JP2015131057A - Image processing device and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To extract an area capturing a target region from a medical dynamic image with a high degree of precision.SOLUTION: A two-dimensional distribution of a statistical value in a time direction related to a feature value of a pixel value for each image area and/or a pixel value for each pixel is calculated for a plurality of frame images and/or a plurality of frame images in which the edge is emphasized. With regard to a first frame image, a first target region related to a target region is detected based on a first target specified region and a first unspecified region detected using template matching. With regard to a second frame image, a second target specified region and a second unspecified region are detected using template matching with a partial image corresponding to the template in the template matching for the first frame image as a template. A second target region related to the target region of the second frame image is detected based on a second target specified region, and an unspecified region including the second unspecified region and a third unspecified region based on the two-dimensional distribution of the statistical value.

Description

  The present invention relates to an image processing apparatus and a program.

  In a medical field, an image that captures an affected part included in a viscera, a skeleton, or the like is acquired by imaging using X-rays (also referred to as X-ray imaging), and various examinations and diagnoses using the image are performed.

  In recent years, by applying digital technology, an image (also referred to as an X-ray dynamic image) that captures the motion of an affected area by X-ray imaging can be obtained relatively easily. For example, an X-ray dynamic image capturing a structure including a part (also referred to as a target part) to be inspected and diagnosed is acquired by photographing using a semiconductor image sensor such as an FPD (flat panel detector). Thereby, an X-ray dynamic image can be acquired with a relatively simple configuration without using a contrast agent. Further, it is possible to make a diagnosis by analyzing the movement of the target part, which cannot be performed by the diagnosis using the still image obtained by the conventional X-ray imaging.

  For example, quantitative analysis of the state of ventilation and blood flow (also referred to as pulmonary blood flow) in the lung field by analyzing the change in concentration in the region related to the lung field in an X-ray dynamic image capturing the movement of the chest Is recognized. Specifically, for example, an abnormal portion of ventilation is identified from a quantitative change regarding the state of ventilation, and chronic obstructive disease (COPD) or the like can be recognized. Further, for example, an abnormal portion of the pulmonary blood flow can be identified from a quantitative change regarding the state of the pulmonary blood flow, and pulmonary embolism or the like can be recognized.

  However, in order to carry out analysis and diagnosis related to the movement of the target part in the diagnosis using the X-ray dynamic image, it is necessary to accurately extract a region that captures the target part (also referred to as the target region) from the X-ray dynamic image. is there.

  As a technique for extracting a target region from an X-ray image that captures an affected area, for example, a region (also referred to as a lung field region) that captures a lung field as a target region by binarization processing using a certain threshold value, labeling, or the like. Techniques for extraction have been proposed (for example, Patent Documents 1 and 2). Further, a technique has been proposed in which a boundary of a predetermined layer constituting the eye part is detected from a tomographic image of the eye part being a subject, and the normal structure of the layer constituting the subject is accurately estimated (for example, Patent Documents). 3 etc.). In this technique, for example, prior knowledge such as a probability atlas or a statistical atlas is used to extract a region such as a graph cut, thereby detecting a boundary line between structures.

JP 09-035043 A JP-A-11-151232 JP 2013-138963 A

  However, in capturing an X-ray dynamic image that captures the target region, the exposure amount required for capturing each frame is kept low so that the exposure amount of the affected part due to X-rays does not increase. It is easy to deteriorate by. Therefore, for example, in the techniques of Patent Documents 1 and 2 in which a simple binarization process is used to extract a lung field region from an X-ray image, a target region such as a lung field region or the like is affected by noise or the like. Is difficult to extract accurately.

  Further, in the technique of Patent Document 3, a tomographic image as a still image is a processing target, and is generally generated from information such as the shape of a structure of a plurality of people and differences in shades so as to be compatible with everyone. Probabilistic atlases are used. For this reason, for example, a part whose shape and position change suddenly with the passage of time, such as a lung field including the diaphragm and the heart, and a part having an unexpected shape in which a part of the structure is removed by surgery or the like If the subject is a subject, it is difficult to accurately extract a target region such as a lung field region from an image.

  The present invention has been made in view of the above problems, and provides a technique capable of accurately extracting a region capturing a target region from a medical dynamic image capturing the movement of a structure in the body. With the goal.

  In order to solve the above problem, an image processing apparatus according to a first aspect configures the dynamic image and an acquisition unit that acquires a medical dynamic image in which a body structure is captured by X-ray irradiation. A feature that indicates a feature of a pixel value for each image region including a plurality of pixels with respect to at least one of a plurality of edge-enhanced frame images in which edges of the structure in the plurality of frame images are respectively emphasized and the plurality of frame images A calculation unit that calculates a two-dimensional distribution of statistical values in the time direction related to at least one of a value and a pixel value for each pixel, and a template matching is used for a first frame image of the plurality of frame images, A first target definite region in which at least a part of the part is captured and a first unconfirmed region in which a part of the target part may be captured are detected. A first detection unit that detects a first target region that extends from the inside of the target part of the first frame image to an outline based on information on the first detection unit and the first target determination region and the first unconfirmed region. In the template matching for the first frame image, the second frame image is different from the first frame image among the plurality of frame images. A second target fixed region in which at least a part of the target portion is captured using template matching using a partial image related to a corresponding region corresponding to the template in the first frame image, and the target portion; Based on a process for detecting a second undefined region in which a part of the statistical value may be captured and a two-dimensional distribution of the statistical value And a process of detecting a third unconfirmed region in which a part of the second unconfirmed region may be captured, and the second detection unit performs the second target confirmed region, the second unconfirmed region, and the third Based on the information on the undetermined area including the undetermined area, a second target area extending from the inside of the target part to the contour in the second frame image is detected.

  The image processing device according to the second aspect is the image processing device according to the first aspect, and the statistical value in the time direction includes a degree of dispersion in the time direction.

  An image processing apparatus according to a third aspect is the image processing apparatus according to the second aspect, wherein the distribution degree includes at least one of variance, standard deviation, and variation coefficient.

  The image processing device according to the fourth aspect is the image processing device according to the first aspect, wherein the time direction statistical value includes a time direction representative value.

  An image processing apparatus according to a fifth aspect is the image processing apparatus according to the fourth aspect, wherein the representative value includes a value related to a difference between a maximum value and a minimum value.

  An image processing apparatus according to a sixth aspect is the image processing apparatus according to the fourth or fifth aspect, wherein the representative value is at least one of the feature value for each image region and the pixel value for each pixel. The value relating to the frequency occupancy in the range of the preset width with the mode as the reference is included.

  An image processing device according to a seventh aspect is the image processing device according to any one of the first to sixth aspects, wherein the feature value is an average value, median value, mode value, maximum value of pixel values. It includes at least one representative value of a value and a minimum value.

  An image processing device according to an eighth aspect is the image processing device according to any one of the first to seventh aspects, wherein the calculation unit includes the plurality of edge-enhanced frame images and the plurality of frame images. At least one of the first statistical value and the second statistical value in the time direction is calculated for at least one of the feature value for each image area and the pixel value for each pixel, and a predetermined rule is followed. Two-dimensional distribution related to the third statistical value obtained by multiplying the first statistical value by each coefficient and the second statistical value obtained by multiplying the second statistical value by the coefficient according to the rule. A two-dimensional distribution of the statistical values is calculated by a combination with a dimensional distribution.

  An image processing device according to a ninth aspect is the image processing device according to any one of the first to eighth aspects, wherein the first detection unit is configured to apply the target to the first and second frame images. Each non-target defined area in which a part is not captured is detected.

  An image processing apparatus according to a tenth aspect is the image processing apparatus according to any one of the first to ninth aspects, wherein the first detection unit is preset in the two-dimensional distribution of the statistical values. A region satisfying the set condition is detected as the third undefined region.

  An image processing device according to an eleventh aspect is the image processing device according to the tenth aspect, in which the first detection unit has the statistical value of the two-dimensional distribution of the statistical value and a preset threshold value. Is detected as the third undetermined region.

  An image processing apparatus according to a twelfth aspect is executed by a control unit included in the information processing apparatus, so that the information processing apparatus functions as an image processing apparatus according to any one of the first to eleventh aspects. It is a program to let you.

  Since the image processing apparatus according to any one of the first to eleventh aspects can obtain an undetermined region wider than the undetermined region corresponding to the detection result of template matching for the second frame image, the internal structure A region in which the target part is captured can be accurately extracted from a medical dynamic image capturing the movement of an object.

  The image processing apparatus according to any of the second to sixth aspects can easily calculate a two-dimensional distribution of statistical values related to temporal changes in a structure captured by a dynamic image.

  According to the image processing device of the sixth aspect, it is possible to reduce the influence of the difference in shading between the plurality of frame images on the statistical value.

  According to the image processing device of the seventh aspect, the feature value indicating the feature of the pixel value for each image region can be easily calculated.

  According to the image processing apparatus of the eighth aspect, a two-dimensional distribution of statistical values that well reflects a temporal change of a structure captured by a dynamic image by combining a plurality of types of statistical values in the time direction. It can be easily calculated.

  With the image processing device according to any of the tenth and eleventh aspects, a wider undetermined region corresponding to the temporal change of the structure captured by the dynamic image can be obtained.

  According to the program according to the twelfth aspect, an effect similar to that of the image processing apparatus according to the first to eleventh aspects is obtained.

It is a block diagram which shows the structure of the imaging | photography system which concerns on one Embodiment. It is a block diagram which shows the functional structure implement | achieved in an image processing apparatus. It is a flowchart which shows the flow of an object area | region extraction process. It is a figure which shows an example of the dynamic image obtained by X-ray imaging. It is a figure for demonstrating the characteristic of the frame image which comprises a dynamic image. It is a figure for demonstrating the characteristic of the frame image which comprises a dynamic image. It is a figure for demonstrating the characteristic of the frame image which comprises a dynamic image. It is a figure for demonstrating the process which correct | amends a frame image. It is a figure for demonstrating the process which correct | amends a frame image. It is a figure for demonstrating the process which correct | amends a frame image. It is a figure for demonstrating the process which correct | amends a frame image. It is a figure for demonstrating the process which correct | amends a frame image. It is a figure for demonstrating the process which correct | amends a frame image. It is a figure for demonstrating the process which correct | amends a frame image. It is a figure for demonstrating the process which correct | amends a frame image. It is a figure for demonstrating the process which correct | amends a frame image. It is a figure for demonstrating the process which correct | amends a frame image. It is a block diagram which shows the functional structure of a statistical value distribution calculation part. It is a figure which illustrates the periodic change of the position of the diaphragm by respiration. It is a figure which illustrates the observation point of the change of the pixel value in the lung field and the lung field periphery. It is a figure which illustrates the observation point of the change of the pixel value in the lung field and the lung field periphery. It is a figure which illustrates change of a pixel value according to progress of time in an observation point. It is a figure which illustrates the pixel as a calculation object of the statistical value in a frame image. It is a figure which illustrates the image area | region as a calculation object of the statistical value in a frame image. It is a figure which shows an example of the statistical value distribution which concerns on a dynamic image. It is a figure for demonstrating the production | generation method of the statistical value distribution which concerns on a dynamic image. It is a figure for demonstrating the production | generation method of the statistical value distribution which concerns on a dynamic image. It is a figure for demonstrating the production | generation method of the statistical value distribution which concerns on a dynamic image. It is a figure which shows an example of the statistical value distribution which concerns on a dynamic image. It is a figure for demonstrating the production | generation method of the statistical value distribution which concerns on a dynamic image. It is a figure for demonstrating the production | generation method of the statistical value distribution which concerns on a dynamic image. It is a figure for demonstrating the production | generation method of the statistical value distribution which concerns on a dynamic image. It is a figure for demonstrating the production | generation method of the statistical value distribution which concerns on a dynamic image. It is a figure for demonstrating the production | generation method of the statistical value distribution which concerns on a dynamic image. It is a figure which illustrates the frame image used as the object of rough detection processing. It is a figure which shows typically an example of the detection result by a rough detection process. It is a block diagram which shows the functional structure of a rough detection part. It is a figure which shows typically an example of the change of the size in the vertical direction of a lung field. It is a figure which shows typically an example of the change of the size in the vertical direction of a lung field. It is a figure which shows typically an example of the change of the size in the vertical direction of a lung field. It is a figure which shows typically an example of the individual difference in the shape of a diaphragm and a heart. It is a figure which shows typically an example of the individual difference in the shape of a diaphragm and a heart. It is a figure which shows typically an example of the individual difference in the shape of a diaphragm and a heart. It is a figure which shows typically an example of the individual difference in the shape of a diaphragm and a heart. It is a figure for demonstrating the production | generation method of a template. It is a figure for demonstrating the production | generation method of a template. It is a figure for demonstrating the production | generation method of a template. It is a figure for demonstrating the production | generation method of a template. It is a figure for demonstrating the production | generation method of 1st partial division information. It is a figure for demonstrating the production | generation method of 1st partial division information. It is a figure for demonstrating the production | generation method of 1st partial division information. It is a figure for demonstrating the production | generation method of 1st partial division information. It is a figure for demonstrating the production | generation method of 1st partial division information. It is a figure which shows typically an example of 1st area | region division information. It is a figure for demonstrating the production | generation method of a self-template. It is a figure for demonstrating the production | generation method of a self-template. It is a figure which shows an example of a self-template. It is a figure which illustrates two-dimensional distribution of the weighting coefficient with respect to a self template. It is a figure which illustrates the detection result of the precision detection process corresponding to a self-template. It is a figure which illustrates the 2nd partial division information according to a self template. It is a figure which illustrates the initial division information determined using a self template. It is a figure which illustrates the 3rd undecided field obtained from two-dimensional distribution of a statistics value. It is a figure which shows typically an example of area | region division information which considered the 3rd undecided area | region. It is a figure which illustrates the frame image used as the object of precise detection processing. It is a figure which shows typically an example of the detection result by a precision detection process. It is a figure which shows an example of the statistical value distribution about the dynamic image which concerns on one modification. It is a figure which shows an example of the statistical value distribution about the dynamic image which concerns on one modification. It is a figure which shows an example of the statistical value distribution about the dynamic image which concerns on one modification.

