JP6120722B2 - Observation device and method of operating observation device - Google Patents

Description

The present invention relates to an observation apparatus such as an endoscope and a method for operating the observation apparatus .

  Currently, in an observation apparatus such as an endoscope, light emitted from a light source is guided by a light guide member such as an optical fiber, and a desired irradiation pattern is provided by a wavelength conversion member provided at an emission side end of the light guide member. A light-emitting device that emits light after converting it into a color is used as a lighting device. In such a light emitting device, since the light guide member is a member having low mechanical strength, the light guide member may be cut (disconnected) for some reason, and light may leak from the cut surface. In particular, when the light emitted from the light source is laser light, there is a risk that the leaked laser light may pose a danger to the human body or the like. Therefore, a mechanism for detecting such disconnection of the light guide member is required.

  For example, in the light emitting device disclosed in Patent Document 1, the light returned to the light source side is detected by the light receiving element among the light converted in wavelength by the wavelength conversion member, and the light guide member is normal when the return light can be detected. However, when the return light is no longer detected, the light guide member is detected as being disconnected.

JP 2008-122838 A

  When the above light emitting device is used as an illumination device for an observation device such as an endoscope, in the configuration as in Patent Document 1, it is necessary to use a member such as a light receiving element for detecting disconnection of the light guide member. This makes it difficult to reduce the size of the observation apparatus.

Therefore, an object of the present invention is to provide an observation apparatus and an operation method of the observation apparatus that can detect a failure of an illumination apparatus such as a disconnection of a light guide member without increasing the size of the apparatus.

  In order to achieve the above object, an aspect of the observation apparatus of the present invention includes a light source that emits primary light, a light guide member that guides the primary light emitted from the light source, and the light guide member. And receiving at least one light conversion member that receives the primary light guided by the light source, converts the light into secondary light having an optical characteristic different from that of the primary light, and emits the secondary light. An illumination device that irradiates the observation target portion with emitted light as illumination light, an image acquisition unit that captures the observation target portion and repeatedly acquires an observation target image composed of a plurality of pixels, and the inside of the observation target image A failure determination region setting unit for setting a failure determination region used for failure determination of the lighting device, a representative value extraction unit for extracting a representative value indicating a state of a pixel value in the failure determination region, and a change in the representative value Based on the above, Comprising a failure determination unit.

Further, according to one aspect of the operation method of the observation apparatus of the present invention, a light source that emits primary light, a light guide member that guides the primary light emitted from the light source, and light guide by the light guide member. And receiving at least one light converting member that receives the primary light and converts it into secondary light having an optical characteristic different from that of the primary light, and emits the light from the light converting member. An operation method of an observation apparatus including an illuminating device that irradiates an observation target portion with light as illumination light, wherein the observation device images the observation target portion and repeats an observation target image including a plurality of pixels. acquired, the observation apparatus sets the failure determination area used for failure determination of the illumination device to the observation target image, the observation apparatus extracts a representative value indicating the state of pixel values in the failure determination area , the observation apparatus, the change of the representative value It intends line failure determination of the illumination device Zui.

The observation device and the operation method of the observation device of the present invention can detect a failure of the illumination device such as disconnection of the light guide member without using the image acquired by the image acquisition unit without increasing the size of the device. .

FIG. 1 is a block diagram showing the configuration of the observation apparatus according to the first embodiment of the present invention. FIG. 2 is a schematic view of the configuration of the light conversion member in FIG. FIG. 3 is a diagram showing the spectral sensitivity characteristics of a known CCD imager. FIG. 4 is a schematic diagram for explaining the operation of the failure determination area setting unit. FIG. 5 is a schematic diagram for explaining the operation of the failure determination unit. FIG. 6 is a diagram illustrating an operation flowchart of the failure determination unit. FIG. 7 is a schematic diagram for explaining the movement of the center of gravity of the distribution of red pixel values on the image. FIG. 8 is a block diagram showing a configuration of an observation apparatus according to the third embodiment of the present invention. FIG. 9 is a diagram for explaining a failure determination area setting interval and a failure determination interval. FIG. 10 is a diagram for explaining a failure determination area setting method.

Hereinafter, embodiments will be described with reference to the drawings.
[First Embodiment]
As shown in FIG. 1, the observation apparatus 1 according to the present embodiment is for observing an observation target portion OE that does not receive external light that is inside the object OJ. The acquisition unit 3, the image display unit 4, the failure determination area setting unit 5, the representative value extraction unit 6, and the failure determination unit 7 are included. The illumination device 2 irradiates illumination light toward the observation target part OE inside the object OJ. The image acquisition unit 3 receives the reflected light from the observation target unit OE and acquires an image (observation target image) of the observation target unit OE. The observation target image acquired by the image acquisition unit 3 is input to the image display unit 4 such as a CRT or a liquid crystal display and displayed. The observation target image is also input to the failure determination area setting unit 5. The failure determination region setting unit 5 sets a failure determination region used for failure determination of the lighting device 2 in the observation target image, and the representative value extraction unit 6 uses the pixel value of the set failure determination region as failure determination region information. To enter. The representative value extraction unit 6 extracts a representative value indicating the state of the pixel value in the failure determination region from the failure determination region information, and inputs the extracted representative value to the failure determination unit 7. The failure determination unit 7 performs failure determination of the lighting device 2 based on the change in the representative value. When it is determined that there is a failure, the failure determination unit 7 inputs a failure determination signal indicating the failure to the lighting device 2. And a failure determination signal is also input to the image display unit 4 to notify the user of the occurrence of the failure.

Details of each part will be described below.
The illuminating device 2 includes a light source 21 that emits primary light, a light conversion member 22 that converts incident primary light into secondary light having different optical characteristics, and emits the light, and light source control that controls the light source 21. And a light guide member that guides the primary light emitted from the light source 21 to the light conversion member 22. The light guide member includes optical fibers 24a, 24b, and 24c, an optical coupler 25, and optical connectors 26a, 26b, and 26c.

  The light source 21 is, for example, a laser diode (LD) that emits laser light as primary light. Specifically, it is a blue LD having a center wavelength of 445 nm and a half-value width on the order of nm.

  The light source control unit 23 is electrically connected to the light source 21 and inputs an LD control signal for controlling ON / OFF of the light source 21, a drive current, a drive voltage, and a drive method to the light source 21. Here, the driving method is, for example, pulse driving or continuous driving (CW). Further, since the light source 21 generates laser light with a light amount corresponding to the input drive current value, the LD light amount information indicating the laser light amount can be input from the light source control unit 23 to the representative value extracting unit 6. Although the details will be described later, the representative value extraction unit 6 also uses this LD light quantity information when extracting a representative value indicating the state of the pixel value in the failure determination area. The light source control unit 23 controls to stop the light source 21 when a failure determination signal is received from the failure determination unit 7.

  Although at least one light conversion member 22 is sufficient, the lighting device 2 in the present embodiment uses two light conversion members 22. Therefore, the light source 21 and the two light conversion members 22 are optically connected via optical fibers 24a, 24b, and 24c, an optical coupler 25, and optical connectors 26a, 26b, and 26c as light guide members. Has been. That is, the optical coupler 25 is an optical branching coupler of 1 input port × 2 output port with a branching ratio of 1: 1, for example, and the input port and one end of the optical fiber 24a are connected by the optical connector 26a, and two outputs are connected. One of the ports and one end of the optical fiber 24b are connected by an optical connector 26b, and the other output port and one end of the optical fiber 24c are connected by an optical connector 26c. The other end of the optical fiber 24 a is connected to the light source 21, the other end of the optical fiber 24 b is connected to one of the two light conversion members 22, and the other end of the optical fiber 24 c is connected to the other end of the light conversion member 22. Has been. Although not shown, a coupling lens for converging the light from the light source 21 and coupling it to the optical fiber 24a is provided between the light source 21 and the optical fiber 24a.

