JPWO2017158901A1 - Image processing apparatus, operation method of image processing apparatus, and image processing program - Google Patents

Abstract

The image processing apparatus includes a representative image setting unit that sets a representative image representing a group of images captured by a capsule endoscope introduced into the digestive tract of a subject, and a predetermined section in the digestive tract A count is assigned to each of the representative images, and based on the count and the total number of the representative images included in the section, the length from the start position of the section to the representative image with respect to the total length of the section A distance ratio calculation unit that calculates a ratio for each representative image. As a result, it is possible to specify at which part in the digestive tract each acquired image is an image taken, and an image that can specify an image corresponding to a characteristic part from an image group taken in the digestive tract. A processing device is provided.

Description

  The present invention relates to an image processing apparatus, an operation method of the image processing apparatus, and an image processing program.

  2. Description of the Related Art Conventionally, a technique for diagnosing a subject using a series of images acquired by a capsule endoscope that images the inside of the digestive tract of the subject in time series is known.

  In this technique, a user such as a doctor visually determines at which part in the digestive tract each acquired image is an image. For example, when diagnosing the large intestine, an image corresponding to a characteristic part (ie, ileocecal region, liver curvature, spleen curvature) that is easy for the user to visually distinguish is set, and the space between the ileocecal region and liver curvature is ascending The colon, hepatic curvature and splenic curvature were determined to be the transverse colon, and the splenic curvature and subsequent were the descending colon.

  Patent Document 1 discloses a position for displaying a rough position where each image is captured in order to assist the user in grasping at which part in the digestive tract each acquired image is an image. An endoscope system that displays a marker on a display unit is disclosed.

Japanese Patent No. 5248834

  However, Patent Document 1 does not disclose a means for specifying at which part in the digestive tract each acquired image is an image taken. In addition, no means for specifying an image corresponding to a characteristic part is disclosed. Therefore, even if the technique of Patent Document 1 is used, the user can specify in which part of the digestive tract each image is captured, or specify the image corresponding to the characteristic part by visual inspection. There was a problem of requiring a lot of time.

  The present invention has been made in view of the above, and specifies which part in the digestive tract each acquired image is an image captured from, and features from an image group captured in the digestive tract An object of the present invention is to provide an image processing apparatus, an operation method of the image processing apparatus, and an image processing program capable of specifying an image corresponding to a part.

  In order to solve the above-described problems and achieve the object, an image processing apparatus according to an aspect of the present invention is a representative of an image group captured by a capsule endoscope introduced into a digestive tract of a subject. A representative image setting unit for setting an image, and assigning a count to each of the representative images included in a predetermined section in the digestive tract, and based on the count and the total number of the representative images included in the section A distance ratio calculating unit that calculates a ratio of the length from the start position of the section to the representative image with respect to the total length of the section for each representative image.

  The image processing apparatus according to an aspect of the present invention further includes a specifying unit that specifies at least one feature part image obtained by imaging the feature part of the digestive tract from the representative image based on the ratio. It is characterized by.

  The image processing apparatus according to an aspect of the present invention further includes a comparison unit that compares a parameter relating to a position of a characteristic site in the digestive tract and the ratio for each representative image, and the specifying unit includes the comparison The at least one representative image is specified as the feature part image based on the comparison result of the parts.

  In the image processing device according to an aspect of the present invention, the ratio of the representative image included in a partial section of the section in which the ratio is determined based on the parameter based on a comparison result of the comparison unit. A candidate image group setting unit that sets a candidate image group that is a set is further provided, and the specifying unit matches the parameter and the ratio based on a comparison result of the comparison unit, or the parameter and the ratio The at least one representative image having a close value is specified as the feature part image from the candidate image group.

  The image processing apparatus according to an aspect of the present invention further includes a storage unit that stores a position of the capsule endoscope when each image of the image group is captured, and the specifying unit includes the position Based on the above, at least one feature part image is specified from the candidate image group.

  In the image processing device according to an aspect of the present invention, the comparison unit compares the change amount of the position with a predetermined threshold value, and the characteristic part image is the change in the position of the subject in the left-right direction. The amount has a value larger than a predetermined threshold, the amount of change in the position of the subject in the vertical direction has a value smaller than a predetermined threshold, and the change in the position in the front-rear direction of the subject The representative image having a value smaller than a predetermined threshold value, or the amount of change in the position of the subject in the left-right direction has a value smaller than a predetermined threshold value, and the position in the vertical direction of the subject. The representative image has a value smaller than a predetermined threshold value, and the amount of change in the position of the subject in the front-rear direction has a value smaller than a predetermined threshold value.

  The image processing apparatus according to an aspect of the present invention further includes a storage unit that stores a posture of the capsule endoscope when each image of the image group is captured, and the specifying unit includes the posture Based on the above, at least one feature part image is specified from the candidate image group.

  In the image processing device according to an aspect of the present invention, the comparison unit compares the elevation angle of the capsule endoscope with a predetermined threshold in the posture, and the feature part image has a predetermined elevation angle. The representative image having a value larger than the threshold value of.

  The image processing apparatus according to an aspect of the present invention further includes a storage unit that stores, for each image, radio wave reception intensity when each image of the image group is captured, and the specifying unit includes the radio wave reception intensity. Based on the above, at least one feature part image is specified from the candidate image group.

  Further, in the image processing device according to an aspect of the present invention, the storage unit stores a first radio wave reception intensity and a second radio wave reception intensity when each image of the image group is captured, A correlation calculation unit that calculates a correlation value between the first radio wave reception intensity and the second radio wave reception intensity; and the specifying unit calculates the correlation calculated by the correlation calculation unit from the candidate image group The representative image having the smallest value is specified as the feature part image.

  The image processing apparatus according to an aspect of the present invention further includes a storage unit that stores acceleration of the capsule endoscope when each image of the image group is captured, and the specifying unit includes the acceleration Based on the above, at least one feature part image is specified from the candidate image group.

  Further, in the image processing device according to an aspect of the present invention, the comparison unit compares the acceleration in the left-right direction of the subject with the predetermined threshold among the accelerations, and the feature part image is in the left-right direction The acceleration is a representative image having a value larger than a predetermined threshold, or the representative image having a value in which the acceleration in the left-right direction is smaller than a predetermined threshold.

  The image processing apparatus according to an aspect of the present invention further includes a storage unit that stores a frame rate when each image of the image group is captured, and the specifying unit is configured to perform the above-described operation based on the frame rate. It is characterized in that at least one feature part image is specified from a candidate image group.

  In the image processing device according to an aspect of the present invention, the comparison unit compares the frame rate with a predetermined threshold value, and the feature part image has a value with the frame rate larger than the predetermined threshold value. The representative image, or the representative image having a frame rate smaller than a predetermined threshold.

  The image processing apparatus according to an aspect of the present invention further includes a storage unit that stores a time when each image of the image group is captured, and the specifying unit is based on the time interval between the images. The at least one feature part image is specified from the candidate image group.

  Further, in the image processing device according to an aspect of the present invention, the comparison unit compares the time interval between the images with a predetermined threshold value, and the feature part image is the time between the images. The representative image having a value smaller than a predetermined threshold, or the representative image having a time interval between the images larger than a predetermined threshold.

  The image processing apparatus according to an aspect of the present invention is based on the ratio in the configuration diagram with respect to a display unit that displays the selected image selected from the image group and the configuration diagram of the digestive tract. And a display control unit for displaying an index indicating a position corresponding to the selected image.

  An operation method of the image processing apparatus according to one aspect of the present invention includes a representative image setting step of setting a representative image representing an image group captured by a capsule endoscope introduced into the digestive tract of a subject; And assigning a count to each of the representative images included in a predetermined section in the digestive tract, and the section with respect to the total length of the section based on the count and the total number of the representative images included in the section A distance ratio calculating step of calculating a ratio of the length from the start position to the representative image for each representative image.

  An image processing program according to an aspect of the present invention includes a representative image setting step of setting a representative image representing a group of images captured by a capsule endoscope introduced into a digestive tract of a subject, and the digestion A count is assigned to each of the representative images included in a predetermined section in the tube, and the start position of the section with respect to the total length of the section based on the count and the total number of the representative images included in the section And a distance ratio calculating step of calculating a ratio of the length from the representative image to the representative image for each representative image.

