (G) and blue (B) wavelength bands are arranged in the annular portion on the inner peripheral side, and the first filter group 6 Composed. The B filters 6B1 and 6B2 are filters having the same characteristics. The B filter 26B1 is a filter used mainly, and the B filter 26B2 is a filter used as a sub.
R, G, and B filters 7R, 7G, and 7B that transmit red (R), green (G), and blue (B) broadband light respectively are disposed on the outer ring portion, and the second fill Group 7 is configured.
FIG. 2 shows transmission characteristics of the G filter 6G and the B filters 6B1 and 6B2. Specifically, the G filter 6G transmits a narrow band G light of 530 to 550 nm centered at 540 nm, and the B filters 6B1 and 6B2 have a narrow band B light of 400 to 430 nm centered at 415 nm, for example. Transmits (also referred to as B1 light or B2 light). In FIG. 2, the code filters 6G, 6B1, and 6B2 are not shown, and their transmission characteristics are indicated by G and B1 (B2).

FIG. 3 shows transmission characteristics of R, G, B filters 7R, 7G, 7B (indicated by Rw, Gw, Bw in the figure). R, G, B filters 7R, 7G, 7B The light passes through a wide wavelength band of G and B, respectively. The first filter group 6 and the second filter group 7 are also referred to as a narrow band filter group and a wide band filter group, respectively.
In this specification, the transmission characteristics of the G filter 6G, B filters 6B1 and 6B2 having the narrow band transmission characteristics shown in FIG. 2 are indicated by G and B1 (B2), and the broadband transmission characteristics shown in FIG. The transmission characteristics of the R, G, and B filters 7R, 7G, and 7B, or the light thereof, are indicated by Rw, Gw, and Bw.
A motor 23 that rotates the rotary color filter 24 is attached to a rack 28, and the rack 28 meshes with a pinion gear 29 a attached to the rotation shaft of the moving motor 29. 1 corresponds to the WLI observation mode in which the R, G, B filters 7R, 7G, 7B (Rw, Gw, Bw) on the outer peripheral side of the rotating color filter 24 are arranged in the illumination optical path. Illumination is performed, and rotation by the motor 23 causes broadband R, G, B plane sequential illumination light to enter the incident end face of the light guide 14.

The operation unit 12 of the endoscope 2 is provided with an observation mode switch 31 as an observation mode switching means for selecting or switching an observation mode, and the control circuit 32 is operated by the light source device 3 in response to the operation of the observation mode switch 31. The illumination light is switched and the operation corresponding to the illumination light by the video processor 4 is controlled. For example, when a user such as an operator operates the observation mode switch 31 to switch from the WLI observation mode to the NBI observation mode, the control circuit 32 in the video processor 4 rotates the motor 29 to rotate the color. The filter 24 and the motor 23 are moved upward as indicated by arrows in FIG.
When the rotating color filter 24 moves upward as indicated by a two-dot chain line, the inner B2G and 6B2 filters 6B2, 6G, and 6B1 are switched to a state where they are arranged in the illumination optical path. In this state, the rotating color filter 24 is rotated by the motor 23 so that the narrow-band B2, G, B1 plane sequential illumination light enters the incident end face of the light guide 14.
The illumination light transmitted by the light guide 14 is emitted into the body cavity as illumination light from the light guide distal end surface facing the illumination window of the distal end portion 33 of the insertion portion 11, and illuminates a subject such as an affected part in the body cavity. An illuminated subject is, for example, a charge imaging element (CCD and CCD) as an imaging element in which an imaging surface is arranged at an imaging position by an objective lens system 34 attached to an observation window provided adjacent to the illumination window. (Abbreviation) 35 is imaged. The image pickup means for picking up the subject illuminated by the illumination means includes an objective lens system 34 that connects an optical image of the subject, and a CCD 35 arranged at the image formation position.

The CCD 35 constituting the image pickup means is connected to an electrical contact of a signal connector 36 provided at an end of the universal cable 13 through a signal line inserted through the insertion portion 11 and the like. The signal connector 36 is detachably connected to the video processor 4. The CCD 35 outputs a signal obtained by photoelectrically converting an optical image formed on the imaging surface of the CCD 35 as an imaging signal by applying a CCD driving signal generated by a CCD driver 41 provided in the video processor 4.
In the case where the light source device 3 generates broadband R, G, B plane sequential illumination light, the CCD 35 performs a photoelectric conversion signal on the basis of the broadband R, G, B plane sequential illumination light. Wide-band R, G, and B imaging signals are output in frame sequential order.
On the other hand, when the light source device 3 generates narrow-band B2, G, B1 plane sequential illumination light, the CCD 35 performs photoelectric conversion under the narrow-band B2, G, B1 plane sequential illumination light, respectively. The narrowband B2, G, and B1 imaging signals as signals are output in frame sequential order.

In the following, in most cases, broadband R, G, B plane sequential illumination light, Rw, Gw, Bw plane sequential light, broadband R, G, B imaging signals, Rw, Gw, Bw It is expressed as an imaging signal, and is distinguished from narrow-band B2, G, B1 plane sequential illumination light and narrow-band B2, G, B1 imaging signals.
Imaging signals (specifically, imaging signals of Rw, Gw, and Bw or imaging signals of B2, G, and B1) output from the CCD 35 in the surface order are input to the analog processing unit 42 in the video processor 4 and are analog. After being amplified with a low noise figure by a preamplifier constituting the processing unit 42, a baseband signal component is extracted by a correlated double sampling circuit (abbreviated as CDS circuit). The baseband signal that has passed through the CDS circuit is converted into a digital imaging signal by the A / D conversion circuit 43, and then the synchronization control unit 44 that synchronizes the frame sequential imaging signal and the brightness of the captured image are calculated. This is input to the brightness calculation unit 45.

The synchronization control unit 44 is connected to a synchronization memory 46 for synchronizing frame-sequential imaging signals. The synchronization memory 46 has R, G, B memories 46a, 46b, 46c. When Rw, Gw, and Bw imaging signals are input, the synchronization control unit 44 stores the Rw, Gw, and Bw imaging signals in the R, G, and B memories 46a, 46b, and 46c, respectively.
On the other hand, when B2, G, and B1 image signals are input, the synchronization control unit 44 stores the B2, G, and B1 image signals in the R, G, and B memories 46a, 46b, and 46c, respectively.
The synchronization control unit 44 stores the broadband Rw, Gw, and Bw imaging signals or the narrowband B2, G, and B1 imaging signals in the synchronization memory 46, and then simultaneously reads the imaging signals (imaging signals) in each frame. (Signal) level is output to an automatic gain control circuit (abbreviated as AGC circuit) 47 as an imaging signal level varying means for changing the level.

The brightness calculation unit 45 calculates the brightness of the captured image by matrix processing (or color conversion matrix processing) by a matrix processing unit 65 provided in the brightness calculation unit 45. The brightness calculation unit 45 calculates the average brightness (or luminance) of the captured image for a time longer than at least one frame period. The user may be allowed to set the time for calculating the average brightness (or luminance).
In the WLI observation mode, the brightness calculation unit 45 calculates one type of brightness corresponding to the imaging signals of Rw, Gw, and Bw with broadband Rw, Gw, and Bw sequential illumination light.
On the other hand, in the NBI observation mode, within a predetermined time (for example, one frame period or two frame periods), the B2, G illumination light and the B1, G illumination light in the narrow band B2, G, B1 plane sequential illumination light Two sets of brightness (first brightness and second brightness) are calculated from imaging signals respectively captured with illumination light.
In the present embodiment, the B1 illumination light is also referred to as master illumination light, and the B2 illumination light is also referred to as slave illumination light that is used as an auxiliary (subordinate) to the master illumination light. B1 and G illumination light are also referred to as master side illumination light, and B2 and G illumination light are also referred to as slave side illumination light.

The output signal of the AGC circuit 47 is input via a changeover switch 52 to a composition processing unit 51 that constitutes an image processing unit 50 as an image processing unit including an enhancement circuit 55 that performs contour enhancement processing as described below. Is done. The image processing unit 50 generates an image signal from the signal output from the AGC circuit 47 serving as the imaging signal level varying unit, and performs edge enhancement processing and the like on the generated image signal.
In the NBI observation mode, the output signal of the AGC circuit 47 is transmitted via the changeover switch 52 to the B1 captured image (captured by the CCD 35 under the B1 illumination light) and (captured by the CCD 35 under the B2 illumination light). The data is input to the synthesis processing unit 51 that synthesizes the B2 captured image. The composition processing unit 51 outputs the image signal generated by the composition to the subsequent stage side.
In the WLI observation mode, the switch 52 is controlled by the control circuit 32 so that the setting a at the contacts a and b is selected. In this case, the output signal of the AGC circuit 47 is input to a noise reduction (abbreviated as NR) circuit 53 that performs noise reduction processing bypassing the synthesis processing unit 51.