  The optical fibers 24a, 24b, and 24c are, for example, multimode single-wire fibers having a core diameter of 50 μm and a numerical aperture FNA = 0.2. The optical connectors 26a, 26b, and 26c couple the optical fibers 24a, 24b, and 24c and the ports of the optical coupler 25 so that their optical axes coincide with each other, and are, for example, FC connectors / adapters. .

  By adopting such a configuration, the light source 21 having a relatively large size is arranged outside the object OJ, and relatively thin optical fibers 24b and 24c and relatively relatively are inserted in the insertion portion inserted into the object OJ. It is only necessary to provide the light conversion member 22 having a small size, and the diameter of the insertion portion can be reduced. Therefore, even if the observation target part OE is a thin and deep part inside the object OJ, it is possible to illuminate.

  The light conversion member 22 is disposed at the distal end portion of such an insertion portion. In the following description, the side opposite to the tip is referred to as the base end. The two light conversion members 22 have the same configuration and have substantially the same characteristics. Specifically, as shown in FIG. 2, the light conversion member 22 includes a holder 221, a light transmission member 222, a phosphor 223, and a reflection member 224.

  The holder 221 has a frustoconical hole formed therein so that its diameter increases toward the tip side. In the hole of the holder 221, the side having a smaller diameter on the proximal end side is referred to as an incident port, and the side having a larger diameter on the distal end side is referred to as an exit port. The other ends of the optical fibers 24 b and 24 c are connected to the entrance of the holder 221. Therefore, the laser light guided through the optical fibers 24 b and 24 c is emitted from the exit end which is the other end, and enters the hole inside the holder from the entrance side of the holder 221.

  A light transmitting member 222 and a phosphor 223 are provided in the hole of the holder 221, and a reflecting member 224 is provided on the inner surface of the hole that is an interface between them. Here, the light transmission member 222 is provided on the entrance side, and the phosphor 223 is provided on the exit side. The light transmitting member 222 is formed of a material having a high light transmittance, such as silicone resin or glass. Therefore, the laser light guided through the optical fibers 24 b and 24 c passes through the light transmitting member 222 and enters the phosphor 223. The phosphor 223 is, for example, a Ce (cerium) activated powder YAG (Yttrium Aluminum Garnet) phosphor. The phosphor 223 is dispersed and mixed in a sealing material such as silicone resin or glass. The phosphor 223 is mixed into the sealing material in a predetermined amount in order to convert it into desired excitation light absorption rate and wavelength conversion efficiency characteristics as illumination light. The phosphor 223 receives the primary light and converts it into secondary light. For example, the phosphor 223 receives blue laser light and converts it into yellow fluorescence. Therefore, the fluorescence converted by the phosphor 223 is emitted from the exit of the holder 221. The reflection member 224 is, for example, a silver or aluminum foil, and reflects light (for example, laser light and fluorescence) inside the holder 221 toward the exit. That is, since the fluorescence converted by the phosphor 223 is emitted in all directions, the reflection member 224 reflects the fluorescence toward the exit.

  As described above, the two light conversion members 22 are connected to the light source 21 by the light guide member. Therefore, the presence or absence of secondary light (fluorescence in this example) from the two light conversion members 22 synchronizes with the turning on / off of the light source 21. That is, the two light conversion members 22 emit secondary light when the light source 21 is turned on, and stop emitting the secondary light when the light source 21 is turned off. Further, not all of the laser light incident on the light conversion member 22 is incident on the phosphor 223, and a part of the laser light is emitted from the exit. The illumination light that the illumination device 2 irradiates the observation target part OE includes laser light and fluorescence.

  As shown in FIG. 1, the image acquisition unit 3 receives reflected light of illumination light emitted from the illumination device 2 to the observation target unit OE and acquires an image (observation target image) of the observation target unit OE. The imaging unit 31 and the image processing unit 32 are included. The imaging unit 31 is an imaging device, such as a CCD imager or a CMOS imager. In the present embodiment, the imaging unit 31 is described as a CCD imager. The image acquired by the image acquisition unit 3 is also referred to as a frame, and the image acquisition unit 3 repeatedly acquires the frame and visualizes it. The speed at which the image acquisition unit 3 acquires a frame that is an image (image acquisition interval) is referred to as a frame rate. The frame rate is, for example, about 30 fps in general image acquisition.

  FIG. 3 is a diagram showing the spectral sensitivity characteristics of a known CCD imager. The longer wavelength side is referred to as the long wavelength side, and the shorter wavelength side is referred to as the short wavelength side. In the same figure, the light reception sensitivity characteristic of the blue pixel with respect to the wavelength λ is represented as b (λ), the light reception sensitivity characteristic of the green pixel is represented as g (λ), and the light reception sensitivity characteristic of the red pixel is represented as r (λ).

  The imaging unit 31 includes a blue pixel, a green pixel, and a red pixel, and has a sensitivity peak in each color region. For example, in a known CCD imager, a blue pixel has a sensitivity peak at a wavelength of 460 nm (λb) in the blue region. Similarly, the green pixel has a sensitivity peak at a wavelength of 540 nm (λg) in the green region, and the red pixel has a sensitivity peak at a wavelength of 630 nm (λr) in the red region.

  In the known CCD imager, as shown in the figure, the sensitivity region of the blue pixel exists up to the wavelength of 540 nm on the long wavelength side. The sensitivity region of the red pixel exists up to a wavelength of 540 nm on the short wavelength side. Therefore, the blue pixel, the green pixel, the green pixel, and the red pixel have a wavelength region in which the sensitivity overlaps in the adjacent wavelength region. A value representing the brightness of each pixel is referred to as a pixel value. In the imaging unit 31, the pixel value of each color takes a value from 0 to 255.

  The image processing unit 32 has a function of performing predetermined image processing on the pixel signals of the blue pixel, the green pixel, and the red pixel received by the imaging unit 31 to generate an image.

  It should be noted that at least the imaging unit 31 of the image acquisition unit 3 may be disposed at the distal end of the insertion unit inserted into the object OJ. In addition, although the imaging part 31 is normally arrange | positioned so that the front-end | tip direction may be imaged at a front-end | tip part, it can also be arrange | positioned so that the side surface direction of an insertion part front-end | tip may be imaged. In this case, the light conversion member 22 may be arranged so that the exit of the light conversion member 22 faces the side surface direction of the insertion portion tip so that the illumination light can illuminate the side surface direction of the insertion portion tip.

  As shown in FIG. 1, the failure determination area setting unit 5 includes an area storage unit 51 for storing an attention area described later. As described above, the failure determination region setting unit 5 sets a failure determination region used for failure determination of the lighting device 2 in the observation target image. In order to set the failure determination region, the region storage unit The attention area stored in 51 is used.

  Specifically, the failure determination area setting unit 5 receives an observation target image from the image acquisition unit 3. As described above, the observation target image includes information on the pixel values of the blue pixel, the green pixel, and the red pixel. The failure determination area setting unit 5 divides the observation target image into a plurality of areas, calculates a focus evaluation value using the contrast of a pixel of a predetermined color, for example, a blue pixel value in each area, and calculates the calculated alignment. Focus determination is performed from the focus evaluation value. Then, the in-focus area determined to be in focus is set as the attention area, and is stored in the area storage unit 51 together with the pixel value information.