  According to the present invention, it is specified in which part in the digestive tract each acquired image is an image taken, and an image corresponding to a characteristic part is specified from a group of images taken in the digestive tract. An image processing apparatus capable of performing the above, an operation method of the image processing apparatus, and an image processing program can be realized.

FIG. 1 is a block diagram illustrating an image processing apparatus according to the first embodiment. FIG. 2 is a schematic diagram showing the structure of the large intestine portion in the digestive tract of a subject. FIG. 3 is a flowchart showing the operation of the image processing apparatus according to the first embodiment. FIG. 4 is a schematic diagram showing a series of images in the digestive tract arranged in time series. FIG. 5 is a schematic diagram showing a distance ratio from the ileocecal part to each representative image. FIG. 6 is a schematic diagram illustrating a representative image group that is a candidate for a feature region image corresponding to liver curvature and a feature region image corresponding to liver curvature. FIG. 7 is a schematic diagram illustrating a representative image group that is a candidate for a feature part image corresponding to a spleen curve and a feature part image corresponding to a spleen curve. FIG. 8 is a schematic diagram illustrating a characteristic part image corresponding to the liver curvature specified by the image processing apparatus according to the modification of the first embodiment. FIG. 9 is a schematic diagram illustrating a characteristic part image corresponding to the splenic curvature specified by the image processing apparatus according to the modification of the first embodiment. FIG. 10 is a block diagram illustrating an image processing apparatus according to the second embodiment. FIG. 11 is a flowchart showing the operation of the image processing apparatus according to the second embodiment. FIG. 12 is a schematic diagram illustrating the movement of the capsule endoscope in the large intestine. FIG. 13 is a diagram illustrating a temporal change in the position of the capsule endoscope in the large intestine. FIG. 14 is a diagram illustrating the amount of change in the position of the capsule endoscope in the large intestine. FIG. 15 is a diagram illustrating a temporal change in the elevation angle of the capsule endoscope in the large intestine. FIG. 16 is a block diagram illustrating an image processing apparatus according to the third embodiment. FIG. 17 is a flowchart showing the operation of the image processing apparatus according to the third embodiment. FIG. 18 is a flowchart showing processing for specifying a feature part image executed by the specifying unit shown in FIG. FIG. 19 is a diagram illustrating the normalized radio wave reception intensity of the first antenna. FIG. 20 is a diagram illustrating the normalized radio wave reception intensity of the second antenna. FIG. 21 is a diagram illustrating a correlation coefficient between radio wave reception intensities of two antennas. FIG. 22 is a diagram illustrating another arrangement example of the antenna. FIG. 23 is a block diagram illustrating an image processing apparatus according to the fourth embodiment. FIG. 24 is a flowchart showing the operation of the image processing apparatus according to the fourth embodiment. FIG. 25 is a diagram illustrating acceleration in the X-axis direction of the capsule endoscope in the large intestine. FIG. 26 is a diagram illustrating a change over time in the cumulative number of captured images of the capsule endoscope. FIG. 27 is a diagram illustrating a temporal change in the imaging time interval of the capsule endoscope. FIG. 28 is a block diagram illustrating an image processing apparatus according to the fifth embodiment. FIG. 29 is a diagram for explaining the operation of the part classification unit. FIG. 30 is a diagram for explaining the operation of the part classification unit. FIG. 31 is a diagram for explaining the operation of the part classification unit. FIG. 32 is a diagram illustrating display on the display unit when the selected image is the ascending colon. FIG. 33 is a diagram illustrating display on the display unit when the selected image is the transverse colon. FIG. 34 is a diagram illustrating display on the display unit when the selected image is the descending colon. FIG. 35 is a block diagram showing an image processing apparatus according to the sixth embodiment. FIG. 36 is a schematic diagram showing the structures of the small and large intestines in the digestive tract of a subject. FIG. 37 is a schematic diagram showing a distance ratio from the ileocecal part to each image. FIG. 38 is a schematic diagram showing a distance ratio from the ileocecal part to each image.

  Embodiments of an image processing apparatus, an operation method of the image processing apparatus, and an image processing program according to the present invention will be described below with reference to the drawings. Note that the present invention is not limited to these embodiments. The present invention can be applied to an image processing device used for a capsule endoscope, an operation method of the image processing device, and an image processing program in general.

  In the description of the drawings, the same or corresponding elements are denoted by the same reference numerals as appropriate. It should be noted that the drawings are schematic, and the relationship between the dimensions of each element, the ratio of each element, and the like may differ from the actual situation. Even between the drawings, there are cases in which portions having different dimensional relationships and ratios are included.

(Embodiment 1)
FIG. 1 is a block diagram illustrating an image processing apparatus according to the first embodiment. The image processing apparatus 1 according to Embodiment 1 extracts an image of a predetermined section from a series of images acquired by sequentially capturing the inside of the digestive tract of a subject with a capsule endoscope, and the section It has a function that can easily specify a characteristic part image corresponding to a characteristic part (characteristic part).

  When the predetermined section is the large intestine, the characteristic part corresponds to liver curvature or spleen curvature. FIG. 2 is a schematic diagram showing the structure of the large intestine portion in the digestive tract of a subject. As shown in FIG. 2, the large intestine portion of the subject's digestive tract has an ileocecal portion A <b> 1 that is a boundary between the ileum and the cecum along the direction in which the contents proceed, and the cecum from above the subject. ) Extending to the ascending colon A2, the curved portion of the liver curved A3, the transverse colon A4 extending in the left-right direction of the subject, the curved portion of the splenic curved A5, and the lower part of the subject (leg) The descending colon A6 extending toward the side) and the anal part A7 which is the exit of the large intestine. The image processing apparatus 1 has a function of easily specifying images corresponding to the liver curvature A3 and the spleen curve A5 from the image group in the digestive tract.

  Returning to FIG. 1, the image processing apparatus 1 acquires image data in the digestive tract of a subject imaged by a control unit 10 that controls the operation of the entire image processing apparatus 1 and a capsule endoscope (not shown). An image acquisition unit 20, an input unit 30 for inputting a signal corresponding to an external operation to the control unit 10, a storage unit 50 for storing image data and various programs acquired by the image acquisition unit 20, and an image And a calculation unit 100 that executes predetermined image processing on the data, and displays various information and images on the display unit 40 connected to the image processing apparatus 1.

  The control unit 10 is realized by hardware such as a CPU (Central Processing Unit), and by reading various programs recorded in the storage unit 50, image data input from the image acquisition unit 20 and input from the input unit 30. In accordance with a signal or the like, instructions to each unit constituting the image processing apparatus 1 and data transfer are performed, and the overall operation of the image processing apparatus 1 is comprehensively controlled.

  The image acquisition unit 20 is appropriately configured according to the mode of the system including a capsule endoscope that images the inside of the subject. For example, when a portable recording medium is used for transferring image data to and from a capsule endoscope, the image acquisition unit 20 detachably mounts the recording medium and records image data of the recorded image. It is comprised with the reader apparatus which reads. In addition, when installing a server that stores image data of an image captured by a capsule endoscope, the server includes an image acquisition unit 20, a communication device connected to the server, and performs data communication with the server. Get image data.

  The input unit 30 is realized by input devices such as a keyboard, a mouse, a touch panel, and various switches, for example, and outputs an input signal generated according to an external operation on these input devices to the control unit 10.

  The storage unit 50 includes various IC memories such as ROM (Read Only Memory) and RAM (Random Access Memory) such as update-recordable flash memory, a hard disk connected with a built-in or data communication terminal, or information such as a CD-ROM. This is realized by a recording device and its reading device. The storage unit 50 operates the image processing apparatus 1 in addition to the image data of the gastrointestinal tract image acquired by the image acquisition unit 20, and causes the image processing apparatus 1 to execute various functions. Stores data used during execution. In addition, the storage unit 50 stores the results calculated by each unit of the calculation unit 100.

  The calculation unit 100 is realized by hardware such as a CPU, and performs a process of specifying a feature part image corresponding to the liver curvature A3 and the spleen curve A5.