  Hereinafter, an embodiment of the present invention will be described with reference to the drawings. In the drawings, parts having the same configuration and function are denoted by the same reference numerals, and redundant description is omitted in the following description. Further, the drawings are schematically shown, and the mutual relationship between the sizes and positions of the images respectively shown in the different drawings is not necessarily accurately described and can be appropriately changed. 5 and 17, an XY coordinate system is attached in which the upper left of the image is the origin, the right direction is the X direction, and the lower direction is the Y direction.

<(1) Overview of shooting system>
FIG. 1 is a block diagram illustrating a configuration of an imaging system 100 according to an embodiment. The imaging system 100 captures a dynamic image in which a structure in the body is captured by irradiation of radiation using a structure in the body of an animal including a human as a subject, and performs various image processing on the dynamic image. In addition, as a structure in a body, skeletons, such as an organ contained in the chest area | regions, such as a lung, and indirect, are mentioned, for example.

  In the present embodiment, the imaging system 100 is a system for the human body, and a part that is a main target of imaging (also referred to as a target part) is a person who is a subject of inspection (also referred to as a subject) M1. The lung field. In this embodiment, the radiation is X-rays, and a dynamic image (also referred to as an X-ray dynamic image) is used to photograph the human body of the subject M1 for medical treatment or medicine, or the human body of the subject M1. The measurement result is an image (also called a medical dynamic image).

  As shown in FIG. 1, the imaging system 100 includes an imaging device 1, an imaging control device (imaging console) 2, and an image processing device (diagnosis console) 3. The imaging device 1 and the imaging control device 2 and the imaging control device 2 and the image processing device 3 are connected to each other via a communication line so as to be able to transmit and receive data. The communication line connecting the image capturing apparatus 1 and the image capturing control apparatus 2 may be a wired line such as a communication cable or a wireless line. Further, the communication line connecting the imaging control apparatus 2 and the image processing apparatus 3 may be either a wired line or a wireless line, and may be a network line NTW such as a LAN (Local Area Network) line.

  In the imaging system 100, for example, a medical image format and a communication protocol between devices that handle images conform to a standard called DICOM (Digital Image and Communications in Medicine).

<(2) Shooting device>
The imaging apparatus 1 is configured by, for example, an apparatus (also referred to as an X-ray imaging apparatus) that acquires a dynamic image for medical use in which a structure inside the body is captured by irradiation with X-rays as radiation. In the imaging device 1, for example, imaging (also referred to as dynamic imaging) is performed on the dynamics of the structure in the chest of the subject M <b> 1 associated with breathing or the like. In the dynamic imaging, for example, a two-dimensional distribution of X-rays transmitted through the subject M1 is detected in time sequence while X-rays are repeatedly irradiated to the chest of the subject M1. Thereby, a plurality of images (also referred to as frame images) are acquired in time sequence. These series of a plurality of frame images show how the physical state (geometric shape, blood flow concentration, etc.) in the region including the lung of the subject M1 (also referred to as the lung field region) changes over time. A dynamic image (moving image) captured sequentially is constructed.

  As shown in FIG. 1, an imaging apparatus 1 includes an irradiation unit (X-ray source) 11, an irradiation control device 12, an imaging unit (X-ray detection unit) 13, a reading control device 14, a cycle detection sensor 15, and a cycle detection device. 16 is provided.

  The irradiation unit 11 irradiates the subject M1 with X-rays according to the control of the irradiation control device 12. The irradiation control device 12 is connected to the imaging control device 2 and controls the irradiation unit 11 based on information indicating the X-ray irradiation conditions input from the imaging control device 2.

  The imaging unit 13 is configured by, for example, a semiconductor image sensor such as an FPD (flat panel detector), and digitally radiates X-rays irradiated from the irradiation unit 11 to the subject M1 and transmitted through the subject M1. Convert to signal.

  The reading control device 14 is connected to the imaging control device 2 and controls the switching unit of each pixel of the imaging unit 13 based on the reading condition input from the imaging control device 2. Thereby, the reading control device 14 sequentially reads digital signals corresponding to the charges accumulated in the respective pixels of the imaging unit 13, and corresponds to the two-dimensional intensity distribution of the X-rays received on the imaging unit 13. Image data is acquired. Then, the reading control device 14 outputs the acquired image data (also simply referred to as an image) to the imaging control device 2 as a frame image constituting a dynamic image. The reading conditions may be, for example, the frame rate, pixel size, image size, and the like. The frame rate may be the number of frame images acquired per second.

  Here, the irradiation control device 12 and the reading control device 14 are electrically connected to each other, and by transmitting and receiving a synchronization signal to each other, the irradiation operation of the X-ray by the irradiation unit 11 and the reading control device 14 are performed. Synchronize with image data reading operation.

  The cycle detection sensor 15 is configured by, for example, a digital camera or the like, and acquires a dynamic image in which the movement of the chest of the subject M1 is captured by photographing the appearance of the chest of the subject M1.

  The cycle detection device 16 analyzes the dynamic image acquired by the cycle detection sensor 15 to acquire information related to the temporal change in the respiratory phase for the respiratory cycle of the subject M1, and performs imaging control. It outputs to the control part 21 of the apparatus 2. Thereby, the information regarding the phase of respiration in the acquisition timing of each frame image by the imaging device 1 can be acquired.

<(3) Shooting control device>
The imaging control device 2 controls the X-ray imaging and frame image reading operations of the imaging device 1 by outputting information indicating the irradiation condition, the reading condition, and the like to the imaging device 1. In the imaging control device 2, for example, an image acquired by the imaging device 1 is appropriately displayed on the display unit 24. Thereby, the positioning including the position and posture of the subject M1 and the suitability for image diagnosis can be confirmed by the imaging technician.

  As shown in FIG. 1, the imaging control device 2 includes a control unit 21, a storage unit 22, an operation unit 23, a display unit 24, and a communication unit 25 that are connected to each other via a bus 26 so as to be able to exchange data. Configured.

  The control unit 21 includes, for example, a processor and a memory. The processor is, for example, a CPU (Central Processing Unit), and the memory may be, for example, a volatile RAM (Random Access Memory). For example, the processor reads out a program stored in the storage unit 22 in accordance with an operation of the operation unit 23, expands the program in the memory, and executes various processes according to the expanded program. The operation of each part of the apparatus 1 and the imaging control apparatus 2 is controlled.

  The storage unit 22 includes, for example, a nonvolatile semiconductor memory or a hard disk. The storage unit 22 stores various programs executed by the control unit 21 and various data necessary for execution of processing. Here, the various data includes data indicating parameters and the like necessary for execution of processing according to the program, and data generated at least temporarily as a result of the arithmetic processing.

  The operation unit 23 includes, for example, a pointing device such as a keyboard and a mouse. In the operation unit 23, a signal (also referred to as an instruction signal) input in response to an operation on a keyboard, a mouse, or the like is output to the control unit 21. The operation unit 23 may employ a configuration such as a touch panel.

  The display unit 24 includes a monitor such as an LCD (Liquid Crystal Display). In the display unit 24, in accordance with a signal input from the control unit 21, the contents of the instruction input from the operation unit 23 and various data are visually output.

  The communication unit 25 includes, for example, a LAN adapter, a modem, or a terminal adapter (TA). The communication unit 25 controls transmission / reception of data to / from each device connected to the network line NTW.

<(4) Image processing device>
<(4-1) Schematic Configuration of Image Processing Device>
The image processing device 3 is an information processing device that performs information processing. For example, a dynamic image obtained by photographing in the photographing device 1 is acquired via the photographing control device 2, and image processing is performed on the dynamic image. An image suitable for interpretation and diagnosis by a doctor or the like is generated. The image is appropriately displayed on the display unit 34. In the present embodiment, the image processing device 3 captures a target site to be diagnosed from a plurality of frame images constituting a dynamic image in which a dynamic aspect of the chest mainly including the heart and both lungs is captured. Lung field regions as regions (also referred to as target regions) are respectively detected. In addition, the information processing apparatus should just have the functional structure similar to the structure of a personal computer (personal computer) or a personal computer, for example.

  As shown in FIG. 1, the image processing apparatus 3 includes a control unit 31, a storage unit 32, an operation unit 33, a display unit 34, and a communication unit 35 that are connected to each other via a bus 36 so as to be able to exchange data. Configured.

  The control unit 31 includes, for example, a processor 31a and a memory 31b. The processor 31a is, for example, a CPU, and the memory 31b may be, for example, a volatile RAM. For example, the processor 31a reads out the program P1 stored in the storage unit 32 in accordance with the operation of the operation unit 33, expands it in the memory 31b, and executes various processes according to the expanded program P1. Thus, the operation of each unit of the image processing apparatus 3 is controlled. That is, the program P1 is executed by the control unit 31 included in the information processing apparatus, thereby causing the information processing apparatus to function as the image processing apparatus 3.

  The storage unit 32 includes, for example, a nonvolatile semiconductor memory or a hard disk. The storage unit 32 stores a program P1 executed by the control unit 31 and various data D1 necessary for execution of processing. Here, the various data D1 includes data indicating parameters and the like necessary for execution of processing according to the program, data generated at least temporarily as a result of the arithmetic processing, and the like. The program P1 includes, for example, a program for executing image processing. The program P1 is stored in the storage unit 32 in the form of readable program code, and the control unit 31 sequentially executes operations according to the program code.

  The operation unit 33 includes, for example, a pointing device such as a keyboard and a mouse. In the operation unit 33, a signal (also referred to as an instruction signal) that is input in response to an operation on the keyboard and the mouse is output to the control unit 31. The operation unit 33 may employ a configuration such as a touch panel.

  The display unit 34 includes, for example, a monitor such as an LCD. In the display unit 34, in accordance with a signal input from the control unit 31, the contents of instructions input from the operation unit 33, various data, and various images are visually output.

  The communication unit 35 includes, for example, a LAN adapter, a modem, or a TA. The communication unit 35 controls transmission / reception of data to / from each device connected to the network line NTW.

<(4-2) Functional Configuration of Image Processing Device>
FIG. 2 is a block diagram showing a functional configuration realized by the control unit 31 when the processor 31a and the like operate according to the program P1 in the image processing apparatus 3, together with other configurations. As shown in FIG. 2, the control unit 31 mainly includes an image acquisition unit 311, an image correction unit 312, a statistical value distribution calculation unit 313, a coarse detection unit 314 as a first detection unit, and a second detection unit. A precision detection unit 315 and a target region extraction unit 316 are provided.

  FIG. 3 is a flowchart showing a flow of information processing (also referred to as a target area extraction process) mainly including the following processes A1 to A8, which is executed in the image processing apparatus 3. By the target region extraction processing, a target region that captures the target region is accurately extracted from a series of a plurality of frame images I0 (see FIG. 4) in which a dynamic aspect of the internal structure obtained by X-ray imaging is captured. Can be done.

  [Processing A1] The image acquisition unit 311 acquires a series of a plurality of frame images I0 constituting a dynamic image from the imaging control device 2.