  In the present embodiment, a contrast method is used in this focus determination process. The contrast is an index representing light and dark on the image. Generally, the degree of focus and the contrast are correlated. The contrast method uses the fact that the contrast of the image is maximized at the in-focus position. The magnitude of the contrast is evaluated by a high frequency component at the spatial frequency of the pixel value. For example, a high-frequency component of the pixel value is extracted by a band pass filter. The focus evaluation value is obtained by integrating the absolute values of the high frequency components extracted in this way. Therefore, the in-focus evaluation value becomes the maximum when the in-focus state becomes the maximum after the in-focus state.

  For example, when the object OJ is a human body, blue light is absorbed by hemoglobin in blood vessels existing on the body cavity surface of the object OJ in the illumination light emitted from the light conversion member 22 of the illumination device 2. It is easy to be done. Therefore, when a blood vessel is present in the observation target part OE, the contrast between the blood vessel part on the surface of the body cavity of the object OJ and the other part is enhanced by using the blue pixel value for calculating the contrast.

  FIG. 4 is a schematic diagram of focus determination regarding the observation target image 33. The black portion 34 in the observation target image 33 is, for example, a lumen that exists in the body cavity. The lumen portion is a portion where the amount of reflected light sufficient for imaging cannot be acquired because the distance to the bottom portion is long.

  The failure determination area setting unit 5 receives an observation target image 33 as shown on the left side in the figure from the image acquisition unit 3. Next, as shown in the middle of the figure, the failure determination area setting unit 5 divides the observation target image 33 into a plurality of divided areas 35. Then, as shown on the right side in the figure, of the plurality of divided areas 35, an area where the observation target portion OE exists at a predetermined distance is set as the attention area 36, and is stored in the area storage section 51 together with the pixel value information. To do. Here, the predetermined distance is a distance at which a sufficient reflected light amount can be obtained for detecting a change in a representative value, which will be described later, indicating the state of the pixel value at the time of failure.

  That is, the failure determination area setting unit 5 sets an area having a focus evaluation value equal to or higher than a predetermined focus threshold (high contrast) among the plurality of divided areas 35 as the attention area 36. Here, the predetermined focus threshold value is based on the focus evaluation value when the subject is at a short distance where the reflected light is strong and it is difficult to detect the change in the representative value and at a long distance where the reflected light is weak and the change in the representative value is difficult to detect. Large value. Therefore, the attention area 36 of the present embodiment is an area excluding an area where there is a lumen, for example, where sufficient reflected light cannot be obtained to detect a change in the representative value at the time of failure.

  When the attention area is not detected in the entire area of the observation target image 33, the failure determination area setting unit 5 stores the attention area 36 of the observation target image (previous frame) at the previous imaging stored in the area storage unit 51. Is set as the attention area 36 of the observation target image (current frame) at the time of imaging. The set attention area 36 and its pixel value information are stored in the area storage unit 51 as the attention area 36 and its pixel value information of the observation target image (current frame) at the time of imaging.

  In the present embodiment, the failure determination region setting unit 5 sets the attention region 36 stored in the region storage unit 51 as a failure determination region, and the pixel value information of the failure determination region is used as the failure determination region information as a representative value extraction unit. 6 is output.

  The representative value extraction unit 6 represents the pixel value state from the failure determination region information received from the failure determination region setting unit 5 and the LD light amount information received from the light source control unit 23 indicating the light amount at the time of obtaining the observation target image. Extract the value. In place of the LD light amount information from the light source control unit 23, the light source control unit 23 gives an LD control signal for controlling the light source 21 to the representative value extraction unit 6, and the representative value extraction unit 6 obtains an observation target image. You may make it extract the laser light quantity at the time. Specifically, for example, the relationship between the drive current and drive voltage sent as the LD control signal from the light source control unit 23 to the light source 21 and the laser light quantity at the time of obtaining the observation target image is tabulated in advance. Then, the representative value extraction unit 6 associates the information on the drive current and the drive voltage in the LD control signal sent from the light source control unit 23 with this table as the amount of laser light when the observation target image is acquired. get.

  For example, the representative value extraction unit 6 first extracts an average value of red pixel values in the failure determination region from the failure determination region information received from the failure determination region setting unit 5, that is, the pixel value information of the failure determination region. It is known that red light is less susceptible to absorption by blood vessels than blue light and green light. Therefore, by using only the red pixel value as the representative value, for example, when the amount of blood vessels in the failure determination region changes due to the movement of the observation device 1 (insertion unit) within the failure determination interval, the change in the representative value is caused. Get smaller.

  Then, the representative value extracting unit 6 calculates “ratio of“ average value of red pixel values in the failure determination region ”to“ laser light amount at the time of observation target image acquisition ””, and extracts this ratio as a representative value. The interval at which the extraction unit 6 extracts the representative value corresponds to the interval at which failure determination is performed by a failure determination unit 7 to be described later, and the representative value extraction unit 6 outputs the extracted representative value to the failure determination unit 7.

As shown in FIG. 1, the failure determination unit 7 includes a representative value storage unit 71 for storing the representative value received from the representative value extraction unit 6, and the previous frame ( Failure determination is performed by comparing the failure determination criterion set based on the representative value at time t _n-1 ) with the representative value of the current frame (time t _n ) received from the representative value extraction unit 6. . Details of the failure criterion will be described later. The failure determination unit 7 repeatedly performs such failure determination at failure determination intervals. This failure determination interval is t _n −t _n−1 and corresponds to the frame rate.

Hereinafter, a failure determination method in the failure determination unit 7 will be described with reference to FIG. When the representative value of the current frame is larger than the failure determination criterion, the failure determination unit 7 determines that the lighting device 2 has not failed (t ₁ to t _{5 in} FIG. 8). In such a case, the failure determination unit 7 uses the representative value in the current frame (time t _n ) received from the representative value extraction unit 6 for the failure determination process in the next frame. Store to 71. After storing the representative value at time t _{n in} the representative value storage unit 71, the failure determination unit 7 continues to perform failure determination on the representative value of the next frame (time t _{n + 1} ). On the other hand, when the representative value of the current frame is smaller than the failure determination criterion, the failure determination unit 7 determines that the lighting device 2 has failed (t _{6 in} FIG. 8). When it is determined that there is a failure as described above, the failure determination unit 7 outputs a failure determination signal for stopping the light source 21 to the light source control unit 23. Although not particularly illustrated, the failure determination unit 7 further includes a latch circuit so that the output of the failure determination signal to the light source control unit 23 can be maintained. The failure determination signal is also output to the image display unit 4. Therefore, the user can confirm the failure of the lighting device 2 by the image displayed on the image display unit 4.

Failure criteria at time t _n is the representative value at time t _n-1, is set as the value obtained by multiplying a predetermined criterion constant alpha. The determination reference constant α is determined from “a change rate of a representative value within a failure determination interval in a state where the lighting apparatus is operating normally” and “a change rate of a representative value at the time of failure”. For example, in FIG. 5, α = 0.8. Here, “the rate of change of the representative value within the failure determination interval when the illumination device is operating normally” is, for example, a general observation target portion OE (preliminarily assumed when the observation device 1 is used). and performing tests observed the subject), in a state where the lighting apparatus 2 is operating normally, to the representative value of the observation target image representative value of the observation target image at time t _n is at time t _n-1 It is the rate of change when the maximum representative value is changed by evaluating how much it can change at maximum, and this is hereinafter referred to as the normal maximum rate of change β. As a factor of the change in the representative value in the normal operating state, there are a case where the user moves the observation apparatus 1 to change the observation range or the observation distance, and a case where the object OJ itself moves. . Such test observation includes such movement of the observation apparatus 1.