  Next, the configuration of the calculation unit 100 will be described. As shown in FIG. 1, the calculation unit 100 includes an ileocecal part specifying unit 110 that specifies an image corresponding to the ileocecal part A1, an anus part specifying unit 120 that specifies an image corresponding to the anus part A7, and a candidate image. A parameter setting unit 130 for setting a parameter (for example, a statistical value) for extracting a group, a representative image setting unit 140 for setting a representative image representing the image group acquired by the image acquisition unit 20, and a predetermined value. A distance ratio calculation unit 150 that calculates, for each representative image, a distance ratio that is a ratio of the length from the start position of the section to each representative image with respect to the entire length of the section (for example, between the ileocecal part A1 and the anal part A7); The comparison unit 170 that compares the distance ratio calculated by the distance ratio calculation unit 150 with the parameter set by the parameter setting unit 130, and the liver curvature A3 and the spleen curve based on the comparison result of the comparison unit 170 5 corresponds to the liver curvature A3 and the spleen curve A5 from the candidate image group based on the comparison result of the candidate image group setting unit 160 that sets the candidate image group that is a candidate for the feature part image corresponding to 5, and the comparison unit 170 And a specifying unit 180 that specifies a feature part image to be performed.

  The ileocecal part specifying means 110 acquires information for specifying an image corresponding to the ileocecal part A1 from the image group. First, a user such as a doctor selects an image corresponding to the ileocecal part A <b> 1 from the image group in the digestive tract displayed on the display unit 40, and performs a predetermined input on the input unit 30. And the ileocecal part identification means 110 acquires the information received by the input unit 30, and identifies the image selected by the user as an image corresponding to the ileocecal part A1. Thereby, the first image (first position) of the section is specified. Note that the ileocecal part specifying means 110 may perform image recognition on the image group in the digestive tract, and an image corresponding to the ileocecal valve may be an image corresponding to the ileocecal part A1. For example, when the capsule endoscope passes through the ileocecal valve, the entire captured image becomes dark. Therefore, a photo in which the entire image in the image group is dark may be detected by image recognition.

  The anal part specifying unit 120 specifies an image corresponding to the anal part A7 from the image group. First, a user such as a doctor selects an image corresponding to the anus A7 from the group of images in the digestive tract displayed on the display unit 40 and makes a predetermined input to the input unit 30. Then, the anus specifying unit 120 acquires the information received by the input unit 30, and specifies the image selected by the user as an image corresponding to the anus A7. As a result, the last image (last position) of the section is specified. The anus specifying unit 120 may perform image recognition on the image group in the digestive tract, and an image corresponding to the anus may be an image corresponding to the anus A7. For example, the last image before the capsule endoscope is ejected from the anus A7 and the entire image becomes bright may be detected by image recognition.

  The parameter setting unit 130 sets parameters used when the candidate image group setting unit 160 sets a candidate image group. Specifically, when the input unit 30 receives a user input, the parameter setting unit 130 acquires the input information and sets parameters used when the candidate image group setting unit 160 sets a candidate image group. Here, it is statistically known that the distance ratio from the ileocecal portion A1 to the liver curvature A3 with respect to the entire length of the large intestine is 11% to 19%. Similarly, it is statistically known that the distance ratio from the ileocecal portion A1 to the splenic curve A5 with respect to the entire length of the large intestine is 39% to 52%. Based on these statistical values, the user inputs parameters so that a representative image having a distance ratio of 11% to 19% is a candidate image group that is a candidate for a characteristic part image corresponding to liver curvature A3, and the distance ratio A parameter is input so that a representative image of 39% to 52% is a candidate image group that is a candidate for a characteristic part image corresponding to the splenic curve A5. Thus, the user can set parameters based on, for example, statistical values. Parameters may be set in advance so that user input can be omitted.

  The representative image setting unit 140 sets a representative image. Here, the representative image is an image picked up when the capsule endoscope moves largely in the digestive tract. The representative image setting unit 140 calculates the similarity by comparing the images before and after the image group arranged in chronological order, and sets the image as the representative image when the calculated similarity is smaller than a predetermined threshold. Alternatively, it may be determined that the capsule endoscope has moved greatly in the digestive tract by various other methods and set as a representative image.

  First, the distance ratio calculation unit 150 sequentially assigns one count to each representative image in the section. That is, the number of representative images in time series in each section is counted. Then, the ratio between the count of each representative image and the total number of representative images in the section is calculated. That is, the distance ratio calculation unit 150 calculates the number ratio of each representative image in the section. Here, since the representative image corresponds to an image with a large movement of the capsule endoscope, the number ratio is from the start position of the section (the ileocecal part A1) to each representative image with respect to the entire length of the section (for example, large intestine). It can be approximated to the distance ratio of the length. Therefore, the distance ratio calculation unit 150 stores the calculated number ratio as a distance ratio in the storage unit 50 in association with each representative image.

  The comparison unit 170 compares the value determined based on the parameter with the distance ratio for each representative image. Specifically, as described above, the statistical value of the distance ratio from the ileocecal portion A1 to the liver curvature A3 is 11% to 19%, and this statistical value is compared with the distance ratio associated with each representative image. . Alternatively, for example, the median value of 15% of the statistical values is compared with the distance ratio associated with each representative image. Similarly, the statistical value of the distance ratio from the ileocecal part A1 to the splenic curve A5 is 39% to 52%, and this statistical value is compared with the distance ratio associated with each representative image. Alternatively, for example, the median value of 45% of the statistical values is compared with the distance ratio associated with each representative image.

  Candidate image group setting section 160 sets a candidate image group that is a set of representative images included in a partial section of the large intestine based on the distance ratio calculated by distance ratio calculation section 150. Specifically, based on the comparison result of the comparison unit 170, the candidate image group setting unit 160 corresponds to the liver curvature A3 a representative image whose distance ratio calculated by the distance ratio calculation unit 150 is 11% to 19%. A candidate image group is set, and a representative image having a distance ratio of 39% to 52% is set as a candidate image group corresponding to the splenic curve A5.

  Based on the distance ratio calculated by the distance ratio calculation unit 150, the specifying unit 180 specifies at least one feature part image obtained by imaging the feature part of the digestive tract. At this time, a feature part image may be specified from the candidate image group. Specifically, the specifying unit 180 specifies a representative image whose distance ratio calculated by the distance ratio calculating unit 150 is 15% as a feature part image corresponding to the liver curvature A3 based on the comparison result of the comparing unit 170. Then, a representative image having a distance ratio of 45% is specified as a feature part image corresponding to the splenic curve A5. If there is no representative image that matches the distance ratio of 15% or 45%, the specifying unit 180 specifies the representative image having the distance ratio closest to these values as the feature part image.

  The display unit 40 is realized by a display device such as a liquid crystal or an organic EL (Electro Luminescence), and displays various information and images including images in the digestive tract under the control of the control unit 10.

  Next, the operation of the image processing apparatus 1 shown in FIG. 1 will be described. FIG. 3 is a flowchart showing the operation of the image processing apparatus according to the first embodiment. As shown in FIG. 3, first, in step S <b> 10, the image processing apparatus 1 acquires image data of a series of gastrointestinal images captured in time series order via the image acquisition unit 20 and records them in the storage unit 50. To do.

In subsequent step S11, the representative image setting unit 140 compares the images before and after the image group arranged in time series order to calculate the similarity, and when the calculated similarity is smaller than a predetermined threshold, the representative image is set as the representative. Set to image. FIG. 4 is a schematic diagram showing a series of images in the digestive tract arranged in time series. The images with the image number n ₁ , the image number n ₂ , and the image number n _M shown in FIG. 4 are representative images. That is, the total number of representative images is M.

In step S12, when the user selects an image corresponding to the ileocecal part A1 from the input unit 30, the ileocecal part specifying unit 110 acquires the information received by the input unit 30, and the image selected by the user is the ileocecal part. It is specified as an image corresponding to A1. That is, the start position of the section is set. 4, the image of the originally captured image number n _a of when images arranged in chronological order is an image corresponding to the ileocecal region A1. Here, it is assumed that the image (image number n _a ) corresponding to the ileocecal part A1 is not a representative image.