  [Process A2] The image correction unit 312 performs correction on each frame image I0, thereby reducing noise and reducing the influence of a structure other than the target part. The corrected frame image IC0 (FIG. 10 and FIG. 11). Here, corrected first to tenth frame images IC1 to IC10 are obtained by correction for the first to tenth frame images I1 to I10 (see FIG. 4).

  [Process A3] A two-dimensional distribution of statistical values in the time direction (statistics) for the plurality of corrected frame images IC0 obtained in Process A2 regarding the movement of the subject captured by the dynamic image by the statistical value distribution calculation unit 313. Value distribution).

  [Process A4] The rough detection unit 314 uses template matching to roughly detect the position of the target region from the corrected first frame image IC0 (first frame image IC1) (both the rough detection process). Say) is done. Note that a template normally used in template matching is, for example, an average image simply generated from a plurality of images, or an image (also referred to as a model image) expressed by a simple model expressed in black and white. For this reason, in template matching for X-ray images, if the number of templates used is limited, the shape of the lung field and heart due to differences in shape such as size between individuals and changes in respiration and heart rate Due to the difference, the lung field region is not accurately detected up to the contour. Therefore, information indicating a rough position of the target portion in the frame image IC0 is obtained from the detection result by template matching.

  [Process A5] Information regarding the position of the target part roughly detected in process A4 by the precision detection unit 315 is set as an initial value, and a target area extending from the inside of the target part to the contour in each frame image IC0 is detected. Processing that is detected in detail (also referred to as precise detection processing) is performed.

  Here, after the processes A4 and A5 for the first frame image IC0 (first frame image IC1) are performed once, the second and subsequent frame images other than the first frame image IC1 constituting the dynamic image are displayed. The process in which the processes A7, A8, and A5 for the IC0 are performed in this order is repeated. At this time, in the precision detection process of process A5, information related to the position of the target portion roughly detected in process A8 is set as an initial value.

  [Process A6] The target area extraction unit 316 extracts an image of the target area from each frame image I0 based on the target area detected in detail in process A5.

  [Process A7] By the coarse detection unit 314, in the process A4 or the latest process A8, the portion corresponding to the template (also referred to as the corresponding region) in the frame image IC0 that is the target of the coarse detection process is the template (self Also called a template). Here, for example, the corresponding region may be a region having the highest degree of matching with the template in the frame image IC0 that is the target of the rough detection process. However, the corresponding region is not limited to the region having the highest matching degree with the template in the frame image IC0 that is the target of the rough detection process. For example, if the matching degree is equal to or higher than a preset threshold, It may be a region having the second highest matching score. Here, a partial image (also referred to as a partial image) corresponding to the corresponding region is extracted from each frame image IC0, and the partial image is used as a self-template.

  [Process A8] The coarse detection unit 314 uses template matching using the self template prepared in the latest process A7, and corrects the second and subsequent frame images IC0 (second to tenth frame images IC2 to IC10). ) Is performed as a target. For example, template matching using a self-template generated from the first frame image IC1 in process A7 is used, and the position of the target portion is roughly detected from the corrected second frame image IC2. In the process A8, template matching using the self template is performed based on the statistical value distribution calculated in the process A3.

  Information indicating the position of the target portion roughly detected in the process A8 is used for the precision detection process in the next process A5. For example, the information indicating the position of the target portion roughly detected with the second frame image IC2 as the target in the process A8 is used for the precision detection process in the precision detection unit 315 for the second frame image IC2. .

  Hereinafter, each part 311 to 316 will be described.

<(4-2-1) Image Acquisition Unit 311>
The image acquisition unit 311 acquires a dynamic image for medical use in which a structure inside the body is captured by X-ray irradiation. In this embodiment, the body structure is a lung field structure. Here, the image acquisition unit 311 acquires image data related to a plurality of frame images I0 constituting an X-ray dynamic image obtained by X-ray imaging in the imaging apparatus 1 via the imaging control apparatus 2. For example, the image acquisition unit 311 acquires a plurality of image data stored in the storage unit 22 of the imaging control device 2 via the communication unit 25, the network line NTW, and the communication unit 35 in this order.

  FIG. 4 is a diagram schematically illustrating an example of an X-ray dynamic image in which the dynamics of the structure associated with breathing in the lung field of the subject M1 is captured. FIG. 4 exemplifies a series of a plurality of frame images I0 obtained by sequentially performing imaging at a fixed timing in one cycle related to the breathing cycle. The plurality of frame images I0 shown in FIG. 4 are obtained by shooting at the timing when the time T is t1, t2, t3,..., T10, respectively. .. It is constituted by I10. Of the series of frame images I1 to I10, the first half of the frame images I1 to I5 capture the dynamics of the lung field expanding due to the inspiration, and the second half of the frame images I6 to I10 show the expiration of the breath. The dynamics of contraction of the lung field by movement are captured. Note that the frame image I0 is obtained by converting a two-dimensional intensity distribution of X-rays into, for example, a two-dimensional distribution of pixel values (also referred to as gray values) indicating black and white.

<(4-2-2) Image Correction Unit 312>
The image correction unit 312 performs correction on the plurality of frame images I0 constituting the X-ray dynamic image acquired by the image acquisition unit 311 to obtain a plurality of corrected frame images IC0. The corrected plurality of frame images IC0 are output to the statistical value distribution calculation unit 313, the coarse detection unit 314, and the fine detection unit 315. The plurality of frame images IC0 are, for example, frames corresponding to the plurality of frame images I1, I2, I3,... I10 obtained by photographing at time T is t1, t2, t3,. Consists of images IC1, IC2, IC3,... IC10.

  5 to 7 are diagrams for explaining the characteristics of the frame image I0. FIG. 5 shows a frame image I0, and FIG. 6 shows the grayscale values in a region HR1 (refer to FIG. 5) in which the background of the subject M1 is captured in the frame image I0 (refer to FIG. 5). A histogram showing the frequency is shown. In the histogram, the horizontal axis indicates the gray value, and the vertical axis indicates the frequency of the gray value. Further, FIG. 7 shows a graph showing the gray value distribution in the vertically long region VR2 (see FIG. 5) including the lung field as the target site in the frame image I0. In the graph, the horizontal axis indicates the coordinate in the Y direction, and the vertical axis indicates the average of the gray value at each Y coordinate in the region VR2. As shown in FIGS. 5 to 7, the frame image I0 has the following problems 1 to 3.

  [Problem 1] As shown in FIG. 6, in the frame image I0, there is a lot of noise even in the background region HR1 where the subject M1 is not captured, compared with an X-ray image as a general still image. The S / N ratio in the entire frame image I0 tends to be low. This is because the amount of exposure required for photographing each frame image is kept low so that the amount of exposure of the subject M1 due to X-rays does not increase in the acquisition of the X-ray dynamic image capturing the target region. If there is a lot of noise in this way, it becomes difficult to accurately extract the lung field region as the target region where the lung field as the target region is captured.

  [Problem 2] As shown in FIG. 7, in the frame image I0, the grayscale values related to structures other than the lungs (structures such as ribs and blood vessels) are superimposed on the grayscale values related to the lungs. Variations occur in the gray value of the object, and the contour of the lung field region as the target region is unclear. Thus, if the outline is unclear, it is difficult to accurately extract the lung field region as the target region where the lung field as the target region is captured.

  [Problem 3] In the frame image I0, the gray value in the vicinity of the region where the horizontal angle is captured is closer to the gray value of the region corresponding to the part outside the lung field than in the lung field, and the region where the heart is captured and the lung The boundary with the field is also unclear. Thus, if the boundary is unclear, it is difficult to accurately extract the lung field region as the target region where the lung field as the target region is captured. Note that the region where the heel lateral angle is captured corresponds to the region where both ends of the diaphragm are captured.

  In order to reduce the occurrence of the above problems 1 to 3, as a process of reducing noise in the frame image I0 and reducing the influence of structures other than the lungs on the frame image I0, a correction process for clarifying the outline of the lung field region Is applied to the frame image I0.

  8 to 11 are diagrams for explaining the content of the correction process performed on the frame image I0 in the image correction unit 312. FIG.

  FIG. 8 shows a frame image I0. FIG. 9 shows the frame image IR0 after the processing for reducing noise existing in the entire image (also referred to as noise reduction processing) is performed on the frame image I0. 10 and 11 show the frame image IC0 after the enhancement process is performed on the frame image IR0. FIG. 10 shows the frame image IC0 after the entire frame image IR0 has been subjected to enhancement processing (also referred to as overall enhancement processing). Further, FIG. 11 shows the frame image IC0 after an enhancement process (also referred to as an in-area enhancement process) is performed on a preset area (also referred to as a predetermined area) of the frame image IR0. 9 to 11, histograms are shown in which the horizontal axis represents the gray value (pixel value) and the vertical axis represents the frequency (unit:%) of the gray value (pixel value). . In the present embodiment, in the enhancement processing in the image correction unit 312, noise reduction processing and enhancement processing are performed in this order.

  Here, the contents of the noise reduction process and the enhancement process will be briefly described.

<(4-2-2-1) Noise reduction processing>
Examples of noise reduction processing include reduction processing. In the reduction process, for example, a selected area (also referred to as a selection area) is replaced with a representative pixel value (also referred to as a representative pixel value) related to a plurality of pixel values (light / dark values) constituting the selection area. The representative pixel value may be, for example, an average pixel value (gray value) in the selection area, or a minimum pixel value in the selection area. When the representative pixel value is an average pixel value in the selected area, the reduction process is realized by, for example, a filter process using an average value filter.

  12 to 16 are diagrams for explaining an example of the reduction process. FIG. 12 shows a state where one selection area R3 is set in the frame image I0. FIG. 13 shows an enlarged view of the selection area R3 in the frame image I0, and shows a pixel Pm having the smallest pixel value in the selection area R3. Here, a reduction process using the pixel value of the pixel Pm as the representative pixel value is performed for each selection area R3. Thereby, the pixel value (that is, noise) on the maximum value side in the lung field region captured in the image I0 is reduced.

  For example, FIG. 14 shows a state where an area R4 is set in the frame image I0. FIG. 15 shows an enlarged view of the area R4 in the frame image I0, and FIG. 16 shows an enlarged view of the area R4 in the frame image IR0 after the reduction processing is performed on the frame image I0. Yes. As shown in FIGS. 15 and 16, by applying the enhancement process to the frame image I0, a small pixel value in the lung field region is enhanced, and noise in the lung field region is reduced. Note that smoothing may be performed when noise on the minimum pixel value side is enlarged.

  Note that the noise reduction process is not limited to the above reduction process, and other processes such as a process of cutting a high-frequency component of the frame image I0 using a bandpass filter or the like may be employed.

<(4-2-2) Enhancement processing>
In the overall enhancement process, for example, pixel values that are equal to or greater than a preset threshold value are replaced with relatively low pixel values that are equal to or less than the threshold value in accordance with a preset rule. Thereby, the influence of structures other than the lungs on the frame image I0 is reduced.

  As shown in FIGS. 9 and 10, in the frame image IC0 subjected to the overall enhancement processing, the white region having a large gray value is relatively less than the frame image IR0 subjected to the noise reduction processing. There are relatively many black areas with small gray values. That is, the whole area including the lung field area in the frame image IR0 is reduced by the whole enhancement process, and the white area corresponding to the structure other than the lung is reduced, and the black area is emphasized.

  The threshold value in the enhancement process may be a fixed value set in advance or may be changed as appropriate. For example, a mode in which the threshold value is changed in accordance with the state of the frame image IR0 that is the processing target can be considered. In this aspect, for example, regarding the histogram of the gray value of the frame image IR0, the frequency is accumulated in order from the smaller gray value, and the gray value at which the accumulated frequency reaches a predetermined value is set as the threshold value. In addition, when the cumulative value of the frequency is expressed as a percentage, the predetermined value may be 80%, for example. Here, for example, a mode in which a threshold is set for each frame image IR0 with respect to a series of frame images IR0 is conceivable. For example, a threshold is set for a series of a plurality of frame images IR0 based on the first frame image IR0, and the once set threshold is applied to enhancement processing for other frame images IR0. Conceivable. If such an aspect is adopted, the time required for image processing can be shortened by reducing the amount of calculation.

  By the way, in the in-area enhancement process, the enhancement process is performed on a predetermined area in the frame image IR0. The predetermined area may be, for example, a desired fixed area, but may be changed according to the distribution (profile) of gray values in the frame image IR0 as a processing target in order to improve robustness. .

  FIG. 17 is a diagram for explaining an example of a predetermined area setting method in the in-area enhancement process. In the right part of FIG. 17, the relationship between the Y coordinate and the accumulated value of the gray value (also referred to as the gray value accumulated value) at each Y coordinate of the frame image IR0 is shown as a vertical profile. Further, in the lower part of FIG. 17, the relationship between the X coordinate and the accumulated value of the gray value at each X coordinate of the frame image IR0 (light gray value) is shown as a horizontal profile.