  The determination reference constant α is determined as α <1- | β | with respect to the normal maximum change rate β. When the failure determination criterion set based on the determination criterion constant α determined in this way is used, the representative value does not become smaller than the failure determination criterion when the lighting device 2 is operating normally. Therefore, when the lighting device 2 is in a normal state, it is possible to prevent an erroneous determination result that the failure determination unit 7 determines as a failure from being output.

At the failure determination interval corresponding to the frame rate, for example, 1/30 second, the observation range and the observation distance for the observation target image do not change significantly. Therefore, if the lighting apparatus 2 is operating normally, the failure determination area at time t _n does not change significantly with respect to the failure determination area at time t _n-1. Further, in this case, does not change significantly with respect to the representative value of the representative value is also a time t _n-1 at time t _n. As a result, the maximum normal change rate β is sufficiently smaller than 1.

  On the other hand, the “change rate of the representative value at the time of failure” is mainly determined by the failure mode. The failure mode assumed in this embodiment is a failure of the illumination device 2 being observed. An example of the failure of the illumination device 2 during observation is that illumination light is not emitted at all due to disconnection of the optical connector 26a or disconnection of the optical fiber 24a. Further, there is a failure in which illumination light is not emitted from one or both of the two light conversion members 22 due to disconnection of the optical connector 26b or 26c or disconnection of the optical fiber 24b or 24c.

  As described above, in the present embodiment, the two light conversion members 22 have substantially the same characteristics and the optical coupler 25 is formed with a spectral ratio of 1: 1, so that illumination light is emitted from one of the two light conversion members 22. In the case of a failure that is not emitted, the average value of the pixel values in the failure determination area is reduced to about half. On the other hand, in a failure where illumination light is not emitted at all, the pixel value in the failure determination area is almost zero.

  Further, as described above, the optical fibers 24a, 24b, and 24c are single-line optical fibers. When the optical fibers 24a, 24b, and 24c are disconnected, the amount of illumination light is instantaneously reduced. As a result, the representative value changes during the failure determination interval corresponding to the frame rate.

  The above-mentioned “rate of change of representative value at the time of failure” refers to the rate of change of the representative value at the time of failure where the change in the representative value is the smallest among the assumed failure modes. Hereinafter, this is referred to as a failure minimum change rate γ.

  Therefore, the determination reference constant α is determined as α> 1− | γ | with respect to the minimum change rate γ at the time of failure.

  By setting a failure criterion based on the determination criterion constant α determined in this way, and using this failure criterion for failure determination, any failure among the assumed failure modes will occur reliably. Can be determined.

  When the characteristics and amount of illumination light emitted from the two light conversion members 22 are different from each other, weighting is set so that failure determination can be performed when either illumination light is not emitted. is necessary. The failure determination criterion α is determined using the weighted failure minimum change rate γ. Alternatively, the change in the representative value in the assumed failure mode may be evaluated in advance to determine the failure minimum change rate γ. As an example of the case where the characteristics and amount of illumination light emitted from the two light conversion members 22 are different, for example, when the characteristics of the two light conversion members 22 are different, the branching ratio of the optical coupler 25 is In the case of unbalance, there are cases where there is a difference in the amount of illumination light emitted from the two light conversion members 22.

  From the above, failure determination set based on the determination reference constant α determined as 1− | γ | <α <1− | β | using the maximum normal change rate β and the minimum change rate γ during failure. Using the criteria prevents wrong decisions. That is, the failure determination unit 7 can reliably determine a failure with respect to the assumed failure mode. In addition, since the frame rate used in an observation apparatus that acquires a moving image is generally about 30 fps, the observation range / observation distance for an observation target image is large at a failure determination interval corresponding to the frame rate, for example, 1/30 seconds. It does not change. Therefore, the normal maximum change rate β is never smaller than the failure minimum change rate γ.

Next, operation | movement of the observation apparatus 1 which concerns on this embodiment is demonstrated according to the flowchart of FIG.
In the observation apparatus 1 according to the present embodiment, first, in step S1, the imaging target 31 of the image acquisition unit 3 acquires an observation target image at a predetermined frame rate. The frame rate is 30 fps, for example. The imaging unit 31 outputs blue, green, and red pixel signals of the captured image to the image processing unit 32. The image processing unit 32 performs image processing on the pixel signals of each color input from the imaging unit 31. The image processing unit 32 outputs the pixel signal after the image processing to the image display unit 4 and the failure determination area setting unit 5 as an observation target image. The image display unit 4 displays this observation target image.

  Next, in step S2, the failure determination area setting unit 5 determines the focus from the focus evaluation value calculated by the contrast method. That is, first, the failure determination area setting unit 5 divides the observation target image into a plurality of areas. Next, the failure determination area setting unit 5 calculates a focus evaluation value by the contrast method using the contrast of the blue pixel value in each divided area. The failure determination area setting unit 5 determines whether or not it is in focus based on the calculated focus evaluation value.

  Next, in step S <b> 3, the failure determination area setting unit 5 determines whether or not an in-focus area exists in the observation target image. If the in-focus area exists, the process proceeds to step S4. If there is no in-focus area, the process proceeds to step S5. In the processing of step 4 and step 5, the failure determination area setting unit 5 sets the attention area 36 at a timing corresponding to the frame rate of the image acquisition unit 3.

  That is, in step S4, the failure determination area setting unit 5 sets the focused area as the attention area 36, and stores the attention area 36 together with the pixel value information in the area storage section 51. That is, among the plurality of divided areas, the failure determination area setting unit 5 sets, as the attention area 36, an in-focus area that has an in-focus evaluation value equal to or higher than a predetermined in-focus threshold (high contrast). And it is memorize | stored in the area | region memory | storage part 51 with the pixel value information.

  In step S5, the failure determination region setting unit 5 sets the attention region 36 of the observation target image stored in the region storage unit 51 at the time of failure determination one frame before as the failure determination attention region 36 of the current frame. To do. That is, when the plurality of divided areas do not have a focus evaluation value equal to or greater than the focus threshold value, the attention area 36 in the observation target image acquired in the previous frame stored in the area storage unit 51 is selected as the attention of the current frame. This is set as the area 36.

  After step S4 or step S5, in step S6, the failure determination area setting unit 5 sets the set attention area as a failure determination area. Since the attention area 36 is set as a failure determination area, the failure determination interval is synchronized with the frame rate. Further, since the failure determination area is the attention area 36, it is an area excluding an area where sufficient reflected light cannot be obtained to detect a change in the representative value at the time of failure. The failure determination region setting unit 5 outputs the pixel value information of the set failure determination region to the representative value extraction unit 6 as failure determination region information.

  Next, in step S7, the representative value extraction unit 6 extracts “a ratio of“ the average value of red pixel values in the failure determination region ”to“ the laser light amount at the time of observation target image acquisition ”” as a representative value. The representative value extraction unit 6 calculates an average value of the red pixel values in the failure determination region from the failure determination region information received from the failure determination region setting unit 5. The representative value extraction unit 6 also obtains the observation target image. Is obtained as LD light quantity information at the time of acquiring the observation target image from the light source control unit 23. Then, a ratio between the laser light quantity acquired from the LD light quantity information and the average value of the red pixel values in the failure determination area is calculated. The value extraction unit 6 extracts the calculated ratio as a representative value, and the representative value extraction unit 6 outputs the extracted representative value to the failure determination unit 7.

  Next, in step S8, the failure determination standard is set by the failure determination unit 7. That is, the failure determination unit 7 sets a failure determination criterion based on the representative value of the previous frame stored in the representative value storage unit 71.