The NR circuit 53 is connected to a ROM 53a that stores processing intensity (also referred to as NR intensity) at the time of noise reduction processing and the value of the synthesis rate by the synthesis processing unit 51 as information that is associated like a lookup table. . Then, the NR circuit 53 refers to the information in the ROM 53a and performs a noise reduction operation with an intensity corresponding to the value of the synthesis rate.
The image signal output from the NR circuit 53 is input to the enlargement / reduction circuit 54. The enlargement / reduction circuit 54 enlarges (electronic zoom) or reduces the image size of the input image signal by electrical signal processing. Do. The output signal of the enlargement / reduction circuit 54 is input to the enhancement circuit 55, and the enhancement circuit 55 performs contour enhancement processing for enhancing the contour.
The image signal output from the enhancement circuit 55 is input to the matrix processing unit 56, and the matrix processing unit 56 outputs R, G, B images output to the RGB channels of the color monitor 5 by matrix processing in units of one frame period. Signals Ro, Go, Bo are generated.

The matrix processing unit 56 becomes a unit matrix in the WLI observation mode, and adopts a matrix including α, β, and γ in the expression (1) described later when changing the color tone in the NBI observation mode.
The image signals Ro, Go, Bo generated by the matrix processing unit 56 are converted into analog image signals Rt ′, Gt ′, Bt ′ by the D / A conversion circuit 57, and then converted into the RGB channels of the color monitor 5. Entered.
On the color monitor 5, the WLI observation image is displayed in color in the WLI observation mode, and the NBI observation image is displayed in color or pseudo color in the NBI observation mode.

Further, the video processor 4 has an operation panel 58 for performing various settings such as a composition ratio setting unit (or composition coefficient setting unit) 58a and a contour enhancement level setting unit 58b as composition coefficient setting means (or composition ratio setting means). The signal set by the operation panel 58 is input to the control circuit 32.
A user such as an operator performs an operation of setting the composition ratio a ′ from the composition ratio setting unit 58a, whereby the composition ratio a ′ is input to the composition processing unit 51 via the control circuit 32, and the composition processing unit 51 Using this synthesis ratio a ′ as the synthesis ratio ac used when actually synthesizing, two sets of imaging signals (B1 imaging signal and B2 imaging signal) are synthesized.
The composition processing unit 51 preferentially uses the composition ratio a ′ set by the user. When this setting is not performed, the composition ratio a described later calculated by the brightness calculation unit 45 is used as the composition ratio. The synthesis process is performed as ac.

In FIG. 1, a line for directly outputting the synthesis ratio information from the brightness calculation unit 45 to the synthesis processing unit 51 is shown, but the information of the synthesis ratio a calculated by the brightness calculation unit 45 is output to the control circuit 32. The control circuit 32 may collectively control the synthesis processing of the synthesis processing unit 51. In the present embodiment, the following description is given assuming the latter case.
Further, the brightness calculation unit 45 outputs the difference of the calculated brightness from the target brightness to the dimming control unit 59. The dimming control unit 59 outputs the input difference signal as a dimming signal to the aperture driving circuit 22B of the light source device 3, and the aperture driving circuit 22B adjusts the amount of light passing through the aperture 22A based on the dimming signal. .
Further, in the video processor 4, the control circuit 32 has, for example, a flash memory 60 as a control information storage unit that stores control information 60 a that limits the range of the AGC gain (hereinafter simply referred to as gain) of the AGC circuit 47. .

The control circuit 32 refers to the control information 60 a stored in the flash memory 60 and controls the gain of the AGC circuit 47 with the gain control signal Cg applied to the gain control terminal of the AGC circuit 47. The AGC circuit 47 is composed of an AGC circuit having three characteristics, and amplifies the imaging signals B2, G, and B1 with the same gain according to the voltage value of the gain control signal Cg applied to the gain control terminal. Output to the rear stage.
As will be described later, in the NBI observation mode, as a synthesis coefficient in the case of synthesizing the B1 imaging signal and the B2 imaging signal in accordance with the difference between the two sets of brightness calculation values and the target brightness, etc. The composite ratio a is variably set.
The brightness calculation unit 45 in the present embodiment calculates the average brightness (luminance) of the captured image captured by the CCD 35 with the configuration shown in FIG. 4, for example. FIG. 4 shows an example of calculating the average brightness when the predetermined time is set to, for example, two frame periods. However, it is set to one frame period T1 and set to one frame period T1. Average brightness may be calculated, or average brightness may be calculated in a plurality of frame periods such as three frame periods. The dotted line in FIG. 4 will be described later.

As shown in FIG. 4, the brightness calculation unit 45 includes, for example, three sets of frame memories 61R, 61R ′, 61G, 61G ′, 61B, 61B ′, and Rw, Gw, Bw imaging signals or B1, G, The imaging signals of / Rw, / Gw, / Bw or / B2, / G, / B1, which are the average values (average signal values) of the imaging signals of a plurality of frames (here, 2 frames) of the imaging signal of B2 are calculated. An average value calculation unit 62 is included.
In the configuration example shown in FIG. 4, when the Rw or B2 imaging signal of the first frame is input, the control circuit 32 switches the changeover switches SW1 and SW2a, and the Rw or B2 imaging signal is sent to the frame memory 61R. Store. The imaging signal of Rw or B2 stored in the frame memory 61R is indicated by Rw-1 (B2-1). When the Gw or G imaging signal of the next first frame is input, the control circuit 32 switches the selector switches SW1 and SW2b and stores the Gw or G imaging signal in the frame memory 61G. The Gw or G imaging signal stored in the frame memory 61G is denoted by Gw-1 (G-1).

Further, when an imaging signal of Bw or B1 in the next first frame is input, the control circuit 32 switches the selector switches SW1 and SW2c and stores the imaging signal of Bw or B1 in the frame memory 61B. An image signal of Bw or B1 stored in the frame memory 61B is indicated by Bw-1 (B1-1).
When the image signal of Rw or B1 in the next second frame is input, the control circuit 32 switches the changeover switches SW1 and SW2a and stores the image signal of Rw or B2 in the frame memory 61R ′. The imaging signal of Rw or B2 stored in the frame memory 61R ′ is indicated by Rw-2 (B2-2). Similarly, an imaging signal indicated by Gw-2 (G-2) is stored in the frame memory 61G ′, and an imaging signal indicated by Bw-2 (B1-2) is stored in the frame memory 61B ′.

The Rw-1 (B2-1) imaging signal stored in the frame memory 61R and the Rw-2 (B2-2) imaging signal stored in the frame memory 61R ′ are added to the corresponding pixels by the adder 63a. Are added together to calculate / Rw (/ B2) which is an average value.
For example, the base-bound imaging signal component that is imaged by the CCD 35 under the first Rw (or B2) illumination light, amplified by the preamplifier, and extracted by the CDS circuit is converted into a digital signal by the A / D conversion circuit 43. When the signal level of the two-dimensional pixel position (i, j) in the imaging signal Rw-1 (B2-1) converted into is expressed as Rw-1 (i, j) (B2-1 (i, j)) , / Rw = (Rw-1 (i, j) + Rw-2 (i, j)) / 2, and / B2 = (B2-1 (i, j) + B2-2 (i, j)) / 2 Become. In addition, although the example which calculates an average value from the image pick-up signal for 2 frames was shown, you may calculate an average value with several frames more than 3 frames.

Similarly, the Gw-1 (G-1) imaging signal stored in the frame memory 61G and the Gw-2 (G-2) imaging signal stored in the frame memory 61G 'are added by the adder 63b and averaged. A value / Gw (/ G) is calculated.
Similarly, the Bw-1 (B1-1) imaging signal stored in the frame memory 61B and the Bw-2 (B1-2) imaging signal stored in the frame memory 61B ′ are added by the adder 63c. The average value / Bw (/ B1) is calculated.
The adders 63a, 63b, and 63c constituting the average value calculation unit 62 respectively calculate the image pickup signals of / Rw (/ B2), / Gw (/ G), and / Bw (/ B1) calculated as R image and G image, respectively. , B image brightness calculation units 64a, 64b and 64c.
The R image, G image, and B image brightness calculation units 64a, 64b, and 64c, for example, output image signals of / Rw (/ B2), / Gw (/ G), and / Bw (/ B1) to pixels in the effective pixel region. Accumulated in units and divided by the number of pixels (of the effective pixel area used for integration) brightness of each component image <Rwf>, <Gwf>, <Bwf>(<B2f>,<Gf>,<B1f>).

なお、（１）式におけるα、β、γは、マトリクス係数である。α、β、γは、上述したマトリクス処理部５６におけるＮＢＩ観察モード時におけるカラーモニタ５においてカラー表示する場合の色調を希望する色調に変更したものに対応している。 The brightness <Rwf>, <Gwf>, <Bwf>(<B2f>,<Gf>,<B1f>) of each component image is input to the matrix processing unit 65. In the WLI observation mode, the matrix processing unit 65 calculates Rm, Gm, and Bm signals using a unit matrix.
That is, the brightness <Rwf>, <Gwf>, <Bwf> of each input component image is output as Rm, Gm, Bm signals. On the other hand, in the case of the NBI observation mode, the matrix processing unit 65 calculates signals of Rm, Gm, and Bm by performing matrix processing shown in the following equation (1). The calculated Rm, Gm, and Bm signals are output to the luminance calculation unit 66.
[Formula 1]

In the equation (1), α, β, and γ are matrix coefficients. α, β, and γ correspond to those in which the color tone in the color monitor 5 in the NBI observation mode in the matrix processing unit 56 is changed to a desired color tone.