  Here, for example, points (also referred to as change points) Px1, Px2, Py1, and Py2 at which the tendency of the accumulated gray value changes greatly in each profile are detected. A rectangular region R5 in which the X coordinate is in the range from the change point Px1 to the change point Px2 and the Y coordinate is in the range from the change point Py1 to the change point Py2 is set as the predetermined region.

<(4-2-3) Statistical Value Distribution Calculation Unit 313>
The statistical value distribution calculation unit 313 executes a process (also referred to as a statistical value distribution calculation process) for calculating a two-dimensional distribution (statistical value distribution) of statistical values in the time direction with respect to a plurality of frame images IC0 constituting the dynamic image. To do.

  FIG. 18 is a block diagram showing a functional configuration realized by the statistical value distribution calculation unit 313 together with other configurations. As illustrated in FIG. 18, the statistical value distribution calculation unit 313 mainly includes a recognition unit 3131, a statistical value calculation unit 3132, and a distribution generation unit 3133.

<(4-2-3-1) Purpose of calculating statistical value distribution>
FIG. 19 is a diagram illustrating an example of a periodic change in the position of the diaphragm due to respiration in the plurality of frame images IC0. In FIG. 19, the horizontal axis indicates time, the vertical axis indicates the position of the diaphragm as a specific portion in the −Y direction, and the temporal change in the position of the diaphragm is indicated by a broken line. As shown in FIG. 19, at time tp1, the diaphragm reaches the position where the diaphragm is most elevated by expiration, and at time tb1, the diaphragm reaches the position where the diaphragm is most lowered by inspiration.

  20 and 21 are diagrams illustrating points (also referred to as first to third observation points) Pt1 to Pt3 for observing changes in pixel values due to breathing in and around the lung field. FIG. 20 shows the first to third observation points Pt1 to Pt3 of the frame image IC0 in which the lung field at time tp1 is captured, and FIG. 21 shows the frame image in which the lung field at time tb1 is captured. First to third observation points Pt1 to Pt3 of IC0 are shown. As shown in FIG. 20 and FIG. 21, the first observation point Pt1 is always located in a region corresponding to a part in the lung field, and the second observation point Pt2 is a region corresponding to a part outside the lung field and in the lung field. A round trip is made to and from the area corresponding to the part. The third observation point Pt3 is always located in a region corresponding to a portion outside the lung field.

  FIG. 22 is a diagram exemplifying a change of the pixel value with the passage of time at the first to third observation points Pt1 to Pt3. In FIG. 22, the horizontal axis indicates the time, and the vertical axis indicates the pixel value (light / dark value). In FIG. 22, the change in the pixel value at the first observation point Pt1 is indicated by a dashed-dotted curve Cv1, the change in the pixel value at the second observation point Pt2 is indicated by a solid curve Cv2, and the third observation point Pt3. The change in the pixel value at is indicated by a dashed curve Cv3.

  As shown in FIG. 22, the first observation point Pt1 is always located in a region corresponding to a portion in the lung field, and the pixel value at the first observation point Pt1 is close to black regardless of the passage of time. It is held at a low value corresponding to low brightness. On the other hand, the third observation point Pt3 is always located in a region corresponding to a portion outside the lung field, and the pixel value at the third observation point Pt3 is high corresponding to high brightness close to white regardless of the passage of time. Held in value. On the other hand, the second observation point Pt2 reciprocates between a region corresponding to a portion outside the lung field and a region corresponding to the portion inside the lung field according to the expansion and contraction of the lung field due to respiration. For this reason, the pixel value at the second observation point Pt2 reciprocates between a high value corresponding to high luminance close to white and a low value corresponding to low luminance close to black.

  As described above, in the dynamic image, the pixel value of the pixel corresponding to the contour of the part having motion and the vicinity thereof is a high value corresponding to high brightness close to white, similar to the second observation point Pt2. To and from a low value corresponding to a low brightness close to black. In other words, the pixel values of the pixels corresponding to the contour of the part having movement and the vicinity thereof change greatly with the passage of time.

  Therefore, in the present embodiment, a statistical value distribution indicating a region where the pixel value greatly changes with the passage of time is acquired by calculating the statistical value in the time direction in the dynamic image. According to this statistical value distribution, if the target part has a movement, a region where the contour of the target part having an abrupt change and an unexpected shape can exist can be easily recognized. Therefore, in the present embodiment, the target region where the contour of the target part is detected in the precision detection process is expanded according to the statistical value distribution, and the region capturing the target part can be accurately extracted from the dynamic image. .

<(4-2-3-2) recognition unit 3131>
The recognition unit 3131 acquires the frame image IC0 from the image correction unit 312, and for each frame image IC0, a feature value indicating a feature of a pixel value for each region (also referred to as an image region) including a plurality of pixels or a pixel for each pixel Recognize the value. Here, with respect to the edge-enhanced frame image Ee0 (see FIGS. 27 and 28) generated by performing edge enhancement processing (also referred to as edge enhancement processing) on each frame image IC0, the feature value for each image region Or the pixel value for every pixel may be recognized. As the edge enhancement processing, for example, filter processing using a Sobel filter or a Laplacian filter can be employed.

  FIG. 23 is a diagram illustrating an example of the pixel Pm0 as a statistical value calculation target in the frame image IC0. FIG. 24 is a diagram illustrating an example of an image region Rv0 as a statistical value calculation target in the frame image IC0.

  As shown in FIG. 24, each image region Rv0 may be any image region Rv0 as long as the frame image IC0 is divided into a plurality of regions according to a preset division rule. The segmentation rule may be any rule that allows the frame image IC0 to be segmented into an area composed of a preset number of pixels. Each image region Rv0 may be, for example, a rectangular region having m pixels (m is a natural number) in the horizontal direction and n pixels (n is a natural number) in the vertical direction. Note that the shape and size of the image region Rv0 are not limited to a fixed one, and may be changed depending on the location in the frame image IC0, for example.

  Examples of the feature value for each image area include a representative value of a pixel value in each image area and a numerical value indicating the degree of dispersion. Here, as representative values of pixel values in each image region, for example, average value (average), median value (moden), mode value (mode), maximum value (maximum), and minimum value (minimum) of pixel values Etc. Thereby, the feature value indicating the feature of the pixel value for each image region can be easily calculated. The average value and the median value of the pixel values are calculated by, for example, filter processing using an average value filter and a median filter, respectively. In addition, examples of numerical values indicating the degree of dispersion of pixel values in each image region include pixel value variance, standard deviation, coefficient of variation (CV), and the like.

  The average value may be, for example, a value (arithmetic average value) obtained by dividing the sum of the pixel values of all the pixels in the image region by the number of pixels. The median value may be, for example, the median value when the pixel values of all the pixels constituting the image area are arranged in order from the largest. For example, if the number of pixels constituting the image area is an odd number, the order is the center value, and if the number of pixels constituting the image area is an even number, the order is the center value. An aspect that is an average value of two values is conceivable. The mode value may be, for example, a pixel value having the largest number of appearances among pixel values of all pixels in the image region.

Here, the variance (σ ² ) is a numerical value indicating variations in pixel values in the image area. For example, the average of the squares of the difference between the pixel value of each pixel in the image area and the average value of the pixel values in the image area Any value is acceptable. The standard deviation (σ) is a numerical value indicating the degree of variation of the pixel value in the image region, and may be a positive square root value of the variance (σ ² ), for example. The variation coefficient may be a numerical value obtained by dividing the standard deviation (σ) by the arithmetic average of the pixel values. In addition, when the standard deviation (σ) is compared, there is a case where the influence of the unit of numerical value may occur. Therefore, the influence of the unit of numerical value is prevented by adopting a variation coefficient that is a dimensionless number. .

<(4-2-3-3) Statistical Value Calculation Unit 3132>
The statistical value calculation unit 3132 calculates statistical values in the time direction related to the plurality of frame images IC0 or the plurality of edge-enhanced frame images Ee0. Here, for example, for a plurality of frame images IC0 or a plurality of edge-enhanced frame images Ee0, statistical values in the time direction are calculated with respect to feature values for each image region recognized by the recognition unit 3131 or pixel values for each pixel.

  The statistical value in the time direction is, for example, a numerical value indicating a scatter degree in the time direction regarding a feature value related to a pixel value for each image region or a pixel value for each pixel for a plurality of frame images IC0 or a plurality of edge-enhanced frame images Ee0. Alternatively, it may be a representative value in the time direction. Thereby, the two-dimensional distribution of the statistical value concerning the temporal change of the structure captured by the dynamic image can be easily calculated.

  Here, examples of numerical values indicating the degree of dispersion in the time direction include pixel value variance, standard deviation, coefficient of variation (CV), and the like. If such a numerical value indicating the degree of dispersion is employed, the influence of noise in the frame image IC0 can be reduced. Moreover, as a representative value, the numerical value etc. which concern on the difference of the maximum value and the minimum value etc. are mentioned, for example. The numerical value related to the difference between the maximum value and the minimum value may be the difference between the maximum value and the minimum value itself, or a value obtained by adding a slight calculation to the difference between the maximum value and the minimum value. May be.

Here, the variance (σ ² ) is a numerical value indicating variation in the time direction of the pixel value related to the same pixel or the feature value related to the same image area. The variance (σ ² ) is, for example, the mean value of the square of the difference between the pixel value in each frame image IC0 for the same pixel and the average value of the pixel value, or the feature in each frame image IC0 for the same image region. The average value of the square of the difference between the value and the average value of the feature values may be used. The standard deviation (σ) is a numerical value indicating the degree of variation in the time direction of the pixel value related to the same pixel or the feature value related to the same image area. For example, the standard deviation (σ) may be a positive square root value of the variance (σ ² ). The variation coefficient is, for example, a numerical value obtained by dividing the standard deviation (σ) by a weighted average of pixel values for the same pixel, or a numerical value obtained by dividing the standard deviation (σ) by a weighted average of feature values for the same image region. I need it. If a coefficient of variation that is a dimensionless number is employed, the influence of numerical units is prevented.

<(4-2-3-4) Distribution Generation Unit 3133>
The distribution generation unit 3133 generates a two-dimensional distribution (statistic value distribution) of statistical values calculated by the statistical value calculation unit 3132. Thereby, for example, for a plurality of frame images IC0 or a plurality of edge-enhanced frame images Ee0, a statistical value distribution is calculated as a two-dimensional distribution of statistical values in the time direction regarding feature values for each image region or pixel values for each pixel. The Data Sv indicating the statistical value distribution generated by the distribution generation unit 3133 is output to the coarse detection unit 314.

  FIG. 25 is a diagram illustrating a statistical value distribution Ds1 as an example of a statistical value distribution related to a dynamic image. The statistical value distribution Ds1 is a two-dimensional distribution of standard deviations as numerical values indicating the degree of dispersion in the time direction with respect to pixel values for each pixel for a plurality of frame images IC0. In FIG. 25, the level of the standard deviation is indicated by the level of luminance. Specifically, the lower the standard deviation, the lower the luminance is near black, and the higher the standard deviation, the higher the luminance near white. As shown in FIG. 25, in the statistical value distribution Ds1, an area mainly near the diaphragm and the outline of the heart is an area having a large standard deviation of high luminance close to white. Here, if a numerical value indicating the degree of dispersion such as standard deviation is used as a statistical value instead of a representative value in the time direction, reflection of noise in the statistical value of the frame image IC0 can be reduced.

<Specific example of (4-2-3-5) statistical value distribution generation method>
FIG. 26 is a diagram for explaining an example of a method for generating a statistical value distribution relating to a dynamic image. FIG. 26 shows an example of calculating a statistical value when the representative value in the time direction as the statistical value in the time direction is the difference between the maximum value and the minimum value of the pixel values. As shown in FIG. 26, for example, the difference Df1 between the maximum value and the minimum value of the curve Cv2 (FIG. 22) indicating the change in the pixel value at the second observation point Pt2 is calculated as a representative value in the time direction. However, according to this generation method, the calculation of the statistical value is easy, but the pixel value related to the noise in the frame image IC0 is easily reflected in the statistical value.

  Therefore, for example, for a plurality of frame images IC0, a difference Df1 between the maximum value and the minimum value as a statistical value is calculated with respect to representative values such as an average value, a median value, and a mode value of pixel values for each image region. For example, reflection of noise in the frame image IC0 in the statistical value can be reduced. Further, for example, the difference Df1 between the maximum value and the minimum value as the statistical value may be calculated with respect to the average value of the same pixels of two or more frame images IC0 in the time direction. Further, for example, an average value of N values from the larger one of the differences between the maximum value and the minimum value may be calculated as the difference Df1 between the maximum value and the minimum value.