  In step S9, the failure determination unit 7 determines whether or not the representative value extracted in step S7 is smaller than the failure determination criterion set in step S8 as failure determination processing. If the representative value extracted in step S7 is greater than or equal to the failure criterion, the process proceeds to step S10. If the representative value is smaller than the failure criterion, the process proceeds to step S11.

  That is, in this failure determination process, the failure determination unit 7 acquires a representative value in the failure determination region of the observation target image one frame before from the representative value storage unit 71. Next, a value obtained by multiplying the representative value “ratio of the average value of the red pixel value to the laser light amount” in the failure determination region of the observation target image one frame before by a determination reference constant α (0 <α <1). Set as failure criteria. Then, the failure determination criterion set based on the representative value acquired from the representative value storage unit 71 is compared with the representative value in the failure determination region of the observation target image of the current frame.

  If the representative value in the failure determination area of the observation target image of the current frame is equal to or greater than the set failure determination criterion, the failure determination unit 7 determines that the lighting device 2 is not defective, that is, is operating normally. In this case, in step S 10, the failure determination unit 7 stores the representative value of the current frame in the representative value storage unit 71. Then, the process proceeds to step S1, and the operation for the next frame is performed.

  On the other hand, when the representative value in the failure determination area of the observation target image of the current frame is smaller than the set failure determination criterion, the failure determination unit 7 determines the lighting device 2 as a failure. In this case, in step S <b> 11, a failure determination signal is output from the failure determination unit 7 to the light source control unit 23 and the image display unit 4.

  In step S12, the light source control unit 23 stops the light source 21 according to the failure determination signal, and the routine ends. The light source control unit 23 may output a signal for reducing the laser light amount to a safe light amount to the light source 21 instead of stopping the light source 21 according to the failure determination signal. Here, the safe light quantity is, for example, a light quantity of laser class 1 or less. Further, since the failure determination signal is also output to the image display unit 4, the image display unit 4 can notify the user of the failure of the lighting device 2.

Next, the effect of this embodiment will be described.
According to the present embodiment, since the failure is determined based on the change in the pixel value of the failure determination region with respect to the laser light quantity from the light source 21 at the time of acquisition by the image acquisition unit 3, the observation target acquired by the image acquisition unit 3 This can be handled by image processing of the image 33 or the like. Therefore, the failure determination of the illumination device 2 can be performed without newly adding an optical member or a detector.

  By setting the region to be set as the failure determination region by the failure determination region setting unit 5 as the attention region 36 with high contrast, except for a region where it is difficult to detect a change in the pixel value such as a lumen and an erroneous determination is likely to occur. Failure determination can be performed.

  Further, in the calculation of the focus evaluation value of the failure determination area setting unit 5, when the reflected light of blue light, that is, the blue pixel value is used for the calculation of the contrast, particularly when the object OJ is a living body such as a human body. In addition, since the contrast between the blood vessel portion on the surface of the body cavity and the other portion is emphasized, the accuracy of focusing determination by the contrast method is improved, and as a result, the accuracy of failure determination is improved.

  Similarly, when the representative value is extracted by the representative value extraction unit 6, only the red pixel value is used, and thus, when the object OJ is a living body such as a human body, the representative when the blood vessel amount changes in the failure determination region. Since the change in the value is small, the difference between the pixel value change at the normal time and the pixel value change at the time of failure becomes large, so that the failure determination is easy.

  Further, in the representative value extraction unit 6, light is emitted from one of the two light conversion members 22 by using a value that can reflect the pixel value of the entire failure determination region, such as an average value in the failure determination region, as the representative value. It becomes easy to detect a failure in which the pixel value of the failure determination area as a whole is reduced.

  Further, the representative value extraction unit 6 controls the laser light amount so that the pixel value of the observation target image becomes equal to or greater than a certain value by taking the ratio of the pixel value to the laser light amount at the time of obtaining the observation target image as a representative value. ), The pixel value in the failure determination area is not erroneously detected as a failure.

  In addition, by setting a failure determination criterion based on the determination criterion constant α determined using the normal maximum change rate β and the failure minimum change rate γ, it is possible to determine the expected failure mode without erroneous determination. Failure is reliably determined.

  The light source 21 may use an LD or an LED that emits light of another desired wavelength instead of the blue LD. When using an LED, it is preferable in terms of coupling efficiency to use a bundle fiber instead of a single optical fiber. The light conversion member 22 may be a diffusion member that converts only the light distribution characteristics of the laser light instead of the phosphor 223.

  In this embodiment, the blue pixel value is used to calculate the focus evaluation value. However, the green pixel value may be used instead of the blue pixel value, or both the blue pixel value and the green pixel value may be used. A value may be used. Alternatively, in particular, when observing an observation target having no blood vessels other than the body cavity, any or all of the pixel values of red, blue, and green may be used.

  In the present embodiment, the failure determination interval is made to coincide with the image acquisition interval. However, if the failure determination interval is within a range where the normal maximum change rate β is not smaller than the failure minimum change rate γ, the failure determination interval is greater than the image acquisition interval. Can be long. In that case, the normal maximum change rate β becomes larger, so the determination reference constant α is made smaller as necessary.

  Further, the failure determination criterion may be the same for all failure determination regions for an object OJ whose representative value does not change significantly with time when the lighting device 2 is operating normally. .

  The representative value is simply “the average value of the red pixel values in the failure determination region” instead of “the ratio of“ the average value of the red pixel values in the failure determination region ”to the“ laser light amount at the time of obtaining the observation target image ””. In this case, the representative value “average value of red pixel values in the failure determination region” and “laser light amount when the observation target image is acquired” are used in the failure determination unit 7 instead of the representative value extraction unit 6. The calculated ratio is further compared with the failure determination criterion in the failure determination unit 7. The ratio between the representative value and the laser light quantity when the observation target image is acquired is the failure determination criterion. Is greater than the ratio, the ratio between the representative value and the amount of laser light when the observation target image is acquired is stored in the representative value storage unit 71.

  Further, when the failure determination unit 7 determines a failure, the ratio between the “average value of red pixel values in the failure determination region” and the “laser light amount” may not be calculated. In particular, in the observation apparatus 1 having the illumination device 2 without the light amount adjustment function, the light amount when the illumination device 2 is in a normal state is constant. Therefore, even if the “average value of red pixel values in the failure determination region” is simply used as a representative value, the light amount value is constant in the failure determination in the failure determination unit 7, so there is no need to consider changes in the light amount. In other words, even if the “average value of red pixel values in the failure determination region” is simply used as a representative value, no erroneous detection of failure determination occurs.

  In the present embodiment, the image display unit 4 is described as a part of the observation apparatus 1, but may be provided outside the observation apparatus 1.

[Modification]
Next, a modification of this embodiment will be described.
In the state in which the illumination device 2 is operating normally, the center of gravity of the distribution of red pixel values on the image in the failure determination region is, for example, illumination light that is irradiated from one light conversion member 22 to the observation target unit OE, and It exists in the vicinity of the center of the portion where the illumination light irradiated from the other light conversion member 22 to the observation target portion OE overlaps. In the observation target image 33, as shown on the left side in FIG. 7, the center of gravity 37 of the distribution of red pixel values in the failure determination region (target region 36) is one of the two light conversion members 22 (light That is, the reflected light 381 of the illumination light from the conversion member 1) and the reflected light 382 of the illumination light from the other of the two light conversion members 22 (light conversion member 2) are present near the center of the overlapping portion. .

  On the other hand, for example, in a failure in which illumination light is not emitted from one of the two light conversion members 22, the center of gravity of the distribution of red pixel values on the image is irradiated from the other light conversion member 22 that has not failed. Move near the center of the illumination light. That is, as shown on the right side in FIG. 7, in the observation target image 33, the center of gravity 37 of the distribution of red pixel values in the failure determination region (target region 36) is reflected light of illumination light from the light conversion member 1. It moves near the center of 381.