These values can be appropriately selected and set from a range of 0.7 to 1.5, for example.
Rf, Gf, and Bf on the right side of the equation (1) are <R1f>, <Gf>, <B1f> and <B1f> respectively corresponding to the case of the master side illumination light and the case of the slave side illumination light in the NBI observation mode. R2f>, <Gf>, <B2f>. Here, Gf is used in common in the case of the master side illumination light and in the case of the slave side illumination light, but it is not limited to the case of using in common. When Gf is shared, there is an advantage that image shift can be reduced.
In the NBI observation mode, Equation (1) calculates two sets of Rm, Gm, and Bm signals. The luminance calculation unit 66 calculates the luminance Y by the following equation (2) in the WLI observation mode.

Y = 0.3Rm + 0.59Gm + 0.11Bm (2)
On the other hand, in the NBI observation mode, the luminance calculation unit 66 calculates the luminance Y1 by the following equation (3) for the case of the master side illumination light.
Y1 = 0.3Rm + 0.59Gm + 0.11Bm
= 0.3 · α <G> + 0.59 · β <B1> + 0.11 · γ <B1> (3)
In the NBI observation mode, the luminance calculation unit 66 calculates the luminance Y2 by the following equation (4) for the case of slave side illumination light.
Y2 = 0.3Rm + 0.59Gm + 0.11Bm
= 0.3 · α <G> + 0.59 · β <B2> + 0.11 · γ <B2> (4)
Further, the luminance calculation unit 66 determines the difference ΔYw (= Yws−Y) between the target brightness (luminance) Yws and the luminance Y in the case of the WLI observation mode, and the target brightness in the case of the NBI observation mode. A difference calculation unit 66a for calculating a difference ΔY1 (= Ys−Y1) between the brightness (luminance) Ys and the luminance Y1.

The target brightness storage unit 67 configured by a memory or the like stores a target brightness (luminance) Yws value in the WLI observation mode and a target brightness (luminance) Ys value in the NBI observation mode. Then, the difference calculating unit 66a calculates the differences ΔYw and ΔY1 with reference to the information stored in the target brightness storage unit 67.
In the case of the WLI observation mode or the NBI observation mode, the brightness calculation unit 45 (the luminance calculation unit 66) outputs the difference ΔYw or ΔY1 to the dimming control unit 59, and the dimming control unit 59 outputs the difference ΔYw. Alternatively, the aperture amount of the diaphragm 22A is adjusted via the diaphragm drive circuit 22B according to ΔY1.
For example, when the difference ΔYw is positive (when the calculated luminance Y is smaller than the target brightness (luminance) Yws), the dimming control unit 59 outputs a dimming signal that controls to open the aperture 22A. However, when the difference ΔYw is negative, a dimming signal for controlling to close the aperture 22A is output.

Even in the NBI observation mode in which the difference or ΔYw is replaced with the difference ΔY1, the dimming control unit 59 performs the same control. Further, in the NBI observation mode, when the difference ΔY1 is 0 or less, the target brightness (luminance) Ys is obtained only by the master side illumination. The synthesis ratio ac as a synthesis coefficient in the case of synthesizing these images is set to zero.
Further, the AGC gain of the AGC circuit 47 is controlled via the control circuit 32 in accordance with the differences ΔYw and ΔY1 calculated by the brightness calculation unit 45. In the present embodiment, the AGC gain control range (maximum gain value) of the AGC circuit 47 is set to be different between the case of the WLI observation mode and the case of the NBI observation mode, as will be described later (FIG. 7A, see FIG. 7B).
In addition, in the NBI observation mode, the brightness calculation unit 66 of the brightness calculation unit 45 calculates the composition ratio a constituting the composition coefficient by the following equation (5) when the difference ΔY1 is ΔY1> 0. And a composite ratio calculation unit 66b that constitutes a calculation means.

ΔY1 = Y2 × a, a = ΔY1 / Y2 (5)
The combination ratio calculation unit 66b constitutes a calculation unit that calculates the combination ratio a from the difference ΔY1 and the ratio of the luminance Y2 (ΔY1 / Y2).
The composition ratio calculation unit 66b outputs the calculated composition ratio a to the composition processing unit 49 via the control circuit 32, for example. The combination processing unit 51 sets the calculated combination ratio a to the combination ratio ac that is actually combined in the combination processing unit 51 (in the case where the combination ratio a ′ is not set by the combination ratio setting unit 58a). The captured image of B1 is synthesized with the captured image of B2 by slave illumination.

On the other hand, when the composition ratio a ′ is set by the composition ratio setting unit 58a, the composition processing unit 51 preferentially sets the composition ratio a ′ to the composition ratio ac and combines the images. Used for. Note that the composition processing unit 51 actually generates a composite image including a G captured image as shown in equation (6) described later or FIG. However, the G captured image is generated by simply multiplying by a coefficient of 1 + ac. For this reason, when the definition of combining a plurality of images captured at different timings is used, a combination of the captured image of B1 and the captured image of B2 is a captured image corresponding to the definition.
For example, the synthesized images output from the synthesis processing unit 51 are set as Rin, Gin, and Bin as the captured images of B2, G, and B1, which are the captured image components of each frame constituting the moving image input to the synthesis processing unit 51. Assuming that the images are Rt, Gt, and Bt, the composition processing unit 51 generates a composite image signal such as the following equation (6) and outputs it to the subsequent stage.

（６）式によれば、合成処理部５１は、Ｒｔ，Ｇｔ，ＢｔとしてＢ２，（１＋ａｃ）Ｇ，ａｃ・Ｂ２＋Ｂ１の画像を生成する。
このように、合成処理部５１は、出力画像成分としてのＢｔの合成画像としてａｃ・Ｂ２＋Ｂ１画像を生成する。また、合成処理部５１は、Ｂｔの合成画像の生成に対応して、Ｇｔの画像成分として（１＋ａｃ）・Ｇの画像を生成する。
合成処理部５１の後段側における画像処理部５０を構成するＮＲ回路５３、拡大／縮小回路５４、強調回路５５等においては、合成処理部５１により生成された合成画像の画像信号に対して、それぞれノイズ低減、拡大／縮小処理、輪郭強調処理等を行う。
なお、本明細書においては、合成処理部５１により生成される主要な合成画像成分Ｂｔを生成するＢ１の撮像画像とＢ２の撮像画像とそれぞれ乗算する係数の比１：ａｃを合成比率とも言う。 [Formula 6]

According to Expression (6), the composition processing unit 51 generates an image of B2, (1 + ac) G, ac · B2 + B1 as Rt, Gt, and Bt.
In this way, the composition processing unit 51 generates an ac · B2 + B1 image as a composite image of Bt as an output image component. Further, in response to the generation of the Bt composite image, the composition processing unit 51 generates a (1 + ac) · G image as the Gt image component.
In the NR circuit 53, the enlargement / reduction circuit 54, the enhancement circuit 55, and the like constituting the image processing unit 50 on the subsequent stage side of the synthesis processing unit 51, the image signal of the synthesized image generated by the synthesis processing unit 51 is respectively applied. Noise reduction, enlargement / reduction processing, contour enhancement processing, and the like are performed.
In the present specification, the ratio 1: ac of coefficients for multiplying the captured image of B1 and the captured image of B2 that generate the main composite image component Bt generated by the composite processing unit 51 is also referred to as a composite ratio.

FIG. 5 shows a configuration example of the composition processing unit 51. The captured image (imaging signal) of B2 passes through the synthesis processing unit 51 and becomes an Rt image signal. In addition, the captured images of G and B1 (the captured signals thereof) are respectively input to one input terminal of the multiplier 71a and the adder 71b, and the multiplier 71a and the adder 71b are combined images of Gt and Bt from the output terminal. Is output to the NR circuit 53 as an image signal.
The other input terminal of the multiplier 71a receives 1 + ac as a coefficient to be multiplied from the control circuit 32. The multiplier 71a multiplies the captured image of G by the coefficient of 1 + ac to synthesize (1 + ac) G. Output an image.
The other input terminal of the adder 71b receives an ac · B2 captured image obtained by multiplying the captured image of B2 by the coefficient ac by the multiplier 71c, and the adder 71b adds both captured images to ac. -A composite image of B2 + B1 is output. The coefficient ac is output by the control circuit 32.