  27 to 29 are diagrams for explaining an example of a method for generating a statistical value distribution using the edge-enhanced frame image Ee0.

  FIGS. 27 and 28 show edge-enhanced frame images Ee1, Ee2 () generated by emphasizing edges related to the structure of the corrected frame image IC0 based on the two frame images I1, I2 constituting the dynamic image. Ee0) is illustrated. Further, for example, a statistical value such as a numerical value indicating a degree of dispersion in the time direction or a representative value is calculated for each pixel or each image region of the plurality of edge-enhanced frame images Ee0 related to the dynamic image, thereby calculating the statistical value distribution Should be generated.

  FIG. 29 shows a statistical value distribution Ds2 generated by calculating an average value as a representative value that is a statistical value in the time direction for each pixel of the plurality of edge-enhanced frame images Ee0 related to the dynamic image. ing. For example, a region AE0 in FIG. 29 shows a state in which a plurality of white streaks that are shifted from each other corresponding to the diaphragm are arranged. That is, for a portion where a large movement occurs, for example, a region having a relatively high statistical value is distributed over a wide range.

  Here, although an example in which the average value of the pixel values is calculated for each pixel of the plurality of edge-enhanced frame images Ee0 related to the dynamic image has been described, the present invention is not limited to this. For example, a statistical value distribution is calculated by calculating a cumulative value or a difference between a maximum value and a minimum value for a pixel value of each pixel of a plurality of edge-enhanced frame images Ee0 related to a dynamic image or a feature value for each image region. May be calculated.

  If such a plurality of edge-enhanced frame images Ee0 are used, a change in the vicinity of the contour of the target part can be more reliably reflected in the statistical value distribution. If the frame rate of the dynamic image is set to a relatively large value, it is difficult to cause a problem in which statistical values with relatively high brightness appear only in a narrow range in the statistical value distribution. Can be recognized well.

  FIG. 30 is a diagram illustrating the first and second observation points Pt1 and Pt2 in the frame image IC0. FIGS. 31 to 34 are diagrams for explaining an example of a method for generating a statistical value distribution in which statistical values in the time direction are calculated from a histogram indicating the frequency of pixel values for each pixel in a plurality of frame images IC0. FIG. 31 is a histogram showing the frequency of pixel values at the first observation point Pt1 for a plurality of frame images IC0. FIG. 32 is a histogram showing the frequency of pixel values at the second observation point Pt2 for a plurality of frame images IC0. 31 and 32, the horizontal axis indicates the pixel value, and the vertical axis indicates the frequency of the pixel value.

  Here, for each pixel, a value having the highest pixel value frequency (also referred to as a mode value) is recognized, and a numerical value relating to the frequency occupancy in a preset range of width based on the mode value is obtained. An example calculated as a representative value that is a statistical value in the time direction will be described. As shown in FIGS. 33 and 34, for example, a numerical value indicating the frequency occupancy in the range of the width Aw1 centered on the mode value Pk0 is calculated as a representative value that is a statistical value in the time direction. For example, in the example shown in FIG. 33, when the maximum value of the frequency occupancy is 1, the frequency occupancy in the range of the width Aw1 is calculated as about 0.8. In the example shown in FIG. The frequency occupancy in the range of the width Aw1 is calculated to be about 0.3. If such an aspect is adopted, the influence on the statistical value due to the difference in shading between the plurality of frame images IC0 can be reduced.

  For example, a numerical value relating to the frequency occupancy in a value range of a preset width based on a value having the highest frequency of feature values (mode value) for each image area, not for each pixel, in the time direction May be calculated as a representative value as a statistical value.

<(4-2-4) Coarse detection unit 314>
The coarse detection unit 314 detects the target confirmed region PR11 (see FIG. 36) and the unconfirmed region PR12 (see FIG. 36) using template matching for each frame image IC0 (see FIG. 35) constituting the dynamic image. To do.

  Examples of template matching methods include known SAD (Sum of Absolute Differences) methods, NCC (Normalized Cross Correlation) methods, POC (Phase-Only Correlation) methods, and RIPOC (Rotation Invariant Phase-Only Correlation) methods. Is adopted. As a result, the target candidate region PR1 (see FIG. 36) including the target confirmed region PR11 and the undetermined region PR12 is detected. Here, the target confirmed region PR11 is a region in which at least a part of the lung field as the target region is captured in the frame image IC0, and the undetermined region PR12 is the target region in the frame image IC0. This is a region where a part of the lung field may be captured.

  In the present embodiment, the coarse detection unit 314 also detects a region (also referred to as a non-target defined region) PR2 where a lung field as a target region is not captured for each frame image IC0. As a result, as shown in FIG. 36, information (also referred to as region division information) ID1 for dividing the frame image IC0 into the target confirmed region PR11, the undefined region PR12, and the non-target defined region PR2 is generated. .

  By the way, assuming the next precise detection process, in the rough detection process, the target fixed region PR11 and the non-target fixed region are usually from the frame image IC0 to the portion as close as possible to the contour of the structure that is difficult to detect. It suffices if PR2 is detected. That is, it is only necessary that the undetermined region PR12 becomes narrow. Therefore, for example, the target part is divided into a plurality of parts, and for the same part, the part according to the passage of time due to the difference in shape such as the difference in part size between individuals and the action such as breathing and heartbeat A mode in which a plurality of types of templates corresponding to the difference in shape is used can be considered. Thereby, in precision detection processing, it is hard to produce the malfunctions, such as an excessive detection of the object area | region beyond the unclear outline of a structure. In addition, as a structure where it is difficult to detect the contour, for example, a structure where the contour is unclear, or a structure that is divided by the presence of another structure, or the like can be considered.

  However, when template matching using a plurality of types of templates for the same portion is executed for each of the plurality of frame images IC0, the time required for the rough detection process increases due to an increase in the amount of calculation. In the rough detection process, the template having the highest correlation differs for each frame image IC0, so that the detection result by template matching may vary, and the detection result in the precise detection process may also vary.

  Therefore, in this embodiment, a partial image related to a corresponding region corresponding to the template in the frame image IC0 in the rough detection process is set as a template matching template (also referred to as a self template) for the next frame image IC0. That is, when M is a natural number, template matching using the self template obtained from the result of the Mth coarse detection process (also referred to as the Mth coarse detection process) is the next (M + 1) th coarse detection process (the first (M + 1) also called rough detection processing). The Mth coarse detection process is a coarse detection process for the Mth frame image ICM, and the (M + 1) th coarse detection process is a coarse detection process for the next (M + 1) th frame image IC (M + 1). It is. That is, for example, the template matching using the self-template obtained from the result of the first rough detection process for the first frame image IC1 is the second frame image IC2 for the second frame image IC2 next to the first frame image IC1. 2 It is performed in the rough detection process.

  Here, in the (M + 1) th coarse detection process for the (M + 1) th frame image IC (M + 1), for example, the contour of the target area (also referred to as the Mth target area) detected in the Mth precision detection process is used. A reference offset range may be detected as the undetermined region PR12. The Mth precision detection process is a precision detection process for the Mth frame image ICM. Specifically, for example, in the second rough detection process for the second frame image IC2, the offset range based on the contour of the first target area detected in the first precise detection process is an undetermined area PR12. May be detected. The offset range may be set according to, for example, a predetermined range set in advance or the movement of the target part. With such a rough detection process using a self-template, for the Mth rough detection process after the second rough detection process, the time required for the process is reduced by reducing the amount of calculation, and template matching is performed for each frame image IC0. Variations in detection results due to are unlikely to occur. That is, a smooth change in the position of the target portion detected from each of the plurality of frame images IC0 is realized by a relatively short time calculation process.

  However, if the rough detection process using the self-template is simply performed, among the parts captured by the plurality of frame images IC0, a part whose shape and position change significantly may be erroneously detected. it is conceivable that. For example, a defect that should originally be detected as the target determination region PR11 is detected as the non-target determination region PR2, and a region that should originally be detected as the non-target determination region PR2 is detected as the target determination region PR11. Can cause problems. In addition, for example, a problem that an area that should originally be detected as the undetermined area PR12 may be detected as the target determined area PR11 or the non-target determined area PR2.

  In response to such a problem, the rough detection unit 314 according to the present embodiment detects the undetermined region PR12 in consideration of not only the template matching detection result using the self-template but also the statistical value distribution. Thereby, an undetermined region PR12 larger than the undetermined region PR12 obtained when the rough detection process using the self-template is simply performed according to the statistical value related to the movement of the structure in the time direction. . As a result, it is possible to accurately extract a region that captures the target region from a medical dynamic image that captures the movement of the structure inside the body.

  FIG. 37 is a block diagram showing a functional configuration realized by the coarse detection unit 314 together with other configurations. As shown in FIG. 37, the coarse detection unit 314 mainly includes a first region detection unit 3141, a first region setting unit 3142, a template generation unit 3143, a second region detection unit 3144, a second region setting unit 3145, and A third area setting unit 3146 is provided.

<(4-2-4-1) First Area Detection Unit 3141>
The first region detection unit 3141 performs template matching (also referred to as first template matching) for the first first frame image IC1 among the plurality of frame images IC0. Here, a process (first configuration region) in which a first template matching is used to detect a region (also referred to as a configuration region) in which each portion (also referred to as a configuration portion) constituting a lung field as a target region is captured (first configuration region). (Also referred to as detection processing).

  Here, the first component region detection process will be described.

  In the lung field, there is a large change in the shape of the period and the difference in shape between individuals. For example, as shown in FIGS. 38 to 40, when the longitudinal size and shape of the lung field and diaphragm change periodically due to respiration, and as shown in FIGS. 41 to 44, the diaphragm and heart The case where the individual difference in a shape arises etc. can be considered. For this reason, for example, a plurality of templates are prepared in consideration of a change in shape and a difference in each part.

  For example, for the lung field, a plurality of templates may be prepared in accordance with phenomena occurring in the entire lung field, the diaphragm, the heart, the collarbone, the rib, the lateral radius, the aorta, and the like captured by the plurality of frame images IC0.

  Regarding the entire lung field, for example, the length of the lung field in the vertical direction varies greatly depending on individual differences and movement due to breathing. For this reason, for example, a plurality of templates in a relationship in which the entire lung field is expanded and contracted in the vertical direction (also referred to as a longitudinal scaling relationship), and an upper portion and a lower portion that divide the lung field in the vertical direction, respectively. A mode in which at least one of a plurality of templates is adopted is conceivable. Further, in the frame image IC0, since there are many structure boundary candidates due to the presence of ribs or the like, for example, a mode in which a plurality of templates respectively corresponding to a plurality of portions obtained by dividing the entire lung field can be considered.

  For the diaphragm, for example, there are many variations in the positional relationship between the left diaphragm and the right diaphragm due to individual differences and movement due to breathing. For example, independent templates are adopted for the left diaphragm and the right diaphragm, respectively. A possible embodiment is conceivable. Further, for example, since the width of the region where the diaphragm is captured in the frame image IC0 differs depending on the difference in the width of the heart, for example, a mode in which a plurality of templates corresponding to the phase of the diaphragm displacement due to respiration is employed is conceivable. .

  For the clavicle, ribs, etc., for example, the clear outline of the frame image IC0 is easily confused with the outline of the lung field and the apex of the lung, and thus covers a wide range including, for example, the area capturing the shoulder. A mode in which a template to be adopted is considered. Moreover, regarding the aorta, for example, since individual differences are large, a mode in which a plurality of templates are employed for the difference in the shape of the aorta is conceivable. Regarding the heart and the lateral angle of the heel, for example, since the outline is unclear in the frame image IC0, a mode in which the undetermined region PR12 is set in a wide range in the second section setting process described later can be considered.

  By the way, in the present embodiment, for example, a relationship in which the difference in the shape of the diaphragm due to the difference in the size of the lung field and the heart and the difference in the positions of the left and right diaphragms are expanded and reduced in the vertical and horizontal directions (vertical and horizontal changes). Multiple templates are also used. In addition, different templates are employed for the left diaphragm and the right diaphragm for differences in the left and right positions of the diaphragm. Further, for example, the shape of the contour of the lung field along the heart varies depending on the difference in the size of the heart due to cardiac hypertrophy as a specific case. A plurality of templates that are enlarged and reduced in the horizontal direction and the horizontal direction are employed.