  In this modification, failure detection is performed on the basis of the “movement amount of the center of gravity of the distribution of red pixel values in the failure determination area” at the time of failure. Here, the reason for using the red pixel value is the same as the reason described in the first embodiment.

は、以下の式（１）のように求める。 The centroid of the distribution of red pixel values on the image in the failure judgment area

Is obtained by the following equation (1).

ここで、観察対象画像３３上の赤色画素の位置を（ｉ，ｊ）とし、観察対象画像３３上の位置（ｉ，ｊ）の赤色画素値をＲ_ｉ，ｊとする。また、故障判定領域内のすべての赤色画素が積算される。

Here, the position of the red pixel on the observation target image 33 is (i, j), and the red pixel value at the position (i, j) on the observation target image 33 is R _{i, j} . In addition, all red pixels in the failure determination area are integrated.

と、時刻ｔ_ｎにおける故障判定領域内の赤色画素値の画像上の分布の重心

との距離を、代表値として抽出する。すなわち、代表値抽出部６は、

を、時刻ｔ_ｎにおける代表値として設定する。 The representative value extraction unit 6 calculates the center of gravity of the distribution of red pixel values in the failure determination area at time t _n−1 on the image.

And the center of gravity of the distribution of red pixel values in the failure determination area at time t _n

Is extracted as a representative value. That is, the representative value extraction unit 6

Is set as a representative value at time t _n .

Failure determination unit 7 determines a failure if the representative value exceeds the failure criteria T _1. This failure determination criterion T ₁ is determined from the change amount β ₁ of the representative value within the failure determination interval when the lighting device 2 is operating normally, and the change amount γ ₁ of the representative value at the time of failure. Therefore, β ₁ <T ₁ <γ ₁ is set. The setting method of β ₁ and γ _{1 is the same as} the setting method of β and γ in the first embodiment.

  As described above, the failure can also be detected by performing the failure determination based on the movement amount of the center of gravity of the distribution of the red pixel values in the failure determination area on the image.

[Second Embodiment]
The observation apparatus 1 according to the second embodiment has the same configuration as that of the observation apparatus 1 according to the first embodiment, but the failure mode and failure determination process are different. Hereinafter, a different part from 1st Embodiment is demonstrated. In a failure in which the phosphor 223 provided on the light conversion member 22 is detached due to an external force or the like, or the holder 221 holding the phosphor 223 is detached, the laser light is irradiated to the observation target portion OE without wavelength conversion. Is done. For example, in the light conversion member 22 shown in the first embodiment, the blue LD is directly emitted to the observation target unit OE by detaching the YAG phosphor. In such a failure, the color of the reflected light also changes due to the change in the color of the emitted illumination light. Therefore, failure determination processing corresponding to the color change is necessary.

  In the second embodiment, the center of the distribution on the chromaticity coordinate of the pixel value in the failure determination area is used as a value characterizing the color of the reflected light. Here, the chromaticity coordinates mean x and y in the xyY color system. The xyY color system is obtained from the RGB color system by a predetermined conversion. Moreover, the failure determination in the failure determination unit 7 is performed based on the movement amount of the center of the distribution on the chromaticity coordinates of the failure determination region. The failure determination flow is the same as that of the first embodiment shown in FIG.

  In the observation apparatus 1 according to the present embodiment, the failure determination region setting unit 5 sets a pixel region whose red pixel value is equal to or greater than a predetermined threshold in the observation target image 33 as the attention region 36. Here, for example, the sample subject is observed in advance, and the minimum red pixel value necessary for detecting a change in the representative value at the time of failure is set as the predetermined threshold value.

  In the representative value extracting unit 6, the representative values are ““ the center of the distribution on the chromaticity coordinate of the pixel value in the failure determination area in the observation target image acquired immediately before ”and“ the failure determination in the observation target image acquired at the present time ”. It is extracted as a “distance” from the center “distribution on the chromaticity coordinates of the pixel value in the region”.

は、以下の式（２）のようにして求められる。 Here, the center of the distribution of pixel values on the chromaticity coordinate in the failure determination area

Is obtained by the following equation (2).

ここで、観察対象画像３３上の位置（ｉ，ｊ）に赤、緑、青色の３つの画素が存在するとし、観察対象画像３３上の位置（ｉ，ｊ）における赤色画素値、緑色画素値、赤色画素値から得られる色度座標を（ｘ_ｉ，ｊ，ｙ_ｉ，ｊ）とし、故障判定領域内の各色の画素数をＮとする。また、故障判定領域内のすべての画素が積算される。

Here, it is assumed that three pixels of red, green, and blue exist at the position (i, j) on the observation target image 33, and the red pixel value and the green pixel value at the position (i, j) on the observation target image 33. The chromaticity coordinates obtained from the red pixel values are (x _{i, j} , y _{i, j} ), and the number of pixels of each color in the failure determination area is N. In addition, all the pixels in the failure determination area are integrated.

と、時刻ｔ_ｎにおける観察対象画像３３の故障判定領域内の画素値の色度座標上の分布の中心

との距離を時刻ｔ_ｎにおける代表値として抽出する。すなわち、代表値抽出部６は、

として代表値を設定する。 The representative value extraction unit 6 then calculates the center of the distribution on the chromaticity coordinates of the pixel values in the failure determination area of the observation target image 33 at time t _n−1 .

And the center of the distribution on the chromaticity coordinate of the pixel value in the failure determination area of the observation target image 33 at time t _n

It is extracted as the representative value at time t _n the distance between. That is, the representative value extraction unit 6

As a representative value.

Failure determination unit 7 determines a failure if the representative value exceeds the failure criteria T _2. Here, the failure determination criterion T ₂ is determined from the change amount β ₂ of the representative value within the failure determination interval in a state where the lighting device is operating normally, and the change amount γ ₂ of the representative value at the time of failure. Therefore, β ₂ <T ₂ <γ ₂ is set. The method for setting β _{2 and} γ ₂ is the same as the method for setting β and γ in the first embodiment.

  According to the second embodiment, the center of the distribution of the pixel values on the chromaticity coordinates is used as a value characterizing the color of the reflected light in the failure determination area. A holder for holding the phosphor 223 by detaching the phosphor 223 provided on the light conversion member 22 by an external force load or the like by performing failure determination based on the movement amount of the center of the distribution of the pixel value on the chromaticity coordinate It is possible to detect a failure such as 221 detaching together.

  In this embodiment, the center of the distribution on the x and y coordinates in the xyY color system is used as a value characterizing the color of the reflected light. However, other coordinates such as the u and v coordinates in the Luv color system are used. A system may be used. In addition to the amount of movement at the center of the distribution, another index such as a change in the absolute value at the center of the distribution may be used.

[Third Embodiment]
In addition to the configuration of the observation apparatus 1 of the first embodiment, the observation apparatus 1 according to the third embodiment further includes a representative value extraction unit and a failure determination unit that have the functions described in the second embodiment. is there. That is, as shown in FIG. 8, the observation apparatus 1 according to the third embodiment includes a plurality of representative value extraction units 6a and 6b and failure determination units 7a and 7b, and includes the representative value extraction unit 6a and the failure determination unit. 7a is the same as the representative value extraction unit 6 and the failure determination unit 7 in the first embodiment, and the representative value extraction unit 6b and the failure determination unit 7b are the representative value extraction unit 6 and the failure determination unit 7 in the second embodiment. Is the same. In addition, the observation apparatus 1 according to the third embodiment transmits a signal corresponding to the risk level of the failure based on the information on the result of the plurality of failure determinations made by the plurality of failure determination units 7a and 7b. And a failure handling determination unit 8 for outputting to the image display unit 4.