In the configuration of equation (6) or FIG. 5, the composition processing unit 51 shows an example of generating an image of B2, (1 + ac) G, ac · B2 + B1 as Rt, Gt, Bt. A matrix coefficient or a multiplication coefficient may be set so as to output ac · B2.
Further, the synthesis processing unit 51 may generate the 0, (1 + ac) G, ac · B2 + B1 images as Rt, Gt, and Bt in a more simplified manner.
In the present embodiment, the control circuit 32 changes the intensity of noise reduction by the NR circuit 53 according to the synthesis ratio ac or the synthesis rate. In the present embodiment, as shown in FIG. 6A or FIG. 6B, control is performed to increase the noise reduction intensity according to the value of the synthesis rate. In FIG. 6A or 6B, the case where the composition ratio ac is 0 is represented as 0% composition rate, and the case where the composition ratio ac is 1 is represented as 100% composition rate.
In FIG. 6A, the noise reduction strength is set to NRmin at a synthesis rate of 0%, and the noise reduction strength is increased stepwise as the synthesis rate increases, and the noise reduction strength is increased at a synthesis rate of 100%. NRmax is set. In FIG. 6B, the noise reduction strength is set to 0 at a synthesis rate of 0%, the noise reduction strength is continuously increased as the synthesis rate increases, and the noise reduction strength is set to NRmax at a synthesis rate of 100%. ing.

なお、Ｒｔ′として０に設定しても良い。 When the synthesis rate is large, the noise reduction strength is increased and the noise reduction function is increased. On the other hand, when the synthesis rate is small, the intensity of noise reduction is made smaller (lower) than when the synthesis rate is high, and deterioration of the resolution of the image during the noise reduction process is prevented. In other words, in this embodiment, noise reduction processing is performed so that good noise reduction can be performed while maintaining the resolution of the image.
Further, in the NBI observation mode, when the input image is Rt ′, Gt ′, and Bt ′, the matrix processing unit 56 generates Ro, G, and Bo images as shown in the equation (7) and performs D / D The data is output to the R, G, and B channels of the color monitor 5 via the A conversion circuit 57.
[Formula 7]

Rt ′ may be set to 0.

In the present embodiment, the control circuit 32 (imaging signal level varying means) according to the observation mode, the enhancement level L when the enhancement circuit 55 performs edge enhancement, and the composition ratio ac in the composition processing unit 51. The gain of the AGC circuit 47 is controlled. That is, the control circuit 32 has a function of a gain control unit 32a that controls the gain of the AGC circuit 47 as the signal level varying means. The gain control unit 32a refers to the control information 60a stored in the flash memory 60 and controls the gain of the AGC circuit 47 so as to reduce noise.
FIG. 7A shows contents for limiting the gain range of the AGC circuit 47 in the NBI observation mode in accordance with the enhancement level L and the composition ratio ac. In this way, by limiting the gain of the AGC circuit 47, the present embodiment makes it possible to obtain an image that is easy to observe without causing noticeable noise in the image displayed on the color monitor 5.

As shown in FIG. 7A, the enhancement level L increases when the contour is enhanced in eight stages from 0 to 7. An enhancement level L of 0 indicates that no enhancement is performed. In the present embodiment, control (limitation) is performed so that the maximum gain value of the AGC circuit 47 decreases as the enhancement level L increases.
Further, when the composition ratio is set (that is, when the composition ratio ac is greater than 0 and equal to or less than 1) compared to when the composition ratio is not set (in other words, when the composition ratio ac is 0), the AGC circuit. Control (limitation) is performed so that the maximum gain value of 47 becomes a smaller value. In the specific example of FIG. 7A, when the synthesis ratio ac is greater than 0 when the synthesis ratio ac is 0, when the synthesis ratio ac is greater than 0, the gain is limited to a half gain maximum value.
In the gain control content shown in FIG. 7A, noise becomes more conspicuous (noisy) as the contour emphasis level L increases, so that the maximum gain value is limited. Further, if the synthesis ratio ac is set to be greater than 0, it becomes noisy than when the synthesis ratio ac is 0. Therefore, the maximum gain value is limited to be smaller than when the synthesis ratio ac is 0.

By limiting the maximum gain value in this way, it is possible to make the noise of the image displayed on the color monitor 5 inconspicuous. In FIG. 7A, the maximum gain value may be limited to a value that more reflects the value of the combination ratio ac, as indicated by the dotted line in accordance with the ratio between the emphasis level L and the combination ratio.
When the maximum gain value is limited as shown by the dotted line in FIG. 7A, the value corresponding to the composition ratio when generating the main image component (B1 + ac · B2) in the composite image is (B1 + ac · B2) ≈ (1 + ac) B1. And (1 + ac) can be regarded as the magnitude of the synthesis ratio. The gain control unit 32a of the control circuit 32 suppresses the maximum gain value as (1 + ac) increases. The reason is that the picked-up images of B1 and B2 are images picked up at different timings. Therefore, when the moving image is displayed on the color monitor 5, a smaller (1 + ac) has a contour and structure. It is considered that the image has good image quality with little deviation.

For this reason, when the value of (1 + ac) becomes large, the deviation of the outline and the structure can be a factor of deteriorating the image quality. Therefore, by suppressing the gain so that the maximum gain value becomes smaller as (1 + ac) becomes larger as described above, it is possible to obtain an image with good image quality in which noise is not conspicuous.
On the other hand, in the WLI observation mode, since the composition processing unit 51 does not operate, the maximum gain value of the AGC circuit 47 is determined only by the enhancement level L. FIG. 7B shows the maximum gain value set corresponding to the enhancement level L in the WLI observation mode. Even in the WLI observation mode, noise becomes more conspicuous (noisy) as the contour emphasis level increases, so that the maximum gain value is limited.

In the WLI observation mode, the amount of illumination light can be made considerably larger than in the NBI observation mode, so that the noise level can be made considerably smaller than in the image obtained in the NBI observation mode.
Therefore, in the specific example shown in FIG. 7B, the gain maximum value is limited to be smaller as the enhancement level L is larger than in the NBI observation mode, but the gain is more relaxed than in the NBI observation mode. It is limited to the maximum value. As described above, the present embodiment has an object to provide an imaging system that can cope with a case where observation is performed in different observation modes by switching illumination light, and that can generate an image with an image quality that is easy to observe without noticeable noise. Achieve. In particular, the illumination light in the first illumination band having the first narrow band (specifically G) and the second narrow band (specifically B1 and B2) in which the illumination light tends to be insufficient. In the case of the NBI observation mode imaged in (1) and (2), the maximum gain value is limited or suppressed so as to generate an image having an image quality that is easy to observe without causing noticeable noise.

In FIG. 7A, the enhancement level L for edge enhancement has been described. However, an AGC circuit is used depending on the type of enhancement processing such as structure enhancement for emphasizing a pattern or structure with a frequency band along with edge enhancement (edge enhancement) in an image. The maximum gain value of 47 may be limited.
For example, in the structure enhancement A, the luminance level in the frequency band below the frequency fa is emphasized, and in the structure enhancement B, the luminance level in the frequency band below the frequency fb (fb> fa) higher than the frequency fa is emphasized. The maximum gain value may be limited as shown in FIG. 8 in the structure enhancement to be performed and the outline enhancement at an enhancement level L (for example, L = 2) that is substantially the same as the enhancement level in these structure enhancements.

In the example shown in FIG. 8, when the enhancement level L is the same, the structure enhancement to the higher frequency fb results in a noisy image, so the structure enhancement B is more than the structure enhancement A. The maximum gain at is limited so as to be small.
Further, when the enhancement level L is the same, the contour enhancement results in a noisy image as compared with the structure enhancement. Therefore, the maximum gain value in the case of the contour enhancement is smaller than that in the case of the structure enhancement B. To be limited. By doing so, it is possible to display an image in which noise is not noticeable. FIG. 8 shows the case of one emphasis level. When the emphasis level L is varied, the maximum gain value may be changed as shown in FIG. 7A. Also, FIG. 8 can be applied to the WLI observation mode.

  As described above, the imaging system 1 of the present embodiment includes the illumination light in the first illumination band having the first narrow band and the second narrow band, and the illumination light in the second illumination band having a plurality of wide bands. The light source device 3 that constitutes the illumination means that performs illumination, the CCD 35 that constitutes the imaging means that images the subject illuminated by the illumination means, and the imaging signal level of the imaging signal output by the imaging means An AGC circuit 47 as a signal level changing means to be changed, and an image processing means including an enhancement circuit 55 for generating an image signal from a signal output from the signal level changing means and performing an edge enhancement process on the generated image signal. As the image processing unit 50, the content of contour enhancement processing performed by the image processing means, and the first illumination band including the first narrow band and the second narrow band in the illumination means Gain control unit as signal level variable control means for controlling the imaging signal level by the signal level variable means according to information of illumination light and illumination light of the second illumination band having the plurality of wide bands 32a.