  In the present embodiment, an example in which a lung field as a target part is divided into a plurality of parts (also referred to as component parts) and a plurality of templates are prepared for each component part will be described. In this embodiment, as the plurality of constituent parts, parts including the left diaphragm, the right diaphragm, the left lung apex, and the right lung apex are employed. Therefore, templates for the left diaphragm, the right diaphragm, the left lung apex, and the right lung apex (also referred to as a left diaphragm template, a right diaphragm template, a left lung apex template, and a right lung apex template) are prepared.

  FIG. 45 to FIG. 48 are diagrams showing a method for generating a plurality of right diaphragm templates T1R and a plurality of left diaphragm templates T1L each having a relationship of vertical / horizontal scaling. A sample image SI0 is schematically shown on the left part of FIGS. 45 to 48, the right diaphragm templates T1Ra to T1Rd (T1R) and the left diaphragm templates T1La to T1Ld (T1L) respectively representing the shapes of the left and right diaphragms of the sample image SI0 are schematically shown. Yes.

  As shown in the sample images SI0 of FIGS. 45 to 48, there are large individual differences in the shape of the diaphragm, and the shape of the diaphragm varies greatly depending on the size of the heart. For this reason, for example, the sample images SI0 are classified into a plurality of groups according to the size of the heart and the like. Here, a mode in which the shape of the heart is classified into four groups, which are small, general, medium and large, is shown.

  Next, as shown in FIGS. 45 to 48, the right diaphragm region R1R and the left diaphragm region R1L surrounded by the broken line in the sample image SI0 are designated and cut out from the sample image SI0. Here, the right diaphragm region R1R is a region including the component region in which the right diaphragm is captured in the sample image SI0 and a region in the vicinity thereof (also referred to as a component-containing region). The left diaphragm region R1L is a component-containing region including a component region in which the left diaphragm is captured in the sample image SI0 and a region in the vicinity thereof. The designation of the right diaphragm region R1R and the left diaphragm region R1L for each sample image SI0 may be performed manually via the operation unit 33, for example. The designation of the right diaphragm region R1R and the left diaphragm region R1L for each sample image SI0 is, for example, template matching in which each one of the right diaphragm region R1R and the left diaphragm region R1L designated manually is used as a template. It may be realized by.

  Then, for example, four types of right diaphragm templates T1Ra to T1Rd (T1R) are generated by generating an average image of images cut out from the right diaphragm region R1R for each of the four types of groups classified in advance. The Here, each pixel value of the average image is calculated, for example, by dividing a cumulative value of pixel values related to pixels at the same position in an image cut out from a plurality of right diaphragm regions R1R by the number of right diaphragm regions R1R. Is done. Similarly, four types of left diaphragm templates T1La to T1Ld (T1L) are generated by generating an average image of images cut out from the left diaphragm region R1L for each of the four types of groups classified in advance. Is done.

  The other templates are also generated by the same method as the right diaphragm template T1R and the left diaphragm template T1L, for example. As for the left apex template and the right apex template, for example, the sample images SI0 are classified into a plurality of groups from the viewpoint of the size of the apex, the inclination of the collarbone, and the like. It only has to be generated.

  Information relating to the template generated in this way may be included in advance in the various data D1 stored in the storage unit 32.

  Then, in the first configuration area detection process, the position of the configuration area that captures the configuration part is detected by template matching using a template for the first frame image IC1. Here, for each component, a plurality of types of templates are used, and the component-containing region having the highest degree of correlation (also referred to as the degree of correlation) with the template in the first frame image IC1 is detected. A constituent area included in the constituent-containing area is detected.

<(4-2-2-4-2) First Area Setting Unit 3142>
The first region setting unit 3142 sets the first partial segment information for each component region detected in the first component region detection process according to the correlation value between the template and the component-containing region (first component) (Also referred to as setting processing). The first partial classification information includes the component-containing regions related to each component region as a target defined region PR11 (first target confirmed region PR11a), an undefined region PR12 (first undefined region PR12a), and a non-target defined region PR2 (first 1 is information classified into non-target defined regions PR2a). In addition, the first area setting unit 3142 executes a process of integrating the first partial section information set for each constituent area (also referred to as a first integration process). Thereby, area | region division information ID1 (1st area | region division information ID1a) is produced | generated. That is, processing for detecting the first target confirmed region PR11a and the second unconfirmed region PR12a for the first frame image IC1 is executed according to the result of the first template matching for the first frame image IC1. The

  In the first segment setting process, the first partial segment is determined for each component area detected in the first component area detection process based on the first partial segment information group included in the various data D1 stored in the storage unit 32. Set the information. Here, the first partial classification information group is a collection of first partial classification information corresponding to the correlation between the template and the component-containing region for each template. For this reason, in the first category setting process, the first partial category information corresponding to the correlation between the template and the component-containing region when the component-containing region is detected in the first component region detecting process is Obtained from the partial segment information group. Thereby, 1st partial division information is set with respect to the structure area contained in this structure containing area | region.

  Here, with reference to FIG. 49 to FIG. 53, a method for generating the first partial section information corresponding to the correlation between the template and the component-containing region for each template will be described.

  As described above, for example, when the template is generated, the sample images SI0 are classified into a plurality of groups, and for each group, the component-containing regions including the component region and the region in the vicinity thereof from the plurality of sample images SI0, respectively. The average image is generated by cutting out. At this time, first, data (also referred to as correct data) indicating whether the target region is a target region in which the target part is captured or a non-target region in which a part other than the target part is captured is generated for each extracted component-containing region. .

  FIG. 49 illustrates correct data A0 in which the target area is expressed in white and the non-target area is expressed in black for a certain sample image SI0. FIG. 50 shows the left diaphragm region R1L as the component-containing region and the correct data A1L (A0) regarding the left diaphragm region R1L, and the correct data A1R regarding the right diaphragm region R1R and the right diaphragm region R1R as the component-containing regions. (A0) is illustrated. 51 shows the left lung apex region R2L as the component-containing region and the correct data A2L (A0) regarding the left lung apex region R2L, the right lung apex region R2R as the component-containing region, and the right lung apex The correct answer data A2R (A0) related to the partial area R2R is illustrated. Here, the left lung apex region R2L is a configuration-containing region including a configuration region in which the left lung apex is captured in the sample image SI0 and a region in the vicinity thereof. The right lung apex region R2R is a configuration-containing region including a configuration region in which the right lung apex in the sample image SI0 is captured and a region in the vicinity thereof.

  Next, a plurality of correct data A0 is overlaid for each group for each configuration area, and the number of times the correct data A0 is overlaid for each pixel is divided by the number of the plurality of correct data A0. The probability that A0 overlaps (also referred to as a cumulative rate) is calculated. The accumulation rate may be shown in a form such as a percentage such as 0 to 100%. In the lower right of FIG. 52, an image (also referred to as a cumulative rate image) Am1L showing the distribution of the cumulative rate for a plurality of correct answer data A1L (A0) related to the left diaphragm region R1L is illustrated. In the lower left of FIG. 52, an accumulation rate image Am1R showing the distribution of accumulation rates for a plurality of correct answer data A1R (A0) related to the right diaphragm region R1R is illustrated. In the upper right of FIG. 52, an accumulation rate image Am2L showing the distribution of accumulation rates for a plurality of correct answer data A2L (A0) related to the left lung apex region R2L is illustrated. In the upper left of FIG. 52, an accumulation rate image Am2R showing the distribution of accumulation rates for a plurality of correct answer data A2R (A0) related to the right lung apex region R2R is illustrated.

  Here, for a certain group, it is certain that an area with a cumulative rate of 100% corresponds to a target area that captures the target part, and a region with a cumulative rate of 0% does not capture the target part. It is certain to correspond to the target area. On the other hand, it is uncertain whether an area with an accumulation rate of 1 to 99% corresponds to a target area or a non-target area. However, for example, for each group for each constituent area, the higher the correlation value between the template and the constituent inclusion area detected in the first constituent area detection process, the higher the cumulative ratio corresponds to the target area. The possibility to do increases.

  Therefore, for each template, first partial section information corresponding to the correlation value between the template and the component-containing region is generated based on the corresponding cumulative rate image.

  For example, for the first partial classification information for the template, a correlation value range may be set in advance, and the first partial classification information may be generated for each correlation value range. For example, a range having a large correlation value (0.9 <correlation value ≦ 1), a range having a medium correlation value (0.8 <correlation value ≦ 0.9), and a range having a small correlation value (0.5 <correlation value). A mode in which ≦ 0.8) is set is conceivable. In this aspect, for a range with a large correlation value (0.90 <correlation value ≦ 1), for example, a region with a cumulative rate of 0% is a non-target defined region, and a cumulative rate exceeds 0% and is 50% or less. A region is an unconfirmed region, and a region having an accumulation rate of more than 50% and not more than 100% is a target confirmed region. An example of the first partial section information PV1 in this case is shown in the left part of FIG. In addition, for an intermediate correlation value range (0.8 <correlation value ≦ 0.9), for example, an area where the accumulation rate is 0% is a non-target determination area, an accumulation ratio exceeds 0% and is less than 70% Is an undetermined area, and an area where the accumulation rate exceeds 70% and is equal to or less than 100% is a target confirmed area. An example of the first partial section information PV2 in this case is shown in the central part of FIG. Furthermore, for a range with a small correlation value (0.5 <correlation value ≦ 0.8), for example, a region with a cumulative rate of 0% is a non-target defined region, and the cumulative rate exceeds 0% and is 99% or less. A region is an unconfirmed region, and a region where the accumulation rate exceeds 99% and is 100% or less is a target confirmed region. An example of the first partial section information PV3 in this case is shown in the right part of FIG.

  In the first integration process, the first partial classification information set for each component area in the first classification setting process is integrated. Accordingly, the first region indicating the target candidate region PR1 (first target candidate region PR1a) including the first target confirmed region PR11a and the first undefined region PR12a and the non-target defined region PR2 (first non-target defined region PR2a). Setting of the division information ID 1a is completed. FIG. 54 is a diagram illustrating an example of the first region classification information ID1a indicating the first target candidate region PR1a and the first non-target determination region PR2a set by the first integration process. The first area classification information ID1a obtained here is sent to the precision detection unit 315.

<(4-2-4-3) Template Generation Unit 3143>
The template generation unit 3143 generates, as a self template, a partial image related to a corresponding region corresponding to a template in the Mth frame image ICM in template matching (also referred to as Mth template matching) for the Mth frame image ICM. . In the present embodiment, the corresponding region is a region having the highest degree of matching with the template in the Mth frame image ICM in the Mth template matching. However, the corresponding region is not limited to the region having the highest matching degree with the template in the Mth frame image ICM in the Mth template matching. For example, if the matching degree is equal to or higher than a preset threshold, It may be a region having the second highest matching score.

  Here, for example, in the first template matching in the first region detection unit 3141, the partial image related to the corresponding region corresponding to the template in the first frame image IC1 is extracted from the first frame image IC1, A self template is generated. Further, for example, in the M-th template matching (here, M is a natural number of 2 or more) in the second region detection unit 3144, the partial image related to the corresponding region corresponding to the template in the M-th frame image ICM is By extracting from the frame image ICM, a self template is generated.

  55 and 56 are diagrams for explaining a self-template generation method.

  FIG. 55 shows corresponding areas Mr1L, Mr1R, Mr2L, and Mr2L that have the highest degree of matching with each template in the first frame image IC1. For example, in the first frame image IC1, the outer edge of the corresponding region Mr1L having the highest degree of coincidence with the left diaphragm template T1L and the outer edge of the corresponding region Mr1R having the highest degree of coincidence with the right diaphragm template T1R are drawn by thick broken lines. It is. Further, for example, in the first frame image IC1, the outer edge of the corresponding region Mr2L having the highest degree of coincidence with the left lung apex template and the outer edge of the corresponding region Mr2R having the highest degree of coincidence with the right lung apex template are It is drawn with a thick broken line. Here, the corresponding region having the highest degree of coincidence with the template in the frame image IC0 has, for example, the highest degree of correlation (correlation degree) with the template in the frame image IC0 as a template matching target. Any region (structure-containing region) may be used.

  FIG. 56 schematically illustrates partial images Ts1L, Ts1R, Ts2L, and Ts2R as self templates. Specifically, the partial image Ts1L related to the left diaphragm corresponding region Mr1L in the first frame image IC1 is schematically shown in the lower right, and the right diaphragm corresponding region Mr1R in the first frame image IC1 in the lower left. A partial image Ts1R according to is schematically shown. Further, the partial image Ts2L related to the corresponding region Mr2L of the left lung apex in the first frame image IC1 is schematically shown in the upper right, and the corresponding region Mr2R of the right lung apex in the first frame image IC1 is shown in the upper left. A partial image Ts2R according to is schematically shown. FIG. 57 is a diagram illustrating an example of a partial image Ts1R as a self template.