Further, in step S13, when the user selects an image corresponding to the anus A7 from the input unit 30, the anus specifying unit 120 acquires information received by the input unit 30, and selects the image selected by the user from the anus A7. It is specified as an image corresponding to. That is, the end position of the section is set. As shown in FIG. 4, the image of the image number n _r captured at the end of the image group arranged in time series is an image corresponding to the anus A7. Here, it is assumed that the image (image number n _r ) corresponding to the anus A7 is not a representative image.

  In step S <b> 14, the parameter setting unit 130 sets parameters used when the candidate image group setting unit 160 sets a candidate image group in accordance with the input received by the input unit 30. If parameters are set in advance, user input to the input unit 30 is not necessary.

In step S15, the distance ratio calculation unit 150 calculates the number ratio RNm for each representative image. In the following equations, m = 1, 2,... M.
RNm = n _m / (M + 1)
Further, since the sheet number ratio RNm can be approximated to the distance ratio RLm, the distance ratio RLm is calculated as RNm = RLm. FIG. 5 is a schematic diagram showing a distance ratio from the ileocecal part A1 to each representative image. As shown in FIG. 5, the distance ratio RLm of the image corresponding to the ileocecal region A1 of the image number _{n a} is 0%, and the distance ratio RLm of the image corresponding to the anal area A7 of the image number _{n r} 100% is there.

In subsequent step S16, the candidate image group setting unit 160, based on the distance ratio RLm and the parameter set by the parameter setting unit 130, represents a representative image group that is a candidate for the characteristic part image corresponding to the liver curvature A3 and the spleen curve A5. Set. FIG. 6 is a schematic diagram illustrating a representative image group that is a candidate for a feature region image corresponding to liver curvature and a feature region image corresponding to liver curvature. As shown in FIG. 6, the distance ratio RLm first representative image of the representative image group G _A that are candidates for feature region image corresponding to the hepatic flexure A3 was 11%, the distance ratio RLm last representative image 19 %. FIG. 7 is a schematic diagram illustrating a representative image group that is a candidate for a feature part image corresponding to a spleen curve and a feature part image corresponding to a spleen curve. As shown in FIG. 7, the distance ratio RLm first representative image 39% of the representative image group G _B which is a candidate of feature region image corresponding to the splenic flexure A5, distance ratio RLm last representative image 52 %.

  Thereafter, in step S17, for each representative image, the comparison unit 170 sets a parameter (for example, a statistical median value) of the distance ratio RLm of the length from the ileocecal portion A1 to the liver curvature A3 and the splenic curve A5 with respect to the entire length of the large intestine. 15% or 45%) is compared with the distance ratio RLm.

Further, in step S18, the specifying unit 180, based on the comparison result of the comparison unit 170, as shown in FIG. 6, the image number substantially distance ratio RLm located in the center is 15% of the representative image group _{G A n A1} Is specified as a feature part image corresponding to the liver curvature A3. Likewise, certain portions 180, based on the comparison result of the comparison unit 170, as shown in FIG. 7, the representative image of the image number n _B1 substantially distance ratio RLm located in the center of 45% of the representative image group G _B Is identified as a feature part image corresponding to the splenic curve A5.

  As described above, in the image processing apparatus 1 according to the first embodiment, based on the distance ratio RLm calculated by the distance ratio calculation unit 150, the specifying unit 180 corresponds to the liver curvature A3 and the spleen curve A5. In order to identify the image, specify in which part of the digestive tract each acquired image is an image taken, and specify an image corresponding to the characteristic part from the group of images taken in the digestive tract An image processing apparatus capable of

In the first embodiment described above, the specifying unit 180 specifies the representative image of the image number n _A1 having a distance ratio RLm of 15% as a feature part image corresponding to the liver curve A3, and the distance ratio RLm is 45%. Although the representative image of the image number n _B1 is specified as the characteristic part image corresponding to the splenic curve A5, the present invention is not limited to this. FIG. 8 is a schematic diagram illustrating a characteristic part image corresponding to the liver curvature specified by the image processing apparatus according to the modification of the first embodiment. Specifying unit 180, based on the comparison result of the comparison unit 170, as shown in FIG. 8, first representative image number n _A2 distance ratio RLm a representative image captured 11% of the representative image group G _A The image may be specified as a feature part image corresponding to liver curvature A3. FIG. 9 is a schematic diagram illustrating a characteristic part image corresponding to the splenic curvature specified by the image processing apparatus according to the modification of the first embodiment. Specifying unit 180, based on the comparison result of the comparison unit 170, as shown in FIG. 9, the end to the representative of the image number n _B2 of the distance ratio RLm a representative image captured 52% of the representative image group G _B The image may be specified as a characteristic part image corresponding to the splenic curve A5. Further, certain portions 180, in response to user input by the input unit 30 has received, identified respectively feature region image corresponding to the hepatic flexure A3 and splenic flexure A5 from the representative image group G _A and the representative image group G _B May be. Specifically, after the candidate image group setting unit 160 sets the candidate image group, each representative image of the candidate image group is displayed on the display unit 40. Then, the user selects a characteristic part image corresponding to the liver curve A3 and the spleen curve A5 from each representative image of the candidate image group displayed on the display unit 40, and performs a predetermined input to the input unit 30. The specifying unit 180 sets the representative image selected by the user based on the input information received by the input unit 30 as a feature part image corresponding to the liver curve A3 and the spleen curve A5.

In the first embodiment described above, the case where the image corresponding to the ileocecal part A1 of the image number n _a and the image corresponding to the anus A7 of the image number n _r are not representative images has been described. image n _a image and the image number n _r of the case is a representative image, it is possible to calculate the distance ratio RLm as follows.
(1) If the image of the image number _{n a} is a representative image _{RLm = (n m -1) /} M
(2) When the image with the image number n _r is a representative image RLm = n _m / M
(3) When the image number n _a and the image number n _r are representative images RLm = (n _m −1) / M−1
By calculating as described above, the distance ratio RLm can be calculated even if the image with the image number n _a or the image with the image number n _r is a representative image.

(Embodiment 2)
FIG. 10 is a block diagram illustrating an image processing apparatus according to the second embodiment. As illustrated in FIG. 10, the image processing apparatus 1A includes a receiving unit 21A, and the arithmetic unit 100A of the image processing apparatus 1A includes a position and orientation calculating unit 210A. Other configurations are the same as those in the first embodiment, and thus description thereof will be omitted as appropriate.

  The receiving unit 21A includes a plurality of antenna units 21Aa, and sequentially receives radio signals from the capsule endoscope via these antenna units 21Aa. The image acquisition unit 20 of the reception unit 21A selects the antenna with the highest received electric field strength from these antenna units 21Aa, and demodulates the radio signal from the capsule endoscope received through the selected antenna. Perform processing. As a result, the image acquisition unit 20 extracts the image data in the digestive tract of the subject from the wireless signal and stores the image data in the storage unit 50.

  The position and orientation calculation unit 210A calculates the position and orientation of the capsule endoscope in the subject for each image based on the strength of the radio signal received by the plurality of antenna units 21Aa of the reception unit 21A. Information related to the position of the endoscope (hereinafter referred to as position information) and information related to the posture (hereinafter referred to as posture information) are generated and stored in the storage unit 50.

  For the position and orientation of the capsule endoscope, for example, the initial value of the position and orientation is set as appropriate, and the process of calculating the estimated value of the position and orientation by the Gauss-Newton method is performed using the calculated estimated value and the previous estimated value. It can be obtained by iterating until the deviation amount becomes less than a predetermined value (see, for example, JP-A-2007-283001).

  Next, the operation of the image processing apparatus 1A shown in FIG. 10 will be described. FIG. 11 is a flowchart showing the operation of the image processing apparatus according to the second embodiment. As shown in FIG. 11, first, in step S <b> 10, the image processing apparatus 1 </ b> A acquires image data of a series of intra-gastrointestinal images captured in chronological order via the image acquisition unit 20 and records them in the storage unit 50. To do.

  In subsequent step S20, the position and orientation calculation means 210A determines the position and orientation of the capsule endoscope in the subject for each image based on the intensity of the radio signal received by the plurality of antenna units 21Aa of the reception unit 21A. calculate. Furthermore, the position and orientation calculation unit 210A stores position information and orientation information when each image is captured in the storage unit 50 for each image.