  Hereinafter, portions of the observation apparatus 1 according to the present embodiment that are different from the first and second embodiments will be described in detail.

  In the present embodiment, the failure determination area setting unit 5 is connected to each of the representative value extraction units 6a and 6b. The failure determination region setting unit 5 sets the attention region 36 based on the contrast calculated from the blue pixel value for the observation target image 33 acquired by the image acquisition unit 3 as in the first embodiment. In the first and second embodiments, the failure determination area setting unit 5 outputs the set attention area 36 as a failure determination area. However, in the third embodiment, the observation object acquired immediately before that is not only that. A plurality of attention areas 36 of the image are also used, and an overlapping area is set as a failure determination area. This will be described with reference to FIGS.

The failure determination region setting unit 5 sets the attention region 36 for the acquired observation target image 33 and stores it in the region storage unit 51 as in the first embodiment. Then, the failure determination region setting unit 5 continuously sets a plurality of observation target images, in this embodiment, three frames as shown in FIG. 9, and acquires the failure determination region setting period as a failure determination region setting period. An overlapping area in the three attention areas 36 stored in is set as a failure determination area. For example, as illustrated in FIG. 9, the failure determination region setting unit 5 sets the attention region 36 in the acquired observation target image 33 at time t _n and stores the attention region 36 in the region storage unit 51. At this time, similarly, at time t _n−2 and time t _n−1 , the attention area 36 is set for each of the acquired observation target images 33 and stored in the area storage unit 51. acquired by moving or the object OJ own device 1 moves, for example, as shown in FIG. 10, the region of interest _{36 n-2} of the observation target image 33 obtained at time _{t n-2,} at time tn-1 The three attention regions of the attention region 36 _n−1 of the observed image 33 and the attention region 36 _n of the observation image 33 acquired at time tn do not match. Therefore, in this embodiment, as the failure determination area 39 for use in the failure determination at time _{t n,} these three regions of interest _{36 n-2, 36 n-} 1, 36 n using overlapping regions in. The failure determination region setting unit 5 outputs the pixel value information of the failure determination region 39 at the time t _n set in this way to the representative value extraction units 6a and 6b as failure determination region information. Note that the failure determination interval is an interval corresponding to the frame rate, as in the first embodiment, and is represented by t _n −t _n−1 .

Like the representative value extracting unit 6 of the first embodiment, the representative value extracting unit 6a calculates the “ratio of the average value of the red pixel values in the failure determination region to the laser light amount” from the failure determination region information as the first representative value. (Representative value A) is extracted and sent to the failure determination unit 7a. Further, the representative value extraction unit 6b, like the representative value extraction unit 6 of the second embodiment, ““ center of distribution on the chromaticity coordinate of the pixel value in the failure determination region in the observation target image acquired immediately before ” And “the distance between the distribution value on the chromaticity coordinate of the pixel value in the failure determination region in the observation target image acquired this time” as the second representative value (representative value B), and the failure determination unit 7b Here, for example, if “the failure determination area in the observation target image acquired this time” is an overlapping area in the three attention areas 36 _n−2 , 36 _n−1 , and 36 _n , “ The “failure determination area in the observation target image” refers to an overlapping area in the three attention areas 36 _n−3 , 36 _n−2 , and 36 _n−1 .

  Each of the failure determination units 7a and 7b performs failure determination with different danger levels based on temporal changes of different representative values. That is, the failure determination unit 7a includes a representative value storage unit 71a that stores the representative value (representative value A) received from the representative value extraction unit 6a, and is similar to the failure determination process of the first embodiment. A failure determination is performed based on a decrease in luminance of reflected light in the region. The failure detected by the failure determination unit 7a is, for example, a failure such as disconnection of the optical connectors 26b and 26c or disconnection of the optical fibers 24b and 24c. The failure determination unit 7a outputs a failure determination signal (failure determination signal A) to the failure handling determination unit 8 when determining that there is a failure. On the other hand, the failure determination unit 7b includes a representative value storage unit 71b that stores the representative value (representative value B) received from the representative value extraction unit 6b. In the failure determination similar to the second embodiment, Failure determination is performed based on the color change of the reflected light. The failure detected by the failure determination unit 7b is, for example, a failure such as detachment of the phosphor 223 or detachment of the holder 221 that holds the phosphor 223. The failure determination unit 7b outputs a failure determination signal (failure determination signal B) to the failure handling determination unit 8 when determining that there is a failure.

  The failure handling determination unit 8 sets the laser light amount to a safe light amount according to the risk level of the failure based on the failure determination result of any of the plurality of failure determination units 7a and 7b, Judgment of response by failure notification.

  In other words, when a failure determination signal is output from the failure determination unit 7a, the failure response determination unit 8 first informs the image display unit 4 to set the failure of the illumination device 2 and the laser light to a safe light amount. A signal for notifying the user is output to the image display unit 4. Thereafter, the failure handling determination unit 8 outputs a signal for causing the light source control unit 23 to set the light amount of the laser light from the light source 21 to a safe light amount to the light source control unit 23. In a failure such as disconnection of the optical connectors 26b and 26c detected by the failure determination unit 7a or disconnection of the optical fibers 24b and 24c, there is a possibility that the laser light leaks to the outside of the optical fibers 24b and 24c. Since a housing exists around the fibers 24b and 24c, there is a low possibility that the laser light will instantaneously leak to the outside and be emitted to the human body or the like. Therefore, the failure level is low.

  On the other hand, when a failure determination signal is output from the failure determination unit 7b, the failure response determination unit 8 instantaneously sets the light amount of the laser light from the light source 21 to a safe light amount from the light source control unit 23. The signal for making it output is output to the light source control part 23. After that, the failure handling determination unit 8 outputs to the image display unit 4 a signal for causing the image display unit 4 to display that the laser light has been set to a safe light amount because a failure has occurred in the lighting device 2. Thus, a failure notification is given to the user. That is, in a failure such as the detachment of the phosphor 223 detected by the failure determination unit 7b or the detachment of the holder 221 that holds the phosphor 223, the laser beam is directly emitted from the illumination light emission surface, and the laser beam is instantaneously emitted. Can be emitted to the human body or the like. Therefore, the failure detected by the failure determination unit 7b is more likely to be fatal than the failure detected by the failure determination unit 7a, and the failure level is high. Therefore, first of all, priority is given to setting the laser light to a safe light quantity.

  As described above, according to the present embodiment, the failure determination region setting unit 5 sets the failure determination region 39 as an overlapping region in the plurality of attention regions in the plurality of observation target images acquired immediately before. The determination area can be set more appropriately than setting from one observation target image. As a result, erroneous determination of failure determination in the failure determination unit 7a and / or 7b is prevented.

  Further, according to the present embodiment, the laser light is set to a safe light amount according to the danger level of the failure based on the information on the result of the plurality of failure determinations made by the plurality of failure determination units 7a and 7b, and the user For example, it is possible to take appropriate measures for each detected failure.

OJ ... object, OE ... observation target part, 1 ... observation device, 2 ... illumination device, 3 ... image acquisition part, 4 ... image display part, 5 ... failure determination area setting part, 6, 6a, 6b ... representative value extraction 7, 7a, 7b ... failure determination unit, 8 ... failure response determination unit, 21 ... light source, 22 ... light conversion member, 23 ... light source control unit, 24a, 24b, 24c ... optical fiber, 25 ... optical coupler, 26a , 26b, 26c ... optical connector, 31 ... imaging unit, 32 ... image processing unit, 33 ... observation target image, 34 ... black portion, 35 ... divided region, 36, 36 n _... attention area, the distribution of 37 ... red pixel value 39 ... failure determination region, 51 ... region storage unit, 71, 71a, 71b ... representative value storage unit, 221 ... holder, 222 ... light transmission member, 223 ... phosphor, 224 ... reflection member, 381, 382 ... reflected light.