<(4-2-4-4) Second Area Detection Unit 3144>
The second region detection unit 3144 targets a M-th frame image ICM (here, M is a natural number greater than or equal to 2) different from the first frame image IC1 among the plurality of frame images IC0, and performs template matching using a self-template. I do. Here, template matching is used to execute a process (also referred to as an M-th component area detection process) in which a component area in which each component part constituting the lung field is captured is detected.

  In the M-th component region detection process, template matching is performed using the self-template obtained from the (M-1) th frame image IC (M-1) for the Mth frame image ICM. For example, template matching using the self template obtained from the first frame image IC1 is performed on the second frame image IC2.

  Specifically, with respect to the Mth frame image ICM, the position of the configuration area capturing the configuration portion is detected by template matching using the self template. For example, for each component, a configuration in which one self-template is used and a region having the highest degree of correlation (also referred to as a degree of correlation) with the self-template in the Mth frame image ICM is captured. It is detected as a component-containing region including the region and the region in the vicinity thereof.

  However, in template matching using a self-template, there is a possibility that the detection accuracy of the constituent area is lowered for a portion where a large movement occurs in the dynamic image. For this reason, for example, in the template matching using the self template in the second region detection unit 3144, a process (also referred to as a weighting process) in which weighting according to the statistical value distribution calculated by the statistical value distribution calculating unit 313 is performed. Executed. As the weighting process, for example, for each pixel or image area, a correlation value between a self-template and an area of the Mth frame image ICM on which the self-template is superimposed (also referred to as a superposition area) is a statistical value distribution. A mode divided by the statistical value in can be considered. As a result, the occurrence of a malfunction in which the position of the constituent area is erroneously detected in a portion where a large movement occurs is reduced. As the weighting process, for example, for each pixel or image area where the statistical value in the statistical value distribution exceeds a preset threshold value, a predetermined coefficient (for the correlation value between the overlapping area of the Mth frame image ICM and the self template is used. A mode in which the weighting coefficient is also multiplied is conceivable. The predetermined coefficient may be a value set in advance (for example, 0).

  FIG. 58 is a diagram illustrating a two-dimensional distribution of weighting coefficients for the self-template. FIG. 58 shows an example in which the weighting coefficient for a pixel whose statistical value exceeds a threshold value in the statistical value distribution is 0, and the weighting coefficient for a pixel whose statistical value is less than or equal to the threshold value is 1 in the statistical value distribution. ing.

<(4-2-4-5) Second area setting unit 3145>
The second area setting unit 3145 applies the Mth frame area ICM (here, M is a natural number equal to or greater than 2) constituting the dynamic image to each of the constituent areas detected in the Mth constituent area detection process. A process of setting M partial section information (also referred to as an Mth section setting process) is executed. The Mth partial division information is information for dividing the configuration-containing region related to each configuration region into a target determination region PR11, an undefined region PR12, and a non-target determination region PR2. For example, the second partial classification information includes the component-containing regions related to each component region as the target confirmed region PR11 (second target confirmed region PR11b), the undefined region PR12 (second undefined region PR12b), and the non-target defined region PR2. The information is classified into (second non-target defined region PR2b).

  The second area setting unit 3145 also integrates the Mth partial segment information set for each constituent area for each Mth frame image ICM constituting the dynamic image (also referred to as Mth integration process). ) To generate the M-th area division information ID1. That is, processing for detecting the target confirmed region PR11 and the undefined region PR12 is executed for the Mth frame image ICM according to the result of the Mth template matching for the Mth frame image ICM. Specifically, for example, the second region segment information ID1b is obtained by executing the second integration process for integrating the second partial segment information set for each component region for the second frame image IC2. Is generated. That is, processing for detecting the second target confirmed region PR11b and the second unconfirmed region PR12b for the second frame image IC2 is executed according to the result of the second template matching for the second frame image IC2. The

  Here, in the M-th segment setting process, with respect to each configuration area detected in the M-th configuration area detection process, the (M-1) target area detected in the (M-1) precision detection process is used as a reference. The partial segment information is set. Specifically, for example, in the second segment setting process, partial segment information based on the first target area detected in the first precision detection process for each component area detected in the second component area detection process Is set.

  FIG. 59 and FIG. 60 are diagrams for explaining a setting method of the second partial section information in the second section setting process. In the second segment setting process, as shown in FIG. 59, for example, the first frame as the self template of the lung field region RT1 as the first target region detected in the first precise detection process and the vicinity region thereof. A portion (also referred to as a partial corresponding region) PA1 corresponding to the partial image extracted from the image IC1 is extracted. As shown in FIGS. 59 and 60, partial segment information PV0 in which the offset range based on the contour ED1 of the lung field region RT1 in the partial corresponding region PA1 is set as the undetermined region PR12 is set. The The offset range may be set according to, for example, a predetermined range set in advance or the movement of the target part. For example, the movement of the target part is predicted from the shift amount of the corresponding area corresponding to the template in the frame image IC0 between the first coarse detection process and the second coarse detection process for the same component. obtain. For example, a mode in which a value obtained by multiplying the deviation amount by a predetermined value (for example, 3) is set as the width of the offset range is conceivable. The offset range is set for each part by dividing the contour ED1 of the part corresponding area PA1 into a plurality of parts, for example.

  In the partial classification information PV0, a region corresponding to the lung field region RT1 and other than the undetermined region PR12 is set as the target confirmed region PR11, and a region not corresponding to the lung field region RT1 and other than the undetermined region PR12. Is defined as a non-target defined region PR2. FIG. 60 illustrates partial segment information PV1R (PV0) corresponding to the self template Ts1R related to the right diaphragm.

  In the Mth integration process, the Mth partial classification information set for each component area in the Mth classification setting process is integrated. For example, in the second integration process, the second partial classification information set for each component area in the second classification setting process is integrated. Thereby, the setting of the target candidate region PR1 including the target confirmed region PR11 and the undefined region PR12 and the Mth region segment information ID1 indicating the non-target determined region PR2 is completed. For example, the setting of the target region PR1b (second target candidate region PR1b) including the second target confirmed region PR11b and the second undefined region PR12b and the second region classification information ID1b indicating the second non-target determined region PR2b is completed. . FIG. 61 is a diagram illustrating an example of the second region segment information ID1b indicating the second target candidate region PR1b and the second non-target defined region PR2b set by the second integration process. The second area classification information ID1b obtained here is sent to the precision detection unit 315.

<(4-2-4-6) Third Area Setting Unit 3146>
The third region setting unit 3146 is based on the statistical value distribution obtained by the statistical value distribution calculating unit 313, and the unconfirmed region PR12 (third unconfirmed region PR12c) in which a part of the target part may be captured. Is executed (also referred to as an indeterminate detection process). In the undetermined detection process, for example, a mode is conceivable in which a region in the statistic value distribution satisfying a preset condition is detected as the third undetermined region PR12c. More specifically, for example, a mode is considered in which a region in which a magnitude relationship between a statistical value in a statistical value distribution and a preset threshold value satisfies a preset condition is detected as the third undetermined region PR12c. It is done. Thereby, the undetermined region PR12 can be easily expanded according to the temporal change of the structure captured by the dynamic image.

  FIG. 62 is a diagram showing a specific example of the third undetermined region PR12c. In FIG. 62, the third undetermined region PR12c is indicated by a white region.

  In addition, the third region setting unit 3146 removes a portion that overlaps the third unconfirmed region PR12c among the target confirmed region PR11 and the non-target confirmed region PR2 of the Mth region classification information ID1 generated by the Mth integration process. A process of changing to the fixed region PR12 (also referred to as a change process) is executed. For example, a portion of the second target determined area PR11b and the second non-target determined area PR2b of the second area classification information ID1b generated by the second integration process overlaps with the third undefined area PR12c in the undefined area PR12. Be changed. FIG. 63 is a diagram illustrating the second region partition information ID1bc generated by performing the change process on the second region partition information ID1b. As shown in FIGS. 61 to 63, the second undefined region PR12b is expanded to an undefined region PR12bc including the second undefined region PR12b and the third undefined region PR12c. Further, the second target confirmed region PR11b is changed to the third target confirmed region PR11bc from which the portion overlapping the third undefined region PR12c in the second target confirmed region PR11b is removed. Further, the second non-target defined region PR2b is changed to the third non-target defined region PR2bc from which the portion overlapping the third undefined region PR12c in the second non-target defined region PR2b is removed.

<(4-2-5) Precision detection unit>
The precision detection unit 315 detects the target region from the inside of the target part to the contour in the frame image IC0 based on the information of the target determination region PR11 and the undetermined region PR12 set by the rough detection unit 314. Perform detection processing. For example, based on the information of the first target confirmed region PR11a and the first unconfirmed region PR12a obtained by the rough detection process for the first frame image IC1, the contour from the inside of the target portion of the first frame image IC1 The first target region that reaches is detected. In addition, based on the information on the second target confirmed region PR11b and the undetermined region PR12bc obtained by the rough detection process on the second frame image IC2, the target region (the first region) related to the target portion in the second frame image IC2 2 target region) is detected. The undetermined region PR12bc is an undetermined region including the second undetermined region PR12b and the third undetermined region PR12c. The precise detection unit 315 of the present embodiment determines whether the undetermined region PR12 is inside or outside the lung field region as the target region.

  FIG. 64 is a diagram illustrating an example of the frame image IC0 input from the image correction unit 312 to the precision detection unit 315. FIG. 65 is a diagram illustrating an example of a lung field region RT1 as a target region detected from the frame image IC0 in the precision detection unit 315. In the precision detection process, the frame image IC0 (see FIG. 64) input from the image correction unit 312 is targeted based on the area classification information ID1 (see FIGS. 54 and 63) as a result of the rough detection process. A lung field region RT1 (see FIG. 65) as a region is detected.

  In the precise detection process, the lung field region RT1 as the target region is detected by a detection method that does not depend on the shape of the target part. Examples of detection methods include methods that use edge information such as level set and snake, methods that use region information such as region growing, and graph cuts (Graph Cut). ) And the like using edge and area information.

  In the Snake process, for example, an initial value of a closed curve is defined so as to surround a region to be detected, and the closed curve is deformed so as to satisfy the following conditions [i] to [iii] by repeated calculation. When the shape no longer changes, the deformation of the closed curve is terminated. By this processing, the contour ED1 of the lung field region RT1 as the target region is detected. Regarding the processing contents of Snake, for example, various known information (Hiroshi Mosquito “Chapter 11“ Regional Processing ””, Kyoto Sangyo University Faculty of Computer Science and Engineering, Department of Network Media “Spring Semester / Tuesday / Three Times” Search on November 18, 2013], Internet <www.cc.kyoto-su.ac.jp/~kano/pdf/course/chapter11.pdf>, and representative: Nobuhiko Yajima "Processing example | sliceOmatic" area / volume Examples of area definition processing for measurement ”disclosed on Image Laboratories website [searched on November 18, 2013], Internet <URL: http://www.imagelabo.com/Appli_sliceO.html>, etc.)

  [I] The shape of the closed curve itself tends to be continuous and smooth.

  [Ii] If the closed curve is on an edge, it tries to stay there.

  [Iii] The closed curve tends to be small.

  Here, in the method using edge information such as level set and snake, if there is information on the target defined region PR11 and the undefined region PR12 in the region classification information ID1, the target region A contour can be detected. In this case, the rough detection unit 314 may set the target candidate region PR1 including the target confirmed region PR11 and the undetermined region PR12 in the frame image IC0.

  In the region growing process, for example, first, a pixel that satisfies a condition relating to a pixel value is manually determined, and a single label is attached to the seed point. And the area | region is expanded by repeating the process in which the same label is attached | subjected to the pixel which satisfy | fills the conditions regarding a pixel value in the point of the seed point vicinity. In accordance with such processing, if the target confirmed region PR11 is used as a seed point and the region is expanded in the undetermined region PR12, the lung field region RT1 as the target region can be detected.

  Regarding graph cut processing, for example, publicly known information (Hiroshi Ishikawa “Theory and Application of Graph Cut”, 14th Image Sensing Symposium SS1108, [retrieved on November 18, 2013], Internet < URL: http://www.f.waseda.jp/hfs/SSIITutorial.pdf>).

  Information Rt2 indicating the result detected by the precision detection unit 315 is sent from the precision detection unit 315 to the target region extraction unit 316.

<(4-2-6) Target area extraction unit>
The target region extraction unit 316 extracts an image of a lung field region as a target region from each frame image I0 based on the target region detected by the precision detection process in the precision detection unit 315. Here, an image of a lung field region as a target region may be extracted from each frame image IC0 instead of each frame image I0.