  FIG. 12 is a schematic diagram illustrating the movement of the capsule endoscope in the large intestine. In FIG. 12, time t = 0 corresponds to ileocecal part A1, time t = t1 corresponds to liver curvature A3, time t = t2 corresponds to splenic curve A5, and time t = t3 corresponds to anal part A7. Correspond. As shown in FIG. 12, the capsule endoscope 2 moves from the ileocecal part A1 to the anal part A7 in the large intestine. In the following, as shown in FIG. 12, the horizontal direction of the subject is the X-axis direction, the vertical direction of the subject is the Y-axis direction, and the front-back direction of the subject is the Z-axis direction.

  FIG. 13 is a diagram illustrating a temporal change in the position of the capsule endoscope in the large intestine. As shown in FIG. 13, the capsule endoscope 2 moves above the subject and moves in front of the subject in the ascending colon A2. On the other hand, in the ascending colon A2, the movement of the subject of the capsule endoscope 2 in the left-right direction is slight. The capsule endoscope 2 moves in the left-right direction of the subject in the transverse colon A4. On the other hand, in the transverse colon A4, the movement of the subject of the capsule endoscope 2 in the vertical direction and the front-back direction is slight. The capsule endoscope 2 moves below the subject and moves behind the subject in the descending colon A6. On the other hand, in the descending colon A6, the movement of the subject of the capsule endoscope 2 in the left-right direction is slight.

  FIG. 14 is a diagram illustrating the amount of change in the position of the capsule endoscope in the large intestine. FIG. 14 shows the amount of change in the position of the capsule endoscope 2 based on FIG.

  Thereafter, the same processing as in the first embodiment is performed, and in step S21, the comparison unit 170 compares the amount of change in position with a predetermined threshold value. Specifically, the comparing unit 170 compares the amount of change (ΔX, ΔY, ΔZ) of each representative image with the threshold value BX1, threshold value BX2, threshold value BY1, threshold value BY2, threshold value BZ1, threshold value BZ2 shown in FIG. To do.

  In subsequent step S18, the specifying unit 180 has ΔX larger than the threshold BX1 among the candidate image groups set by the candidate image group setting unit 160 based on the comparison result of the comparison unit 170, and ΔY A representative image having a value smaller than the threshold BY1 and ΔZ smaller than the threshold BZ1 is specified as a feature part image corresponding to the liver curvature A3. At this time, when there are a plurality of representative images that satisfy these conditions, the representative image captured at the earliest time may be specified as the feature part image corresponding to the liver curvature A3. Similarly, the specifying unit 180 includes a candidate image group set by the candidate image group setting unit 160, wherein ΔX has a value smaller than the threshold BX2, ΔY has a value smaller than the threshold BY2, and A representative image having a value ΔZ smaller than the threshold value BZ2 is specified as a feature part image corresponding to the splenic curve A5. At this time, when there are a plurality of representative images that satisfy these conditions, the representative image captured at the earliest time may be specified as the characteristic part image corresponding to the splenic curve A5.

  As described above, in the image processing apparatus 1A according to the second embodiment, based on the position calculated by the position and orientation calculation unit 210A, the specifying unit 180 has a characteristic part image corresponding to the liver curve A3 and the spleen curve A5. In order to identify the image, it is possible to identify in which part of the digestive tract each acquired image is an image captured, and to identify an image corresponding to a characteristic part from a group of images captured in the digestive tract. It is an image processing apparatus that can be used.

  In the second embodiment described above, the specifying unit 180 specifies the feature part images corresponding to the liver curvature A3 and the spleen curve A5 based on the position calculated by the position and orientation calculation unit 210A. I can't. The specifying unit 180 may specify the characteristic part images corresponding to the liver curvature A3 and the splenic curve A5 based on the posture calculated by the position and posture calculation unit 210A.

  FIG. 15 is a diagram illustrating a temporal change in the elevation angle of the capsule endoscope in the large intestine. In FIG. 15, time t = 0 corresponds to ileocecal part A1, time t = t11 corresponds to liver curvature A3, time t = t12 corresponds to splenic curve A5, and time t = t13 corresponds to anal part A7. Correspond. In the following, the elevation angle is an in-plane angle along a plane orthogonal to the direction of gravity, and the direction toward the upper side (head side) of the subject is 0 ° and the lower side (leg side) of the subject. The direction is defined as 180 °. Specifically, in the capsule endoscope 2, the elevation angle changes slightly in the ascending colon A2, but the elevation angle changes abruptly in the liver curvature A3. Similarly, in the capsule endoscope 2, the elevation angle changes slightly in the transverse colon A4, but the elevation angle changes abruptly in the splenic curve A5.

  In this configuration, the comparison unit 170 compares the elevation angle with a predetermined threshold value. Then, the specifying unit 180 specifies a representative image having an elevation angle greater than the threshold value BA1 as a feature part image corresponding to the liver curve A3. At this time, when there are a plurality of representative images that satisfy this condition, the representative image captured at the earliest time may be specified as the characteristic part image corresponding to the liver curvature A3. Similarly, the specifying unit 180 specifies a representative image having an elevation angle greater than the threshold value BA2 as a feature part image corresponding to the splenic curve A5. At this time, when there are a plurality of representative images satisfying this condition, the representative image captured at the earliest time may be specified as the feature part image corresponding to the splenic curve A5.

(Embodiment 3)
FIG. 16 is a block diagram illustrating an image processing apparatus according to the third embodiment. As shown in FIG. 16, the image processing apparatus 1B includes an antenna 70B attached to the body surface of the subject 3 swallowing the capsule endoscope 2, and a radio wave intensity receiving unit that receives the radio wave reception intensity from the antenna 70B. 71B, and the calculation unit 100B of the image processing apparatus 1B includes a correlation calculation unit 310B. Other configurations are the same as those in the first embodiment, and thus description thereof will be omitted as appropriate.

  The radio wave intensity receiving unit 71B stores the radio wave reception intensity when each image is captured in the storage unit 50 for each image.

  The correlation calculation unit 310B calculates, for example, a correlation coefficient as the correlation value of the radio wave reception intensity of each antenna 70B received by the radio wave intensity reception unit 71B. However, the correlation value is not limited to the correlation coefficient, and may be a value representing the correlation of the plurality of antennas 70B such as a difference or product of radio wave reception intensity of each antenna 70B.

  Next, the operation of the image processing apparatus 1B shown in FIG. 16 will be described. FIG. 17 is a flowchart showing the operation of the image processing apparatus according to the third embodiment. As shown in FIG. 17, first, in step S <b> 10, the image processing apparatus 1 </ b> B acquires image data of a series of intra-gastrointestinal images taken in time series order via the image acquisition unit 20 and records them in the storage unit 50. To do.

  In subsequent step S30, the radio wave intensity receiving unit 71B receives the signal intensity (radio wave reception intensity) received by the plurality of antennas 70B. Further, the radio wave intensity receiving unit 71B stores the radio wave reception intensity when each image is captured in the storage unit 50 for each image.

  Thereafter, the same processing as in the first embodiment is performed, and in step S18, the specifying unit 180 specifies a feature part image. FIG. 18 is a flowchart showing processing for specifying a feature part image executed by the specifying unit shown in FIG.

  First, in step S31, the radio wave reception intensity of the first antenna closest to the ileocecal part A1 is normalized. FIG. 19 is a diagram illustrating the normalized radio wave reception intensity of the first antenna. As shown in FIG. 19, the radio wave reception intensity of the first antenna decreases as the capsule endoscope 2 advances through the digestive tract and moves from the ileocecal part A1 toward the liver curve A3.

  In subsequent step S32, the radio wave reception intensity of the second antenna closest to the liver curvature A3 is normalized. FIG. 20 is a diagram illustrating the normalized radio wave reception intensity of the second antenna. As shown in FIG. 20, the radio wave reception intensity of the second antenna increases as the capsule endoscope 2 advances through the digestive tract and moves from the ileocecal portion A1 toward the liver curvature A3. Furthermore, the radio wave reception intensity of the second antenna decreases as the capsule endoscope 2 travels through the digestive tract and passes through the liver curve A3, and the capsule endoscope 2 moves away from the liver curve A3.