  In the present embodiment, the illumination in the first illumination band is a predetermined time N × T1 which is an integer N times the one frame period T1, where the time for one rotation of the rotating color filter 24 is one frame period T1. And a combination of the first M1 illumination with the first narrowband illumination light and the second M2 (= 2 × M1) illumination with the second narrowband illumination light, Furthermore, the first imaging signal by the imaging means based on the illumination of the first predetermined number M1-1 (= M1) of the first number of times M1, and the second number of the second number of times M2. First brightness is calculated by color conversion matrix processing using the second imaging signal by the imaging means based on illumination of M2-1 (= M1) predetermined times, and the first brightness of the first number M1 is calculated. No. 3 based on M3-1 (= M1) illumination The second brightness is obtained by color conversion matrix processing using one imaging signal and a third imaging signal based on the illumination of the fourth predetermined number M4-1 (= M1) of the second number of times M2. On the basis of a ratio between the first brightness and a predetermined target brightness, and the ratio between the second brightness and the brightness calculation unit 45 serving as a brightness calculation means for calculating A synthesis ratio calculating unit 66b as a calculation unit that calculates a synthesis coefficient when the first and third imaging signals are combined with the first and second imaging signals, and the image processing unit includes: The first and second imaging signals in each one frame period based on the synthesis coefficient by the arithmetic means or the synthesis coefficient set by the synthesis ratio setting unit 58a as the synthesis coefficient setting means. And third shot The AGC circuit 47 serving as the signal level variable means is composed of a combination of the first narrow band and the second narrow band. In the case of an imaging signal imaged under illumination light in the first illumination band, the first, second, and third are based on the contents of the edge enhancement performed by the image processing means and the synthesis coefficient. The signal level varying means for processing the image pickup signal is controlled. The above N, number of times M1, M2, etc. may be set to values such as once or twice, or may be set so as to satisfy the relationship indicated by ().

Next, the operation of this embodiment will be described with reference to the flowchart of FIG. When the power supply of the imaging system 1 is turned on, the control circuit 32 performs control so that illumination and imaging are performed in the WLI observation mode, for example, as the initial setting process in the first step S1. That is, as shown in step S2, the control circuit 32 controls the motor 29 of the light source device 3 so that the wide band filter group 7 on the outer peripheral side of the rotating color filter 24 is arranged in the illumination optical path. The subject is illuminated with broadband illumination light by the broadband filter group 7, that is, broadband R, G, B illumination light (in other words, Rw, Gw, Bw illumination light).
Further, as shown in step S3, the control circuit 32 images the subject illuminated with the broadband illumination light with the CCD 35, and the video processor 4 generates a broadband imaging signal or image signal.
In that case, in step S4, the gain control unit 32a of the control circuit 32 suppresses the maximum gain value by the AGC circuit 47 according to the value of the enhancement level L. That is, the higher the enhancement level L, the more the maximum gain value by the AGC circuit 47 is suppressed.

Then, as shown in step S5, the color monitor 5 displays the WLI image.
As shown in step S <b> 6, the control circuit 32 monitors the observation mode switching operation by the observation mode switch 31. When the surgeon desires to observe the running state of the blood vessel of the affected tissue as a subject more clearly, the operator operates the observation mode switch 31 to switch to the NBI observation mode.
If the control circuit 32 does not detect the observation mode switching operation, the control circuit 32 returns to the process of step S3. If the observation mode switching operation is detected, the control circuit 32 starts from the WLI observation mode as shown in step S7. Set to switch to NBI observation mode.

That is, as shown in step S8, the control circuit 32 controls the motor 29 of the light source device 3 so that the narrow band filter group 6 on the inner peripheral side of the rotating color filter 24 is arranged in the illumination optical path. The subject is illuminated with narrowband illumination light from the narrowband filter group 6.
Further, as shown in step S9, the control circuit 32 images the subject illuminated with the narrowband illumination light by the CCD 35, and the synchronization control unit 44 of the video processor 4 performs the narrowband B1, G, B2 imaging signals or An image signal is generated.
Further, as shown in step S10, the luminance calculation unit 66 of the brightness calculation unit 45 calculates the brightness, and the combination ratio calculation unit 66b in the luminance calculation unit 66 calculates the combination ratio a. The calculated composition ratio a is input to the control circuit 32.
As shown in step S11, the control circuit 32 determines whether or not the composition ratio a ′ is set by a setting input from the composition ratio setting unit 58a by the user.

In the case of a determination result in which the composition ratio a ′ is not set, the control circuit 32 determines to adopt the composition ratio a as the composition ratio ac when the composition processing unit 51 performs composition as shown in step S12. (Ac = a).
On the other hand, in the case of the determination result in which the composition ratio a ′ is set, as shown in step S13, the control circuit 32 sets the composition ratio ac when the composition processing unit 51 performs composition (prior to the composition ratio a). ) It is determined to adopt the synthesis ratio a ′ (ac = a ′).
In step S14 after step S12 or S13, the control circuit 32 determines the setting of the enhancement level L. When the enhancement level L is set by the contour enhancement level setting unit 58b or the like, the control circuit 32 confirms the enhancement level L. If the enhancement level L is set, the enhancement level L may be stored in a memory (not shown), the flash memory 60, or the like.

In the next step S15, the gain controller 32a of the control circuit 32 limits the maximum gain value of the AGC circuit 47 with reference to the control information 60a according to the enhancement level L and the synthesis ratio ac. Specifically, the gain control unit 32a constituting the signal level variable control unit is configured to suppress the gain with respect to the imaging signal level input to the AGC circuit 47 as the enhancement level L is larger or the synthesis ratio ac is larger. A control signal Cg is applied to the gain control terminal of the AGC circuit 47.
From the signal that has passed through the AGC circuit 47, an image signal of a composite image is generated by using the composite ratio ac or the like in the composite processing unit 51, and passed through the NR circuit 53, the enlargement / reduction circuit 54, the enhancement circuit 55, the matrix processing unit 56, or the like. Thereafter, the data is output to the color monitor 5. Then, as shown in step S16, the NBI observation image is displayed on the color monitor 5.
By controlling in this way, it is possible to prevent the NBI image displayed on the color monitor 5 when such control is not performed from becoming noisy, and to obtain an easily observable image with no noticeable noise.

In step S17, the control circuit 32 monitors the observation mode switching operation. If the determination result indicates that the observation mode has been switched, the process returns to step S1. On the other hand, in the case of the determination result that the observation mode switching operation has not been performed, in the next step S18, the control circuit 32 determines whether or not the user has performed an operation to end the endoscopy, and the end of the operation is completed. If the determination result indicates that no operation has been performed, the process returns to step S9. If an end operation has been performed, the power is turned off and the process in FIG. 9 ends.
According to this embodiment which operates in this way, the content of the contour enhancement processing by the enhancement circuit 55, the combination of the observation mode (the illumination light by the narrowband filter group 6 corresponding to the observation mode, and the illumination light by the wideband filter group 7). By suppressing (limiting) the maximum value of the gain for amplifying the imaging signal level by the AGC circuit 47 constituting the signal level variable means according to the information of the combination), the image quality is easy to observe without conspicuous noise. An image can be generated.

In particular, in the NBI observation mode in which the amount of illumination light tends to be insufficient, the brightness (from the imaging signal by the combination of the narrow-band G illumination and the B2 illumination and the B1 illumination that precede and follow this G illumination at the same time interval ( Brightness) and a composition ratio for composing the captured images of B2 and B1 according to the calculated brightness can be set, and a composition ratio that a user such as an operator composes in the composition processing unit 51 Priority is given to the manual setting.
The AGC circuit prevents the noise from being noticeable even when the emphasis level L for the edge emphasis processing in the emphasis circuit 55 is set by the user in accordance with the synthesis ratio actually used in the synthesis processing in the synthesis processing unit 51. The maximum value of the gain due to 47 is suppressed so that an image with good image quality can be obtained. Therefore, the surgeon can observe an image with no noticeable noise even in the NBI observation mode, and it is easy to perform a smooth diagnosis.

Supplementally, when the NBI observation mode is set, even if the light amount of the illumination light from the light source device 3 is maximized, the light amount is smaller than that in the WLI observation mode, resulting in a dark image. In the conventional example, since the insufficient amount of light is covered by increasing the gain by the AGC circuit, the gain increases, and the noise increases as the gain increases. Further, when the edge enhancement is performed, an image with more noticeable noise tends to be obtained.
On the other hand, in the present embodiment, in the NBI observation mode, two captured images B1 and B2 are acquired as B captured images in which the light amount tends to be insufficient with respect to the G captured image. : By synthesizing at a synthesis ratio of ac (specifically, synthesizing with B1 + ac · B2), it is possible to obtain a brighter captured image than one case, and AGC according to the synthesis ratio and the enhancement level L By suppressing the gain maximum value of the circuit 47, an image in which noise is not noticeable can be obtained. In the present embodiment, as shown by the dotted line in FIG. 7A, when the composition ratio is large, an image with good image quality can be obtained by suppressing the maximum gain value compared to when the composition ratio is small. It can also be made.