<(5) Summary of one embodiment>
As described above, in the image processing device 3 according to the present embodiment, for example, in template matching for the first frame image IC1, the partial image related to the corresponding region corresponding to the template in the first frame image IC1 is self- A template. Then, the second undetermined region PR12b corresponding to the detection result of template matching using the self template for the second frame image IC2 is expanded according to the statistical value distribution, so that the undetermined region PR12bc is changed. can get. As a result, the occurrence of a problem in which the position of the target part is erroneously detected for a part in the body where a large movement is occurring is reduced. As a result, it is possible to accurately extract a region that captures the target region from a medical dynamic image that captures the movement of the structure inside the body.

<(6) Modification>
Note that the present invention is not limited to the above-described embodiment, and various modifications and improvements can be made without departing from the gist of the present invention.

  For example, in the above-described embodiment, for the plurality of frame images IC0 or the plurality of edge-enhanced frame images Ee0, the two-dimensional statistical values in the time direction relating to the feature values related to the pixel values for each image region Rv0 or the pixel values for each pixel Pm0 The distribution was calculated as a statistical value distribution. However, it is not limited to this. For example, for at least one of the plurality of frame images IC0 and the plurality of edge-enhanced frame images Ee0, there is a two-dimensional distribution of statistical values in the time direction regarding at least one of the feature value for each image region Rv0 and the pixel value for each pixel Pm0. It may be calculated as a statistical value distribution.

  In the above-described embodiment, a two-dimensional distribution of statistical values in the time direction is calculated by calculating one type of statistical value in the time direction for each frame image. However, the present invention is not limited to this. . For example, the first statistical value and the second statistical value in the time direction are calculated for at least one of each pixel Pm0 and each image region Rv0, and a combination of the first statistical value and the second statistical value is used to calculate a space. A two-dimensional distribution of statistical values in the direction may be calculated. Here, for example, the first statistical value and the second statistical value in the time direction relating to at least one of the feature value related to the pixel value for each image region Rv0 and the pixel value for each pixel Pm0 may be calculated. The combination of the first statistical value and the second statistical value is, for example, a two-dimensional distribution related to the third statistical value obtained by multiplying each first statistical value by the first coefficient according to a preset rule, Any combination by superimposing with a two-dimensional distribution related to the fourth statistical value obtained by multiplying each second statistical value by the second coefficient may be used. Accordingly, a two-dimensional distribution of statistical values that well reflects temporal changes of the structure captured in the dynamic image can be easily calculated by combining a plurality of types of statistical values in the time direction.

  Here, the coefficient may be a value (also referred to as a weighting coefficient) indicating weighting according to the priority order of statistical values. For example, the first coefficient and the second coefficient may be set so that the sum of the first coefficient multiplied by the first statistical value and the second coefficient multiplied by the second statistical value is 1. For example, when the first statistical value has a higher priority than the second statistical value, the first coefficient is set to a relatively large value (for example, 0.7 or more and 1.0 or less, etc.). Thus, a mode in which the second coefficient is set to a relatively small value (for example, not less than 0 and not more than 0.3) can be considered.

  Here, when the two-dimensional distribution related to the third statistical value and the two-dimensional distribution related to the fourth statistical value are combined, the sum of the third statistical value and the fourth statistical value is obtained for each pixel or each image region. As a result, the two-dimensional distribution of statistical values is calculated, but the present invention is not limited to this. For example, the two-dimensional distribution of statistical values may be calculated by adopting either the first statistical value or the second statistical value for a plurality of preset regions in each frame image. That is, the method of generating the statistical value in the time direction may be different for a plurality of preset regions in each frame image.

  In the above embodiment, the statistical value in the time direction is a numerical value or a representative value indicating the degree of dispersion in the time direction regarding the feature value related to the pixel value for each image region or the pixel value for each pixel. Not limited to. If the statistical value in the time direction is, for example, a numerical value indicating at least one of the characteristic value related to the pixel value for each image region and the pixel value for each pixel, the numerical value indicating the degree of scattering in the time direction and the representative value good. For example, the types of statistical values in the time direction may be different for a plurality of preset regions in each frame image.

  Further, in the above embodiment, an example in which the numerical value indicating the distribution degree as the feature value related to the pixel value for each image region and the distribution degree in the time direction is one of dispersion, standard deviation, and variation coefficient. However, the present invention is not limited to this. For example, the numerical value indicating the scatter degree as the feature value related to the pixel value for each image region and the scatter degree in the time direction may include at least one of variance, standard deviation, and variation coefficient.

  Moreover, in the said one Embodiment, although the dynamic image in which the structure in the body was caught by X-ray imaging was obtained, it is not restricted to this. For example, a dynamic image in which a structure in the body is captured may be obtained by other methods using CT, MRI, ultrasound, and the like.

  Moreover, in the said one Embodiment, although the object site | part was the lung field which exhibits a periodic motion, it is not restricted to this. For example, the target site may be other sites that cause displacement in the body, such as various joints.

  Further, in the above-described embodiment, the second frame image IC2 next to the first frame image IC1 with the partial image including the first target region detected by the precision detection processing for the first frame image IC1 as a self template. A rough detection process of the second target region was performed. However, it is not limited to this. For example, in a plurality of frame images constituting the dynamic image, the first frame image may be a frame image next to the second frame image. That is, the second image targeting the second frame image IC2 before the first frame image IC1 with the partial image including the first target region detected by the precision detection process targeting the first frame image IC1 as a self template. A rough detection process of the target area may be performed. In other words, the partial image including the target area detected by the precision detection process for the (M + 1) th frame image IC (M + 1) is used as a self-template, and the first image before the (M + 1) th frame image IC (M + 1). A rough detection process may be performed on the M frame image ICM.

  In the above-described embodiment, the statistical value distribution is used as it is in the coarse detection unit 314. However, the present invention is not limited to this. For example, the statistical value distribution may be binarized with reference to a preset threshold value before being used by the coarse detection unit 314. Further, for example, for each pixel, in the histogram indicating the frequency of the gray value, the numerical value relating to the frequency occupancy in the range of the preset width based on the mode value is based on the preset threshold. A statistical value distribution may be calculated by binarization. Further, for example, before being used in the coarse detection unit 314, a region in which a statistical value satisfies a preset condition in a statistical value distribution, and a region that forms a preset lump has a large movement. A process (labeling process) for making an area corresponding to the part may be performed.

  66 and 67 are diagrams illustrating how the statistical value distribution is binarized using a preset threshold as a reference. FIG. 66 shows an example of the statistical value distribution Ds3 before binarization, and FIG. 67 shows the statistical value distribution after binarization of the statistical value distribution Ds3. An example of Ds4 is shown. FIG. 68 shows an example of the statistical value distribution Ds5 after the statistical value distribution Ds3 is subjected to the labeling process.

  Further, in the above-described embodiment, a portion of the target determined area PR11 and the non-target determined area PR2 of the Mth area segment information ID1 generated by the Mth integration process is overlapped with the third undefined area PR12c. Although changed to PR12, it is not limited to this. That is, for example, the second undetermined region PR12b is expanded to the undetermined region PR12bc including the second undetermined region PR12b and the third undetermined region PR12c, but is not limited thereto. For example, the third unconfirmed region PR12c obtained from the statistical value distribution is set first, and then the third unconfirmed region PR12c is unconfirmed generated by the third unconfirmed region PR12c and the M-th integration process. The region may be expanded to an undetermined region PR12bc including the region PR12. For example, the third unconfirmed region PR12c obtained from the statistical value distribution is set first, and then the third unconfirmed region PR12c includes the third unconfirmed region PR12c and the second unconfirmed region PR12b. It may be expanded to the fixed region PR12bc.

  In the above-described embodiment, the number of frame images IC0 constituting the dynamic image has been described as 10 for convenience, but the present invention is not limited to this. The number of frame images IC0 constituting the dynamic image may be plural and may be set as appropriate.

  Needless to say, all or a part of each of the above-described embodiment and various modifications can be combined as appropriate within a consistent range.

DESCRIPTION OF SYMBOLS 1 Shooting device 2 Shooting control device 3 Image processing device 31 Control unit 31a Processor 31b Memory 32 Storage unit 100 Shooting system 311 Image acquisition unit 312 Image correction unit 313 Statistical value distribution calculation unit 314 Rough detection unit 315 Precision detection unit 316 Target region extraction Part Ds1 to Ds5 Statistical value distribution Ee0, Ee1, Ee2 Edge-enhanced frame image IC0 Frame image IC1 to IC10, ICM 1st to 10th, M frame image PR11 Target defined area PR11a, PR11b First and second target defined areas PR12, PR12bc Not yet Determined areas PR12a to PR12c First to third undefined areas PR2 Non-target defined areas PR2a, PR2b First and second non-target defined areas Ts1L, Ts1R, Ts2L, Ts2R Partial images (self-template)

Claims (12)

An acquisition unit for acquiring a dynamic image for medical use in which a structure inside the body is captured by X-ray irradiation;
Pixels for each image region including a plurality of pixels with respect to at least one of the plurality of edge-enhanced frame images and the plurality of frame images in which edges related to the structure in the plurality of frame images constituting the dynamic image are respectively enhanced A calculation unit that calculates a two-dimensional distribution of statistical values in the time direction related to at least one of a feature value indicating a feature of the value and a pixel value for each pixel;
With respect to the first frame image of the plurality of frame images, the template matching may be used to capture a first target determination region in which at least a part of the target part is captured and a part of the target part. A first detector for detecting a first undetermined region having
A second detection unit that detects a first target region from the inside of the target portion of the first frame image to an outline based on information on the first target fixed region and the first undefined region;
With
The first detection unit is
For a second frame image different from the first frame image among the plurality of frame images, in a corresponding region corresponding to the template of the first frame image in the template matching for the first frame image. Using template matching using such a partial image as a template, a second target determination region in which at least a part of the target part is captured, and a second unsuccessful part in which a part of the target part may be captured Performing a process of detecting a definite region and a process of detecting a third undetermined region in which a part of the target part may be captured based on the two-dimensional distribution of the statistical values,
The second detection unit is
Second from the inside of the target part of the second frame image to the contour based on the information on the second target fixed region, the second undefined region including the second undefined region and the third undefined region. An image processing apparatus for detecting a target area. The image processing apparatus according to claim 1,
The statistical value in the time direction is
An image processing device including the degree of dispersion in the time direction. The image processing apparatus according to claim 2,
The spreading degree is
An image processing apparatus including at least one of variance, standard deviation, and coefficient of variation. The image processing apparatus according to claim 1,
The statistical value in the time direction is
An image processing apparatus including a representative value in the time direction. The image processing apparatus according to claim 4,
The representative value is
An image processing apparatus including a value related to a difference between a maximum value and a minimum value. The image processing apparatus according to claim 4 or 5, wherein
The representative value is
An image processing apparatus including a value related to a frequency occupancy in a value range of a preset width with respect to a mode value, regarding at least one of the feature value for each image region and the pixel value for each pixel. The image processing apparatus according to any one of claims 1 to 6, wherein
The feature value is
An image processing apparatus including at least one representative value of an average value, a median value, a mode value, a maximum value, and a minimum value of pixel values. The image processing apparatus according to any one of claims 1 to 7, wherein
The calculation unit
For at least one of the plurality of edge-enhanced frame images and the plurality of frame images, a first statistical value in the time direction and a second value regarding at least one of the feature value for each image region and the pixel value for each pixel A statistical value is calculated, and a two-dimensional distribution relating to a third statistical value obtained by multiplying each first statistical value by a coefficient according to a preset rule; An image processing apparatus that calculates a two-dimensional distribution of the statistical values by a combination with a two-dimensional distribution related to a fourth statistical value obtained by multiplying the statistical values. The image processing apparatus according to any one of claims 1 to 8, wherein
The first detection unit is
An image processing apparatus that detects a non-target fixed region in which the target part is not captured for each of the first and second frame images. An image processing apparatus according to any one of claims 1 to 9, wherein
The first detection unit is
An image processing apparatus that detects a region satisfying a preset condition in the two-dimensional distribution of the statistical values as the third undetermined region. The image processing apparatus according to claim 10,
The first detection unit is
An image processing apparatus that detects, as the third unconfirmed region, an area in which a magnitude relationship between the statistical value and a preset threshold value in the two-dimensional distribution of the statistical value satisfies a preset condition.   A program for causing the information processing apparatus to function as the image processing apparatus according to any one of claims 1 to 11 when executed by a control unit included in the information processing apparatus.

Images