  In step S33, correlation calculation section 310B sets the maximum calculation range of the correlation coefficient. Specifically, the correlation calculation unit 310B sets the maximum calculation range C in FIGS. 19 and 20 corresponding to the candidate image group (image group having a distance ratio of 11% to 19%) set by the candidate image group setting unit 160. Set as the maximum calculation range.

  In subsequent step S34, correlation calculation section 310B sets a section for calculating a correlation coefficient. Initially, the first representative image captured in the candidate image group is set as the first image, and the first representative image to the second representative image are set as a section for calculating the correlation coefficient.

  In step S35, correlation calculation section 310B calculates a correlation coefficient between the radio wave reception intensity of the first antenna and the radio wave reception intensity of the second antenna in the set section.

  Thereafter, in step S36, correlation calculation section 310B increases the interval for calculating the correlation coefficient. Specifically, the first representative image to the third representative image in the candidate image group are set as a section for calculating the correlation coefficient.

  In step S37, the correlation calculation unit 310B determines whether or not the section to be calculated exceeds the maximum calculation range C. If the interval to be calculated does not exceed the maximum calculation range C (step S37: No), the calculation of the correlation coefficient is repeated while increasing the interval to be calculated. And when the area to calculate exceeds the maximum calculation range C (step S37: Yes), it progresses to the next step. That is, the correlation calculation unit 310B calculates correlation coefficients for all images within the maximum calculation range C. FIG. 21 is a diagram illustrating a correlation coefficient between radio wave reception intensities of two antennas. The correlation calculation unit 310B calculates a correlation coefficient within the maximum calculation range C (the thick broken line in FIG. 21) indicated by the solid line in FIG. Note that the correlation coefficient outside the maximum calculation range C indicated by the broken line in FIG. 21 may not be calculated in actual processing.

  In subsequent step S <b> 38, the specifying unit 180 specifies a representative image having the smallest correlation coefficient calculated by the correlation calculating unit 310 </ b> B as a feature part image from the candidate image group. Thereafter, the operation of the image processing apparatus 1B returns to the main routine, and a series of operations ends.

  Similarly, by selecting the antenna closest to the liver curve A3 as the first antenna and setting the antenna closest to the spleen curve A5 as the second antenna, the feature part image corresponding to the spleen curve A5 is specified. can do.

  As described above, in the image processing device 1B according to the third embodiment, the specifying unit 180 uses the correlation value calculated by the correlation calculation unit 310B to display the characteristic part images corresponding to the liver curvature A3 and the spleen curve A5. In order to specify, it is possible to specify at which part in the digestive tract each acquired image is an image taken, and to specify an image corresponding to a characteristic part from a group of images taken in the digestive tract An image processing apparatus.

  In the third embodiment described above, the configuration in which the antennas 70B are randomly arranged has been described. However, the present invention is not limited to this. FIG. 22 is a diagram illustrating another arrangement example of the antenna. As shown in FIG. 22, only three antennas, that is, the antenna 70Ba close to the ileocecal part A1, the antenna 70Bb close to the liver curve A3, and the antenna 70Bc close to the spleen curve A5 may be arranged. In FIG. 22, the antenna 70Ba is disposed on the back side, and the antenna 70Bb and the antenna 70Bc are disposed on the chest side.

(Embodiment 4)
FIG. 23 is a block diagram illustrating an image processing apparatus according to the fourth embodiment. As illustrated in FIG. 23, the image processing device 1C includes a reception unit 21A, and the calculation unit 100C of the image processing device 1C includes an acceleration information acquisition unit 410C. Other configurations are the same as those in the first embodiment, and thus description thereof will be omitted as appropriate.

  The receiving unit 21A may have the same configuration as that of the second embodiment, but receives an acceleration signal output from an acceleration sensor arranged in the capsule endoscope.

  The acceleration information acquisition unit 410C causes the storage unit 50 to store the acceleration of the capsule endoscope in the subject for each image based on the acceleration signal received by the reception unit 21A.

  Next, the operation of the image processing apparatus 1C shown in FIG. 23 will be described. FIG. 24 is a flowchart showing the operation of the image processing apparatus according to the fourth embodiment. As shown in FIG. 24, first, in step S <b> 10, the image processing apparatus 1 </ b> C acquires a series of intra-gastrointestinal images imaged in chronological order via the image acquisition unit 20 and records them in the storage unit 50. To do.

  In subsequent step S40, the acceleration information acquisition unit 410C stores the acceleration in the X-axis direction of the capsule endoscope in the subject for each image based on the acceleration signals received by the plurality of antenna units 21Aa of the reception unit 21A. Stored in the unit 50.

  FIG. 25 is a diagram illustrating acceleration in the X-axis direction of the capsule endoscope in the large intestine. In FIG. 25, time t = 0 corresponds to the ileocecal part A1, time t = t21 corresponds to the liver curvature A3, and time t = t22 corresponds to the splenic curve A5. As shown in FIG. 25, the capsule endoscope has slight movement in the X-axis direction in the ascending colon A2. In the following transverse colon A4, acceleration is first performed, and then proceeds at a substantially constant speed. Furthermore, when the capsule endoscope 2 reaches the splenic curve A5, the traveling direction is changed, so that the capsule endoscope 2 is once decelerated.

  Thereafter, the same processing as in the first embodiment is performed, and in step S41, the comparison unit 170 compares the acceleration with a predetermined threshold value. Specifically, the comparison unit 170 compares the acceleration ax with the threshold values D1 and D2 shown in FIG.

  In subsequent step S18, the specifying unit 180 selects a representative image having an acceleration ax greater than the threshold value D1 from the candidate image group set by the candidate image group setting unit 160 based on the comparison result of the comparison unit 170. The characteristic part image corresponding to the liver curvature A3 is specified. At this time, when there are a plurality of representative images that satisfy this condition, the representative image captured at the earliest time may be specified as the characteristic part image corresponding to the liver curvature A3. Similarly, the specification unit 180 specifies a representative image having an acceleration ax smaller than the threshold value D2 as a feature part image corresponding to the splenic curve A5 from the candidate image group set by the candidate image group setting unit 160. . At this time, when there are a plurality of representative images satisfying this condition, the representative image captured at the earliest time may be specified as the feature part image corresponding to the splenic curve A5.

  As described above, in the image processing device 1C according to the fourth embodiment, the specifying unit 180 generates the characteristic part images corresponding to the liver curvature A3 and the spleen curve A5 based on the acceleration calculated by the acceleration information acquisition unit 410C. In order to specify, it is possible to specify at which part in the digestive tract each acquired image is an image taken, and to specify an image corresponding to a characteristic part from a group of images taken in the digestive tract An image processing apparatus.

  Note that, as a configuration similar to that of the fourth embodiment, the characteristic part image may be specified based on the cumulative number of captured images. First, the capsule endoscope includes a frame rate control means, and the frame rate increases as the speed increases, and the frame rate decreases as the speed decreases. FIG. 26 is a diagram illustrating a change over time in the cumulative number of captured images of the capsule endoscope. In FIG. 26, time t = 0 corresponds to the ileocecal part A1, time t = t31 corresponds to the liver curvature A3, time t = t32 corresponds to the splenic curve A5, and time t = t33 corresponds to the anal part A7. Correspond. First, in the ascending colon A2, since the speed of the capsule endoscope is low, the frame rate of the capsule endoscope is low, and the increase in the number N of accumulated captured images is moderate. Subsequently, in the transverse colon A4, since the speed of the capsule endoscope is high, the frame rate of the capsule endoscope is high, and the increase in the number N of accumulated captured images increases. In the descending colon A6, since the speed of the capsule endoscope is low, the frame rate of the capsule endoscope is low, and the increase in the cumulative number of captured images N is moderate.

  The specifying unit 180 specifies a representative image in which the amount of change in the cumulative number of captured images N is greater than a predetermined threshold from the candidate image groups set by the candidate image group setting unit 160 as a feature part image corresponding to the liver curvature A3. To do. Similarly, the specifying unit 180 selects a representative image corresponding to the splenic curve A5 from the candidate image group set by the candidate image group setting unit 160, with a change amount of the cumulative number of captured images N being smaller than a predetermined threshold. Specify as an image.