FIG. 10 shows an overall configuration of an imaging system 1B according to a modification of the first embodiment. The imaging system 1B of the present modification is the same as the imaging system 1 in FIG. 1 except that one lens constituting the objective lens system 34 is arranged in the distal end portion 33 of the endoscope 2 as a movable lens 34a, and the optical axis of the objective lens system 34. An endoscope 2B provided with an actuator 81 having a moving portion 81a that moves in the direction, and a video processor 4B provided with an actuator drive circuit 82 that electrically drives the actuator 81 in the video processor 4 are provided.
A movable lens 34a (lens frame) is attached to a moving portion 81a of the actuator 81. Accordingly, when the moving unit 81a is moved by the actuator 81, the movable lens 34a is also moved in the optical axis direction together with the moving unit 81a, and the focus distance at which the objective lens system 34 forms an object image on the imaging surface of the CCD 35 in a focused state is set. Change.

The actuator 81 can be configured using a piezoelectric element as disclosed in, for example, FIG. 3 of Japanese Patent Application Laid-Open No. 2001-174714. Not only the thing using a piezoelectric element but you may comprise by the electromagnetic plunger etc. which have arrange | positioned the movable magnet connected with the spring as a moving part inside the solenoid coil.
Further, for example, the operation unit 12 in the endoscope 2B includes a first focus distance state in which the focus position of the objective lens system 34 is focused on a medium-distance object and a second focus distance in which a short-distance object is focused. A focus changeover switch 31b is provided as an operation switch for performing an instruction operation for switching to the above state. A focus switching signal output when the user operates the focus switching switch 31b is input to the control circuit 32, and the control circuit 32 outputs a control signal corresponding to the focus switching signal to the actuator drive circuit 82.
For example, in the state indicated by the solid line in FIG. 10, the objective lens system 34 is in a first focus distance state (hereinafter referred to as “NORMAL” focus state or simply NORMAL state) for focusing on a medium-distance subject. .

In the NORMAL state, the depth of field of the objective lens system 34 is, for example, about 5 to 100 mm. In this NORMAL state, when a focus switching signal is output, the control circuit 32 outputs a control signal corresponding to this signal to the actuator driving circuit 82, and the actuator driving circuit 82 positions the movable lens 34a at the position indicated by the dotted line. The objective lens system 34 is moved to a second focus distance state (hereinafter referred to as “NEAR FOCUS” focus state or simply NEAR state).
In the NEAR state, the depth of field of the objective lens system 34 is, for example, about 2 to 5 mm, so that the affected part or the like can be magnified and observed in detail at a short distance close to the subject. The control circuit 32 stores the operation information of the focus changeover switch 31b with time in a memory or the like, and grasps whether the objective lens system 34 is set to NEAR or NORMAL in any use state. To do. In this modification, the movable lens 34a is driven to two positions. However, the movable lens 34a may be selectively set to a plurality of three or more positions.

In this modification, the gain control unit 32a of the control circuit 32 further limits the maximum gain value of the AGC circuit 47 according to the state of NEAR and NORMAL.
FIG. 11 shows the contents for limiting the maximum gain value in the NBI observation mode in this modification. The control information shown in FIG. 11 is stored in the flash memory 60 as in the first embodiment.
In the control information of FIG. 11, an item of focus distance (focus state) is added as an item for regulating the maximum gain value in the case shown in FIG. 7A.
When the enhancement level L and the synthesis ratio are the same, the maximum gain value is set (the maximum gain value) in the case of NEAR is larger than that in the case of NORMAL. The reason for this is as follows.

In the case of the NEAR focus distance state, since the short distance is illuminated and observed, the noise is not noticeable (there are few dark parts and no attention even if there are dark parts). For this reason, the maximum gain value is set to be larger in the case of NEAR than in the case of NORMAL.
In other words, in the case of the NEAR focus distance state, it is easier to increase the enhancement level L than when NORMAL (for example, when the composition ratio is the same value). I have to.
When the NBI observation mode is set, it is a state in which capillary blood vessels and the like are desired to be observed favorably, so that the enhancement level L is increased to make it easy to realize optimal observation.
On the other hand, in the case of the NORMAL focus state, compared to the case of the NEAR focus state, dark portions are more likely to occur and noise is more conspicuous, so the enhancement level L is a lower value. It is set as follows. Other configurations are the same as those shown in FIG.

According to this modification, in addition to the effects of the first embodiment, even when the focus distance is changed, noise is not noticeable at the changed focus distance, and the capillaries and the like near the surface layer of the biological mucous membrane are more The state can be easily observed in detail. Therefore, it is not necessary to perform an operation for facilitating observation along with the change of the operator's focus distance, and the operability can be improved.
Further, as described below in this modification, the image displayed on the color monitor 5 is divided into a plurality of regions, and the brightness calculation unit 45 in the case of the NORMAL focus distance state, the brightness in each region. By calculating (average luminance) and lowering the enhancement level L in a dark area where the brightness is low (small), noise may be made inconspicuous.

FIG. 12 shows a state in which an image captured by the CCD 35 (or an image displayed on the color monitor 5) is divided into a plurality of areas Ai (i = 1, 2,..., 16) in the case of a NORMAL focus distance state. Show. In FIG. 12, although divided into 12 areas, it is not limited to this number. In addition, an area setting unit that can selectively set the number of the plurality of areas Ai and the size of the area Ai may be provided so that the user can set the number of the plurality of areas Ai and the size of the area Ai. .
Hereinafter, although the case where it imaged under the Rw, Gw, and Bw illumination light as a some broadband illumination light is demonstrated, it is applicable similarly when imaged under a narrowband illumination light.
In the brightness calculation unit 45 shown in FIG. 4, the adders 63a, 63b, and 63c add the imaging signals Rw-1 and Rw-2, Gw-1 and Gw-2, and Bw-1 and Bw-2, respectively. When calculating the average brightness (average brightness), the adders 63a, 63b, 63c are area brightness calculation circuits as area brightness calculation means for calculating the average brightness in each area Ai shown in FIG. Form. Then, the calculated average brightness in each area Ai is output to the control circuit 32 as indicated by a dotted line, and is also output to the R image, G image, and B image brightness calculation circuits 64a, 64b, and 64c.

The area brightness determination unit 32b indicated by the dotted line in the control circuit 32 compares the average brightness in each area Ai with the threshold value Vt, and determines whether the area is brighter than the threshold value Vt (denoted as a bright part or bright area) or the threshold value Vt. It is determined by a comparison circuit or the like whether the area is less than a dark area (denoted as dark area or dark area). For an area determined to be a dark area less than the threshold value Vt, the area brightness determination unit 32b of the control circuit 32 sets the enhancement level L when performing edge enhancement or structure enhancement in the dark area to the enhancement level L of the bright area. The emphasis level reduction control signal Sout (indicated by a dotted line) that is controlled to be lowered is output to the emphasis circuit 55. In other words, the area brightness determination unit 32b of the control circuit 32 performs control so that the enhancement level L is higher for the image area in the bright area than in the image area in the dark area.
By controlling in this way, the enhancement level L is reduced in the dark region, so that noise in the image when the enhancement level L is not reduced can be prevented from being noticeable. It is possible to generate an image that is clearer and easier to diagnose. Further, when the enhancement level L is controlled according to the bright region and the dark region, the enhancement level L set by the user may be performed, or the enhancement level L specified in advance (for example, a threshold value). You may make it perform with respect to the emphasis level L) used as the emphasis level VL or more.
Note that the threshold value Vt may be set according to the case of the WLI observation mode and the case of the NBI observation mode, respectively, so that noise is not noticeable according to the observation mode.

In addition, in this modification, spatial frequency characteristics in the case where enhancement is performed on the structure portion in each region according to the bright region or the dark region in the region in the case of the WLI observation mode or the NBI observation mode. May be changed.
FIG. 13A shows the contents in this case. For the bright region, when structure enhancement is performed, the strength of structure enhancement is increased on the high frequency side (enhancement level L is increased), and the strength of structure enhancement is decreased on the low frequency side (reduction level L is decreased). )
On the other hand, in the dark region, when structure enhancement is performed, the intensity of structure enhancement on both the high-frequency side and the low-frequency side is reduced (weak).
In this way, for a bright region that is a portion (image region) that the user wants to see, by increasing the enhancement level L of the structure enhancement on the high frequency side, the structure of the bright bright region can be observed in detail. In addition, noise in a dark dark region can be reduced so as not to be noticeable.

In addition, in the case of the WLI observation mode or the NBI observation mode, the structure enhancement is performed according to the electronic zoom (electronic enlargement) by the enlargement / reduction circuit 54 instead of changing the spatial frequency characteristics when the structure enhancement is performed. The emphasis level L may be changed.
FIG. 13B shows the contents in this case. In the case of no electronic zoom, the strength for structure enhancement is increased for the bright region (enhancement level L is increased), and the strength for structure enhancement is decreased for the dark region (the enhancement level L is decreased). Small). In other words, for the bright region, the strength for performing structure enhancement is set larger than that in the dark region.
On the other hand, in the case of the electronic zoom, the strength for performing structure enhancement is reduced (the enhancement level L is reduced) in the bright region and the dark region.
When electronic zoom (electronic enlargement) is performed, noise particles or lumps increase, so that the strength of structure enhancement is reduced so that the noise is not noticeable. Note that the frequency characteristics when the structure is emphasized may be adjusted according to the magnification of the electronic zoom. For example, as the magnification of the electronic zoom is increased, the maximum frequency in the frequency band for structure enhancement may be lowered.
In addition, in embodiment containing the modification mentioned above, embodiment or modification which combined a partially different structure also belongs to this invention.