Claims (25)

A light source that emits primary light; a light guide member that guides the primary light emitted from the light source; and the primary light guided by the light guide member to receive the primary light. An illumination device that irradiates the observation target part with illumination light as light emitted from the light conversion member;
An image acquisition unit that images the observation target unit and repeatedly acquires an observation target image including a plurality of pixels;
A failure determination region setting unit for setting a failure determination region used for failure determination of the lighting device in the observation target image;
A representative value extracting unit that extracts a representative value indicating a state of a pixel value in the failure determination region;
A failure determination unit that determines a failure of the lighting device based on the change in the representative value;
An observation apparatus comprising:   The failure determination region setting unit is a region in the observation target image in which the observation target unit exists at a distance at which sufficient reflected light can be obtained to detect a change in the representative value when the lighting device fails. The observation apparatus according to claim 1, wherein the failure determination area is set based on the attention area in the observation target image.   The failure determination region setting unit sets the attention region for each of the observation target images that are periodically and repeatedly acquired, and each of the plurality of observation target images acquired during the failure determination region setting period. The observation apparatus according to claim 2, wherein the failure determination area is set based on the set plurality of attention areas.   The failure determination region setting unit sets an overlapping region as the failure determination region in the plurality of attention regions set for the plurality of observation target images acquired in the failure determination region setting period. The observation apparatus described.   The observation apparatus according to claim 2, wherein the failure determination area setting unit sets an area of high contrast in the observation target image as the attention area. The image acquisition unit includes an imaging unit having a red pixel, a green pixel, and a blue pixel,
The failure determination area setting unit calculates a contrast based on a pixel value of one of the green pixel value and the blue pixel value of the observation target image, and sets a high contrast area as the attention area. The observation apparatus according to claim 5.   The observation apparatus according to claim 2, wherein the failure determination area setting unit sets an area of a pixel having a predetermined pixel value or more in the observation target image as the attention area. The image acquisition unit includes an imaging unit having a red pixel, a green pixel, and a blue pixel,
The observation apparatus according to claim 7, wherein the failure determination area setting unit sets an area of a pixel in which a pixel value of the red pixel is equal to or larger than a predetermined pixel value in the observation target image as the attention area.   The failure determination area setting unit, when the attention area does not exist in the observation target image, sets the attention area of the observation target image acquired immediately before as the attention area for the current failure determination. The observation apparatus according to any one of 8.   The failure determination unit sets a failure determination criterion based on the representative value extracted from the failure determination region in the observation target image acquired immediately before, and the failure in the observation target image acquired at the present time The observation device according to claim 1, wherein a failure determination of the lighting device is made based on a relationship between the representative value extracted from the determination region and the set failure determination criterion. The failure determination unit repeatedly performs a failure determination of the lighting device at a predetermined failure determination interval,
Within the failure determination interval, a change in the representative value during a normal operation of the illumination device due to a change in the distance between the observation device and the observation target portion and a change in the imaging range, and the failure in the illumination device Based on the change in the representative value, the failure determination is performed so that a failure is not determined in the change in the representative value during the normal operation and a failure is determined in the change in the representative value during the failure. The observation apparatus according to claim 10, wherein a reference is set.   The observation apparatus according to claim 10 or 11, wherein the representative value extraction unit extracts the representative value from information on a predetermined pixel value in the failure determination region. The representative value extraction unit sets the ratio of the predetermined pixel value in the failure determination region to the light amount of the primary light at the time of obtaining the observation target image as the representative value,
The observation device according to claim 12, wherein the failure determination unit determines that a failure occurs when the representative value is smaller than the failure determination criterion.   The observation apparatus according to claim 12, wherein the representative value extraction unit uses, as the representative value, a value indicating a change in a position of a distribution of the predetermined pixel value on the image in the failure determination region. The image acquisition unit includes an imaging unit having a red pixel, a green pixel, and a blue pixel,
The observation device according to claim 13 or 14, wherein the predetermined pixel value is a pixel value of the red pixel. The image acquisition unit includes an imaging unit having a red pixel, a green pixel, and a blue pixel,
The observation apparatus according to claim 12, wherein the representative value extraction unit uses, as the representative value, a value indicating a change in color of reflected light of the illumination light from the observation target unit in the failure determination region. A light source controller for controlling the light source;
When the failure determination unit determines as a failure, it outputs a failure determination signal to the light source control unit,
The observation apparatus according to claim 1, wherein the light source control unit controls the light source so as to set the light amount of the primary light to a safe light amount when receiving the failure determination signal. An image display unit for displaying the observation target image and the result of the failure determination;
When the failure determination unit determines that a failure has occurred, the failure determination unit outputs a failure determination signal to the image display unit,
The observation apparatus according to claim 1, wherein the image display unit notifies the user of a failure when receiving the failure determination signal. It further comprises a failure response determination unit that determines the response at the time of failure,
The failure correspondence determination unit extracts different representative values from the failure determination area, and outputs a signal corresponding to a failure risk level based on a plurality of failure determination results made based on changes in the different representative values. The observation apparatus according to claim 1 which outputs. The illumination device further includes at least one light conversion member other than the light conversion member,
The light guide member branches the primary light and guides it to the plurality of light conversion members,
The observation apparatus according to claim 1, wherein illumination light emitted from the plurality of light conversion members is turned on or off in synchronization with each other when the light source is turned on or off. A light source that emits primary light; a light guide member that guides the primary light emitted from the light source; and the primary light guided by the light guide member to receive the primary light. An illumination device that irradiates the observation target portion with illumination light as light emitted from the light conversion member. Operating method of the observation device ,
The observation device imaging the observation target portion, repeatedly acquires the observation target image composed of a plurality of pixels,
The observation device sets a failure determination region used for failure determination of the illumination device in the observation target image,
The observation apparatus extracts a representative value indicating the state of pixel values in the failure determination area,
The observation apparatus, cormorants line failure determination of the illumination device based on a change of the representative value,
How to operate the observation device . The failure determination of the illumination device performed by the observation device sets a failure determination criterion based on the representative value extracted from the failure determination region in the observation target image acquired immediately before, and the acquired at the present time The method of operating an observation apparatus according to claim 21, wherein the operation is performed based on a relationship between the representative value extracted from the failure determination region in the observation target image and the set failure determination criterion. In the failure determination of the illumination device performed by the observation device, the observation device controls the light source so that the light amount of the primary light is set to a safe light amount when the illumination device is determined as a failure. 23. A method for operating the observation apparatus according to 21 or 22. 23. The method of operating an observation apparatus according to claim 21 or 22 , wherein when the illumination apparatus determines that the illumination apparatus has failed in the failure determination of the illumination apparatus performed by the observation apparatus, the observation apparatus notifies the user of the failure. . In the extraction of the representative value performed by the observation device , the observation device extracts a plurality of different representative values from the failure determination area,
In the failure determination of the illumination device performed by the observation device, the observation device performs a plurality of failure determinations with different risk levels based on changes in the plurality of representative values that are different from each other,
The observation apparatus according to security level of the fault on the basis of the information of the plurality of failure of the determination result, the setting of the light intensity of the primary light to a safe amount or intends line failure notification to the user according to claim 21 or 23. A method of operating the observation apparatus according to 22.

Images