  Similarly, the specifying unit 180 may specify the feature part image based on the time interval between the captured images. FIG. 27 is a diagram illustrating a temporal change in the imaging time interval of the capsule endoscope. In FIG. 27, time t = 0 corresponds to the ileocecal part A1, time t = t41 corresponds to the liver curvature A3, and time t = t42 corresponds to the splenic curve A5. First, in the ascending colon A2, since the speed of the capsule endoscope is low, the frame rate of the capsule endoscope is low, and the time interval Δt between captured images is large. Subsequently, in the transverse colon A4, since the speed of the capsule endoscope is high, the frame rate of the capsule endoscope is high, and the time interval Δt between the captured images is small. In the descending colon A6, since the speed of the capsule endoscope is low, the frame rate of the capsule endoscope is low, and the time interval Δt between captured images is large.

  The specifying unit 180 selects a representative image having a time interval Δt between captured images smaller than the threshold E1 from the candidate image group set by the candidate image group setting unit 160 as a feature part image corresponding to the liver curvature A3. Identify. Similarly, the specifying unit 180 is a feature corresponding to the splenic curve A5, among the candidate image groups set by the candidate image group setting unit 160, for a representative image whose time interval Δt between captured images is larger than the threshold value E2. It is specified as a part image.

(Embodiment 5)
FIG. 28 is a block diagram illustrating an image processing apparatus according to the fifth embodiment. As illustrated in FIG. 28, the control unit 10 of the image processing apparatus 1D includes a display control unit 11D, and the calculation unit 100D of the image processing apparatus 1D includes a part classification unit 510D. Other configurations are the same as those in the first embodiment, and thus description thereof will be omitted as appropriate.

  The display unit 40 includes a main display and a sub display. A selected image selected by the user from the image group is displayed on the main display of the display unit 40. A configuration diagram of the digestive tract is displayed on the sub-display of the display unit 40.

  The display control unit 11D displays a selection image determined based on the position corresponding to the characteristic part image specified by the specifying unit 180 and the distance ratio on the configuration diagram of the digestive tract displayed on the sub-display of the display unit 40. An index indicating the corresponding position is displayed.

Part classification unit 510D provides a distance ratio to images other than the representative image based on the distance ratio calculated by distance ratio calculation unit 150. FIGS. 29-31 is a figure for demonstrating operation | movement of a site | part classification part. First, as shown in FIG. 29, it is assumed that there are images with image numbers n ₁₁ , n ₁₂ , and n ₁₃ that are not representative images between image numbers n ₁ and n ₂ that are representative images. . The distance ratio of the image number _{n 1} is 0%, the distance ratio of the image number _{n 2} is assumed to be 10%. At this time, as shown in FIG. 30, the part classification unit 510D sets all the distance ratios of the image number n ₁₁ , the image number n ₁₂ , and the image number n ₁₃ to 0%. However, as shown in FIG. 31, the part classification unit 510D is distributed evenly distance ratio, the distance ratio of the image number _{n 11} 2.5%, the distance ratio of the image number _{n 12} 5%, the image number _{n 13} The distance ratio may be 7.5%.

  Next, an image displayed on the sub display of the display unit 40 will be described in detail. FIG. 32 is a diagram illustrating display on the display unit when the selected image is the ascending colon. As shown in FIG. 32, a configuration diagram of the digestive tract (large intestine) is displayed on the sub-display of the display unit 40. Then, on the configuration diagram, the display control unit 11D causes the indicator M1 corresponding to the ileocecal portion A1, the indicator M2 corresponding to the liver curvature A3, the indicator M3 corresponding to the splenic curvature A5, and the indicator M4 corresponding to the selected image. Are displayed. Further, the distance ratio display part F displays the distance ratio from the ileocecal part A1 to the selected image by a color-coded bar and a numerical value (11). Further, since the selected image is in the ascending colon A2, an area corresponding to the ascending colon A2 is highlighted.

  FIG. 33 is a diagram illustrating display on the display unit when the selected image is the transverse colon. FIG. 34 is a diagram illustrating display on the display unit when the selected image is the descending colon. Similarly to FIG. 32, each index and the distance ratio (24, 80) are displayed on the sub-display of the display unit 40, and the transverse colon A4 or the descending colon A6 each having a selected image is highlighted.

  As described above, according to the fifth embodiment, various indicators are displayed on the display unit 40 corresponding to the distance ratio and the characteristic part image, so that the user can intuitively grasp the image currently being viewed. It's easy to do.

(Embodiment 6)
FIG. 35 is a block diagram showing an image processing apparatus according to the sixth embodiment. As shown in FIG. 35, the calculation unit 100E of the image processing apparatus 1E includes a duodenum specifying unit 610E. Then, the specifying unit 180 of the image processing apparatus 1E specifies a feature part image corresponding to the ileocecal part A1 as a feature part. Other configurations are the same as those in the first embodiment, and thus description thereof will be omitted as appropriate.

  FIG. 36 is a schematic diagram showing the structures of the small and large intestines in the digestive tract of a subject. As shown in FIG. 36, the small and large intestines in the digestive tract of the subject include the pylorus A11 at the boundary between the stomach and the small intestine, the duodenum, the jejunum, the ileum, and the like along the direction in which the contents proceed. It consists of the small intestine A12, the ileocecal part A1 which is the boundary between the small intestine A12 and the large intestine A13, the large intestine A13, and the anal part A7 which is the exit of the large intestine A13.

  Similarly to the ileocecal part specifying unit 110, the duodenum specifying unit 610E specifies an image corresponding to the ileocecal part A1 from the image group in the digestive tract.

  The parameter setting unit 130 sets parameters used when the candidate image group setting unit 160 sets a candidate image group in accordance with a user input to the input unit 30. Here, it is statistically known that the distance ratio from the pyloric part A11 to the ileocecal part A1 with respect to the entire length of the small intestine A12 and the large intestine A13 is 75% to 82%. Based on this statistical value, the user inputs parameters so that a representative image having a distance ratio of 75% to 82% is a candidate image group that is a candidate for a characteristic part image corresponding to the ileocecal part A1.

  The distance ratio calculation unit 150 calculates a distance ratio that is a ratio with respect to the distance from the pylorus A11 to the anus A7 for each representative image.

  As described above, the comparison unit 170 compares the statistical value (75% to 82%) with the distance ratio associated with each representative image. Alternatively, for example, the median value of 79% of the statistical values is compared with the distance ratio associated with each representative image.

  Based on the comparison result of the comparison unit 170, the candidate image group setting unit 160 sets a representative image having a distance ratio calculated by the distance ratio calculation unit 150 of 75% to 82% as a candidate image group corresponding to the ileocecal part A1. Set.

  Based on the comparison result of the comparison unit 170, the identification unit 180 identifies a representative image having a distance ratio calculated by the distance ratio calculation unit 150 of 79% as a feature part image corresponding to the ileocecal part A1.

  As in the sixth embodiment, if the statistical value is clear, the corresponding feature part image can be specified not only for the ileocecal part A1, liver curvature A3, and spleen curve A5, but also for any feature part. In the sixth embodiment, an image corresponding to the boundary between the duodenum and the jejunum or the boundary between the jejunum and the ileum may be specified as the feature part image.