  In recent years, devices combined with processors have been diversified. There are two types of recording, moving image recording and still image recording. There are various types of still image recording, such as printers, photographs, and electronic files. On the other hand, the number of connectors of the processor is limited, and a simple connector that can be used in any device is required. Therefore, it is desirable to transmit release control with a few contacts such as a stereo pin jack. Since the release control is an example, it is preferable that the control can be arbitrarily selected in the setting to perform the recording control or the capture control. Depending on the device, the signal may be H or L, so the polarity may be selected as H or L. Not only the stereo pin jack but also those having contacts of about 15 pins may be arbitrarily assigned by setting, and depending on the device, the power source and GND (ground) are connected, so it is also possible to select them.

In recent years, an image may be recorded by connecting an external memory to a processor. Some external memories are password protected. Therefore, it is better to ignore the password protection when recording an image and to protect it with a password when reading the image. In recent years, the miniaturization of CCDs has progressed. Accordingly, there has been a problem of pixel defects. For pixel defects, a method has been devised in which an address of a pixel is held in advance and interpolation is performed from pixels around the pixel defect.
However, pixel defects may occur not only during manufacturing but also during long-term use. Therefore, a method for detecting pixel defects has been a problem. By comparing the signal levels of the target pixel and the surrounding pixels, it was considered that if the target pixel was very bright, it was a white scratch, and if it was very dark, it was a black scratch. It is preferable to determine whether the target pixel is a defective pixel based on such a determination criterion.

At that time, the address of the target pixel may be held in the internal memory. However, since the internal memory is limited, the determination may be made only in the central area of the observation screen.
In addition, when the image sensor is increased in size and miniaturized, it is almost impossible to manufacture pixels that have defects in the image sensor or pixels that generate noise.
For this reason, many techniques for correcting noise of defective pixels and fixed patterns that generate pixel signals, which are different from normal pixels of the image sensor, have been proposed by image processing.
As a defective pixel, it is known that the potential of a signal after photoelectric conversion is not desired, and becomes a signal including a potential of a certain offset. If the offset potential is a positive value, it appears on the observation screen as white dots called white scratches, and if the offset potential is a negative value, it appears on the observation screen as black dots called black scratches. The image processing for correcting the defective pixel can estimate and remove the offset potential. Further, the position information of the defective pixel is stored in the endoscope, and the operation is performed so as to estimate the offset potential with respect to the position information.

In order to estimate the potential of the offset, an average value is calculated using as many pixel signals as possible from the 8 pixels around the defective pixel, and then the degree of deviation of the defective pixel from the average value is estimated. I think it is possible. However, some image sensors have a multi-line type and need to be considered as the surrounding 8 pixels. For example, in the case of a two-wire system, one line including a defective pixel and the upper and lower lines are different channels. If the former is called the A channel and the latter is called the B channel, it is common to form an image by arranging the A channel and the B channel in order up and down by image processing different from correction.
The A channel and the B channel indicate that the channels for transmitting signals from the image sensor are separate. In such a case, the sensitivity may change due to variations in each channel. That is, even if the same amount of light is incident on the A channel and the B channel, the signals obtained are different. In the image processing described above, there is a method of obtaining and multiplying necessary coefficients so that the variation for each channel can be canceled. Specifically, this is achieved by acquiring white balance. Alternatively, necessary coefficients may be stored in advance in a memory (scope ID) provided in the endoscope.
Therefore, it is desirable to perform the image processing to be corrected after canceling the variation for each channel. If image processing to be corrected is applied before canceling the variation for each channel, the following is desirable.

If attention is paid to one defective pixel and its surrounding eight pixels, it becomes three lines as shown in FIG. 14. For example, the A channel is located at the top, the B channel, and the A channel at the bottom. Become. The defective pixel is in the B channel. A case where the sensitivity of the A channel is smaller than that of the B channel, specifically, a case where the sensitivity of the B channel is 1.0 and the sensitivity of the A channel is 0.9 will be described.
When the defective pixel is a white dot, the potential of the offset which is a positive value is calculated and subtracted, so that the defective pixel does not become a white dot and can be improved to an appropriate one. However, since there is a difference in sensitivity for each channel, the offset potential is estimated to be large, and the defective pixel is not corrected properly, resulting in a black dot. This is because the sensitivity of the A channel is smaller than the sensitivity of the B channel, and the average value of the 8 pixels around the defective pixel becomes small. Similarly, if the defective pixel is a black dot, I would like to correct the offset potential, which is a negative value, and subtract it to correct the defective pixel so that it does not become a black dot. Pixels are black dots.
Even when the defective pixel is a white dot or a black dot, if the sensitivity of the A channel is larger than the sensitivity of the B channel, the correction is not appropriate, and a white dot is obtained.

In order to improve this phenomenon, it is necessary to detect whether the pixel is a white dot or a black dot after correcting the defective pixel. The detection method is assumed to be detected if the corrected defective pixel has a difference equal to or greater than a threshold with respect to the average value of the two pixels on the left and right of the defective pixel. For white dots, the threshold value is subtracted, and for black dots, the threshold value is added, so that it can be corrected so that it is no longer visible as a white or black dot.
What to do when the position information of the defective pixel is not stored in the endoscope in advance, or when the position information is not stored when a new defective pixel occurs even if it is stored in advance It is as follows.
One proposal is to detect defective pixels from an image signal obtained from an endoscope. Specific methods have been proposed so far. In this way, if a defective pixel can be detected, position information can be obtained, and processing may be performed so that it is not visually recognized as white or black dots, as described above. In the above description, the threshold value is added and subtracted. However, it may be replaced with the average value of the two left and right pixels. For example, in the case of processing from position information stored in an endoscope, a threshold value is added or subtracted, and in the case of processing from detected position information, the average value of two left and right pixels may be replaced. .

As another proposal, there is a method in which a defective pixel is detected from an image signal obtained from an endoscope, and then position information is obtained and stored in the endoscope. As a feature, it is desirable to carry out after using a predetermined subject and preparing conditions for easily detecting defective pixels. To detect white dots, the illumination light is extinguished and the tip of the endoscope is covered with a dark screen, and then the white dots appear stably by operating for about 10 minutes. It is only necessary to compare and detect if there is a difference greater than or equal to the threshold.
When detected, the position information of the pixels is stored in the endoscope, so that white dots are improved when the endoscope is used. However, unintended conditions must be taken into account. For example, if it cannot be covered with a black curtain, the operation is not correct. Therefore, error determination is important. Although it is an example of error determination, when the position information exceeds a predetermined number, a condition to be confirmed may be notified while notifying as an error. This is because it can be assumed that there are at most about 20 white dots to be detected by this method because there are many publications on the reason why defective pixels occur in the CCD. This method can be applied to a frame sequential type using a monochrome CCD or a simultaneous type using a color chip CCD. Since the method described here is based on the frame sequential method, in order to apply it simultaneously, it is necessary to consider the relationship between the color chip and the pixel so that the white dot is a dot such as blue.

In the case of the frame sequential type as an example, it may be necessary to consider the video system and the signal processing of the CCD. Examples of the video system include NTSC and PAL. The former displays 60 frames (60 Field) images per second to reproduce a moving image, and the latter displays 50 frames of images to reproduce a moving image. On the other hand, the signal processing of the CCD may be 60 frames per second like NTSC. In order to apply to PAL, it is necessary to rate convert 60 frames of images to 50 frames. This process is called 65 conversion. An image memory is used for the 65 conversion, and in the case of the frame sequential type, a synchronization memory is used to obtain a synchronized color image.
The synchronized memory may have a pre-freeze function for obtaining a still image. As shown in FIG. 15, an image memory is used to perform 65 conversion, and 60 frames of images input to the image memory are rate converted to 50 frames and output. For example, in the image memory, as shown in the upper part of FIG. 15, the frame sequential RGB signals of RO, GO, BO, R1, G1, B1,... , C65I_B_VALA is sequentially input and written (some need not be written as described below). On the other hand, the frame sequential RGB images of RO, GO, BO, R1, G1, B1,... Written in the image memory are read at 50 frames / sec (50 Field / sec) as shown in the lower part of FIG. (However, images of RO, R2, etc. are through output at the timing at which 60 frames / second and 50 frames / second are synchronized) and are converted into 65.
Also, the synchronization memory shown in FIG. 16 has a memory configuration for three for RGB, for example, and writes the G0 image after the rate conversion to the G synchronization memory to obtain an R0 image of 60 frames / second. Is written to the R synchronous memory. At this time, since the image format of the R0 image is different from that of the G0 image in terms of horizontal and vertical frequencies, it is desirable to use a line memory or an internal memory of the FPGA and to deal with it at low cost. In this way, 65 conversion and synchronization can be realized at low cost. Since the synchronization memory has a pre-freeze function, a still image with little color misregistration can be provided.