Note that the image in the above embodiment, a section from the image corresponding to the ileocecal region A1 of the image number n _a to the representative image of the first (image number n _1), and from the representative image of the last (image number n _M) The distance ratio RLm is calculated in consideration of the section to the image corresponding to the anal part A7 of the number n _r , but may be calculated by excluding these sections. 37 and 38 are schematic diagrams showing the distance ratio from the ileocecal part A1 to each image. First, as shown in FIG. 37, when calculating by excluding the section from the representative image of the image number n _M to the image of the image number n _r, the distance ratio RLm of the representative image of the image number n _M is 100% . Therefore, the distance ratio RLm can be calculated as follows.
RLm = n _m / M
However, when the image of image number na is _a representative image, it can be calculated as follows.
RLm = (n _m −1) / M−1
In this case, since the calculated excluding the section up image of the image number n _r from the representative image of the image number n _M, the image of the image number n _r is independent of whether or not the representative image The formula is the same. Next, as shown in FIG. 38, calculation is performed by excluding the section from the image with the image number n _a to the representative image with the image number n ₁ and the section from the representative image with the image number n _{M to} the image with the image number n _r. to case, next distance ratio RLm of the representative image of the image number _{n 1} is 0%, the distance ratio RLm of the representative image of the image number _{n M} is 100%. Therefore, the distance ratio RLm can be calculated as follows.
RLm = (n _m −1) / M−1
In this case, is calculated by excluding the section from the image number n _a of the image segment of the representative image of the image number n _1, and a representative image of the image number n _M to the image of the image number n _r Therefore, the equation is the same regardless of whether the images of the image number n _a and the image number n _r are representative images. Further, the image number n _a and the image number n _r may always be representative images.

  Further effects and modifications can be easily derived by those skilled in the art. Accordingly, the broader aspects of the present invention are not limited to the specific details and representative embodiments shown and described above. Accordingly, various modifications can be made without departing from the spirit or scope of the general inventive concept as defined by the appended claims and their equivalents.

1, 1A, 1B, 1C, 1D, 1E Image processing device 2 Capsule endoscope 3 Subject 10 Control unit 11D Display control unit 20 Image acquisition unit 21A Reception unit 21Aa Antenna unit 30 Input unit 40 Display unit 50 Storage unit 70B , 70Ba, 70Bb, 70Bc Antenna 71B Radio wave intensity receiving unit 100, 100A, 100B, 100C, 100D, 100E Arithmetic unit 110 ileocecal part specifying unit 120 anal part specifying unit 130 parameter setting unit 140 representative image setting unit 150 distance ratio calculating unit 160 Candidate Image Group Setting Unit 170 Comparison Unit 180 Identification Unit 210A Position and Posture Calculation Unit 310B Correlation Calculation Unit 410C Acceleration Information Acquisition Unit 510D Site Classification Unit 610E Duodenum Identification Unit A1 Ileal Blindness A2 Ascending Colon A3 Liver Curve A4 Transverse Colon A5 Spleen Curve A6 Descending colon A Anal A11 pylorus A12 small intestine A13 colon C maximum calculation range D1, D2, E1, E2 threshold F distance ratio display section G _{A, G B} representative image group M1, M2, M3, M4 index N cumulative captured image number n _1, n ₂ , n _a , n _r , n ₁₁ , n ₁₂ , n ₁₃ , n _M , n A ₁ , n A ₂ , n _{B 1} , n _{B 2} image number

Claims (19)

A representative image setting unit for setting a representative image representing a group of images captured by a capsule endoscope introduced into the digestive tract of a subject;
A count is assigned to each of the representative images included in a predetermined section in the digestive tract, and based on the count and the total number of the representative images included in the section, A distance ratio calculation unit that calculates a ratio of a length from a start position to the representative image for each representative image;
An image processing apparatus comprising:   The image processing apparatus according to claim 1, further comprising: a specifying unit that specifies at least one feature part image obtained by imaging the feature part of the digestive tract from the representative image based on the ratio. A comparison unit that compares the ratio of the parameter regarding the position of the characteristic site in the digestive tract with the ratio for each representative image;
The image processing apparatus according to claim 2, wherein the specifying unit specifies at least one representative image as the feature part image based on a comparison result of the comparison unit. A candidate image group setting unit that sets a candidate image group that is a set of the representative images included in a partial section of the section, the ratio of which is determined based on the parameter based on the comparison result of the comparison unit Further comprising
Based on the comparison result of the comparison unit, the specifying unit selects at least one representative image in which the parameter and the ratio match or the parameter and the ratio are close to each other in the candidate image group. The image processing apparatus according to claim 3, wherein the image is specified as the feature part image. A storage unit for storing the position of the capsule endoscope when each image of the image group is captured;
The image processing apparatus according to claim 4, wherein the specifying unit specifies at least one feature part image from the candidate image group based on the position. The comparison unit compares the amount of change in the position with a predetermined threshold,
The characteristic part image is
The amount of change in the position of the subject in the left-right direction has a value greater than a predetermined threshold, and
The amount of change in the vertical position of the subject has a value smaller than a predetermined threshold, and
The representative image having a value of a change amount of the position in the front-rear direction of the subject smaller than a predetermined threshold;
Or
The amount of change in the left-right direction of the subject has a value smaller than a predetermined threshold, and
The amount of change in the vertical position of the subject has a value smaller than a predetermined threshold, and
The representative image having a value of a change amount of the position in the front-rear direction of the subject smaller than a predetermined threshold;
The image processing apparatus according to claim 5, wherein: A storage unit for storing the posture of the capsule endoscope when each image of the image group is captured;
The image processing apparatus according to claim 4, wherein the specifying unit specifies at least one feature part image from the candidate image group based on the posture. The comparison unit compares the elevation angle of the capsule endoscope with a predetermined threshold in the posture,
The characteristic part image is
The image processing apparatus according to claim 7, wherein the elevation image is the representative image having a value larger than a predetermined threshold value. A storage unit that stores, for each image, the radio wave reception intensity when each image of the image group is captured;
The image processing apparatus according to claim 4, wherein the specifying unit specifies at least one feature part image from the candidate image group based on the radio wave reception intensity. The storage unit stores a first radio wave reception intensity and a second radio wave reception intensity when each image of the image group is captured,
A correlation calculation unit for calculating a correlation value between the first radio wave reception intensity and the second radio wave reception intensity;
The said specific | specification part specifies the said representative image in which the said correlation value calculated by the said correlation calculation part has the smallest value from the said candidate image group as the said characteristic part image. Image processing apparatus. A storage unit for storing acceleration of the capsule endoscope when each image of the image group is captured;
The image processing apparatus according to claim 4, wherein the specifying unit specifies at least one feature region image from the candidate image group based on the acceleration. The comparison unit compares the acceleration in the left-right direction of the subject with the predetermined threshold among the accelerations,
The characteristic part image is
The representative image in which the acceleration in the left-right direction has a value larger than a predetermined threshold;
Or
The representative image in which the acceleration in the left-right direction has a value smaller than a predetermined threshold;
The image processing apparatus according to claim 11, wherein: A storage unit that stores a frame rate when each image of the image group is captured;
The image processing apparatus according to claim 4, wherein the specifying unit specifies at least one feature part image from the candidate image group based on the frame rate. The comparison unit compares the frame rate with a predetermined threshold,
The characteristic part image is
The representative image, wherein the frame rate has a value greater than a predetermined threshold;
Or
The representative image, wherein the frame rate has a value smaller than a predetermined threshold;
The image processing apparatus according to claim 13, wherein: A storage unit that stores a time at which each image of the image group is captured;
The image processing apparatus according to claim 4, wherein the specifying unit specifies at least one feature region image from the candidate image group based on the time interval between the images. The comparison unit compares the time interval between the images with a predetermined threshold,
The characteristic part image is
The representative image having a value in which the time interval between the images is smaller than a predetermined threshold;
Or
The representative image having a value in which the time interval between the images is larger than a predetermined threshold;
The image processing apparatus according to claim 15, wherein:   An indicator representing a position corresponding to the selected image based on the ratio on the configuration diagram is displayed on a display unit that displays the selected image selected from the image group and the configuration diagram of the digestive tract. The image processing apparatus according to claim 1, further comprising a display control unit. A representative image setting step for setting a representative image representative of a group of images captured by a capsule endoscope introduced into the digestive tract of a subject;
A count is assigned to each of the representative images included in a predetermined section in the digestive tract, and based on the count and the total number of the representative images included in the section, A distance ratio calculating step of calculating a ratio of the length from a start position to the representative image for each representative image;
A method for operating an image processing apparatus, comprising: A representative image setting step for setting a representative image representative of a group of images captured by a capsule endoscope introduced into the digestive tract of a subject;
A count is assigned to each of the representative images included in a predetermined section in the digestive tract, and based on the count and the total number of the representative images included in the section, A distance ratio calculating step of calculating a ratio of the length from a start position to the representative image for each representative image;
An image processing program for causing a computer to execute.

Images