Further, the present invention includes the contents of the following supplementary notes that are slightly different from the claims.
Appendix 1.
An observation mode switching means for switching between the two observation modes;
Depending on the switching of the observation mode switching means, the illumination light of the first illumination band having the first narrow band and the second narrow band or the second illumination having a plurality of wide bands in each frame period Illumination means for illuminating with a band of illumination light;
Imaging means for imaging a subject illuminated by the illumination means;
Signal level variable means for changing the imaging signal level of the imaging signal output by the imaging means;
Image processing means for generating an image signal from the signal output from the signal level varying means, and performing contour enhancement processing on the generated image signal;
When illuminating with illumination light in the first illumination band, two captured images captured under the second narrow-band illumination light at a timing before and after the first narrow-band illumination light. A synthesis processing means for performing a process of combining;
The outline emphasis processing content performed by the image processing means, the combination of the first narrow band and the second narrow band forming the first illumination band in the illumination means, and the second illumination band A signal for controlling the imaging signal level by the signal level varying means according to information on the combination of the plurality of widebands to be formed and a synthesis coefficient (synthesis ratio) when the two captured images are synthesized by the synthesis processing means. Level variable control means;
An imaging system comprising:

Appendix 2. In Supplementary Note 1, the illumination in the first illumination band includes the first illumination by the illumination light in the first narrow band and the second narrow band within a predetermined time that is an integral multiple of one frame period. It consists of a combination with the second illumination which is twice the first illumination with illumination light having a band,
Further, a first imaging signal by the imaging means based on the first predetermined number of times of the first number of times and a second predetermined number that is the same as the first predetermined number of times of the second number of times. Calculating the first brightness by color conversion matrix processing using the second imaging signal by the imaging means based on the illumination of the times,
A third imaging signal based on the third predetermined number of illuminations that is the same as the first predetermined number of times in the first number of times, and a fourth same as the second predetermined number of times in the second number of times. Brightness calculation means for calculating the second brightness by color conversion matrix processing using a fourth imaging signal based on a predetermined number of illuminations;
Based on the difference between the first brightness and a predetermined target brightness, and the ratio between the second brightness and the third and fourth image signals for the first and second imaging signals in each frame period. Computing means for calculating a synthesis coefficient when synthesizing the imaging signals of
With
The image processing means is configured to output the third and third image signals for the first and second imaging signals in each one frame period based on a synthesis coefficient by the computing means or a synthesis coefficient set by a synthesis coefficient setting means. 4 image pickup signals are combined, and the generated image signal is subjected to edge enhancement processing,
In the case where the signal level varying means is an imaging signal imaged under illumination light of the first illumination band consisting of a combination of the first narrow band and the second narrow band, The signal level varying means for processing the first to fourth imaging signals is controlled based on the outline emphasis processing content performed by the image processing means and the synthesis coefficient.

Appendix 3. In Appendix 2,
The first imaging signal and the third imaging signal are common imaging signals.
Appendix 4. In Appendix 2 or Appendix 3,
The imaging means includes an objective lens system that forms an optical image of a subject, and
An imaging device in which an imaging surface is arranged at a position where the optical image is formed, and photoelectrically converts the optical image;
An actuator for driving the movable lens constituting the objective lens system to at least two positions, ie, a first position and a second position in the optical axis direction of the objective lens system;
With
Furthermore, the imaging system sets the movable lens to the first position by the actuator, whereby the objective lens system is in a state of a first focus distance focused on a close object, and the actuator An operation switch for performing an instruction operation for switching the state of the second focus distance in which the objective lens system focuses on a medium-distance object by setting the movable lens to the second position;
When the objective lens system is set to the second focus distance, the signal level varying unit suppresses the gain with respect to the imaging signal level more than when the objective lens system is set to the first focus distance.

  DESCRIPTION OF SYMBOLS 1 ... Imaging system, 2 ... Endoscope, 3 ... Light source device, 4 ... Video processor, 5 ... Color monitor, 6 ... 1st filter group (narrowband filter group), 7 ... 2nd filter group (wideband filter group) DESCRIPTION OF SYMBOLS 11 ... Insertion part, 14 ... Light guide, 21 ... Lamp, 22A ... Diaphragm, 24 ... Rotation filter, 31 ... Observation mode switch, 32 ... Control circuit, 32a ... Gain control part, 35 ... CCD, 44 ... Synchronization control 45 ... Brightness calculation unit 46 ... Synchronization memory 47 ... AGC circuit 50 ... Image processing unit 51 ... Composition processing unit 53 ... NR circuit 54 ... Enlargement / reduction circuit 55 ... Enhancement circuit 56 ... Matrix processing unit 58 ... Operation panel 58a ... Combination ratio setting unit 59 ... Dimming control unit 60 ... Flash memory 60a ... Control information 62 ... Average value calculation unit 64a, 64b, 64c ... Brightness Calculator, 65 ... matrix processing unit, 66 ... luminance calculation unit, 66a ... difference calculation unit, 66b ... synthesis ratio calculation unit

Claims (6)

Illuminating means for performing illumination by switching between illumination light of a first illumination band having a first narrow band and a second narrow band and illumination light of a second illumination band having a plurality of wide bands;
Imaging means for imaging a subject illuminated by the illumination means;
Signal level variable means for changing the imaging signal level of the imaging signal output by the imaging means;
Image processing means for generating an image signal from the signal output from the signal level varying means, and performing contour enhancement processing on the generated image signal;
The outline emphasis processing content performed by the image processing unit, the combination of the first narrow band and the second narrow band forming the first illumination band in the lighting unit, and the second illumination band are formed. Signal level variable control means for controlling the imaging signal level by the signal level variable means in accordance with information of the plurality of broadband combinations;
An imaging system comprising: The illumination in the first illumination band includes a first illumination with the first narrow-band illumination light and a second illumination with the illumination light having the second narrow band within a predetermined time. A combination of
Further, the first imaging signal by the imaging unit based on the first predetermined number of times of the first number of times and the imaging unit based on the second predetermined number of times of illumination of the second number of times. The first brightness is calculated by color conversion matrix processing using the second imaging signal by
The first imaging signal based on the third predetermined number of times of illumination in the first number of times and the third imaging signal based on the fourth predetermined number of times of illumination in the second number of times are used. Brightness calculation means for calculating the second brightness by color conversion matrix processing;
Based on the difference between the first brightness and a predetermined target brightness and the ratio of the second brightness, the first and third imaging signals for the first and second imaging signals in each one frame period. Computing means for calculating a synthesis coefficient when synthesizing the imaging signals of
With
The image processing unit is configured to perform the first and second imaging signals on the first and second imaging signals in each frame period based on a synthesis coefficient by the calculation unit or a synthesis coefficient set by a synthesis coefficient setting unit. 3 image pickup signals are combined, and contour enhancement processing is performed on the generated image signals.
In the case where the signal level varying means is an imaging signal imaged under illumination light of the first illumination band consisting of a combination of the first narrow band and the second narrow band, 2. The signal level varying means for processing the first, second, and third imaging signals is controlled based on the content of the edge enhancement process performed by the image processing means and the synthesis coefficient. The imaging system described.   The signal level varying unit suppresses the gain with respect to the imaging signal level as the enhancement level of the edge enhancement process is larger or as a synthesis ratio based on the synthesis coefficient is larger. Imaging system. The imaging means includes an objective lens system that forms an optical image of a subject, and
An imaging device in which an imaging surface is arranged at a position where the optical image is formed, and photoelectrically converts the optical image;
An actuator for driving the movable lens constituting the objective lens system to at least two positions, ie, a first position and a second position in the optical axis direction of the objective lens system;
With
Furthermore, the imaging system sets the movable lens to the first position by the actuator, whereby the objective lens system is in a state of a first focus distance focused on a close object, and the actuator An operation switch for performing an instruction operation for switching the state of the second focus distance in which the objective lens system focuses on a medium-distance object by setting the movable lens to the second position;
When the objective lens system is set to the second focus distance, the signal level varying unit suppresses the gain with respect to the imaging signal level more than when the objective lens system is set to the first focus distance. The imaging system according to claim 3. Furthermore, area brightness calculation means for calculating an average brightness in each of a plurality of areas obtained by dividing the captured image generated from the first imaging signal and the second imaging signal;
A determination unit that determines a bright region in which the average brightness calculated by the region brightness calculation unit is equal to or greater than a threshold value and a dark region that is less than the threshold value;
The imaging system according to claim 3, wherein in the bright region determined by the determination unit, control is performed so that the enhancement level of the contour enhancement processing is higher than that in the dark region.   6. The first narrow band is a narrow band in a green wavelength band, and the second narrow band is a narrow band in a blue wavelength band. The endoscope apparatus described in 1.

Images