JP6379374B1 - Composite server device for remote evaluation of capillaries and capillary evaluation program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To expand the usability of a system for analyzing, diagnosing, and judging capillaries, and for leaking personal information of persons who have actually performed analysis, diagnosis, etc. of capillaries and unauthorized copying of the software Be able to eliminate the risk.
A cloud server system comprising a composite server group (0006) (0007) of a capillary blood vessel evaluation center (0001) constituting a cloud and the capillary blood vessel evaluation center (0001) via a communication network such as the Internet. Information terminals (0002) to (0004) such as personal computers that access the server groups (0006) and (0007), and a microscopic capillary blood vessel observation device connected to each of these information terminals (0002) to (0004) (0005).
[Selection] Figure 1

Description

  The present invention provides a composite in a remote evaluation system for capillary blood vessels in which a capillary image captured and acquired on the information terminal side is uploaded to the server side by a cloud server system and the diagnosis result and the like can be viewed on the information terminal side. It relates to server devices and the like.

  In the present application, the term “capillary vessel” includes not only subcutaneous blood vessels but also submucosal blood vessels.

  In general, capillaries change with age and physical condition, and it is said that observing the state of subcutaneous capillaries such as fingertips and fundus can grasp health conditions, lifestyle-related diseases, diseases, etc. Since blood vessels are very delicate and cannot be observed with the naked eye, diagnosis is performed using an observation device such as a microscope device.

  Conventionally, for example, as a health condition evaluation support system, an image processing means for processing a captured image of a capillary and recognizing the shape of the capillary in the image, and the abnormality of the capillary from the recognized shape of the capillary There is known a system that includes a calculation means for calculating a degree and finally determines an abnormality degree of a capillary vessel.

  In addition, a system for observing blood vessels and blood flow with a color moving image obtained by enlarging capillary blood vessels of the nail epithelium of the fingertip, and the density, thickness, shape, and blood flow velocity of the blood vessel in the obtained moving image are determined. Digitize, compare the numerical value with the numerical data of the reference table stored in the system in advance, classify it, analyze and evaluate all the classified data by analytical evaluation means, and the result as a comprehensive evaluation of health condition There is also known a medical diagnosis support system that outputs to a display means.

  In addition, the applicant of the present application has already used a capillary blood vessel observation device to sharpen and reveal the captured capillary image by special image processing based on the reaction-diffusion equation and quantify its morphological characteristics. We have proposed an analytical diagnostic device that evaluates capillaries.

JP, 2015-157071, A JP 2006-325714 A JP, 2006-202442, A

  According to the system for analyzing, diagnosing and evaluating capillaries described in the previous section, the state of various capillaries can be objectively determined based on a predetermined standard. In order to use it, in addition to an observation device such as a microscope device for photographing a capillary blood vessel, a personal computer or the like in which software for calculating and processing images of capillary blood vessels and other data obtained by the observation device is installed is required. . For this reason, the installation of such personal computers is usually limited to medical institutions such as health management centers, hospitals, and clinics that perform health examinations and simple examinations. Affects the accuracy of judgment results, and therefore, a certain level of skill is required for the users of the system. Therefore, the above-described capillary analysis / diagnosis / evaluation system should be widely used in various places. There was an inconvenience that it was not possible.

  For this reason, it is possible to expand the above-mentioned system using information communication such as the Internet. In this case, personal information related to imaging, observation and evaluation determination of capillaries may be leaked due to unauthorized access, or evaluation determination There is a risk that the software that performs the process is illegally copied.

  The purpose of the present invention is to expand the usability of a system for analyzing, diagnosing, and determining capillaries, and to leak out personal information of a person who has actually performed analysis, diagnosis, etc. of capillaries by the system. It is an object of the present invention to provide a composite server device or the like in a capillary remote evaluation system that can eliminate the risk of unauthorized copying.

The present invention described in claim 1 is a web interface server having a capillary blood vessel evaluation receiving unit that receives a capillary blood vessel evaluation from an information terminal, and an evaluation result web screen configuration unit that displays a capillary blood vessel evaluation result, and the capillary blood vessel evaluation reception. And a capillary blood vessel evaluation server having a capillary blood vessel image calculation unit having a capillary blood vessel image determination and evaluation unit for determining whether the capillary image uploaded to the unit is correct or not, and a capillary blood vessel analysis processing unit for analyzing the capillary blood itself This is a composite server device.

According to a second aspect of the present invention, in the composite server device according to the first aspect, the correctness / incorrectness of the capillary image in the capillary image discriminating / evaluating unit of the capillary blood vessel evaluation server is determined by the color tone of the image by auto white balance processing. It is characterized in that it is performed by standardization of the difference, standardization of the brightness difference of the image by automatic brightness correction processing, and determination of a predetermined range of HSV average value and standard deviation.

According to a third aspect of the present invention, in the composite server device according to the first or second aspect, the capillary image discrimination evaluation unit of the capillary blood vessel evaluation server also performs image quality judgment of the capillary blood vessel image. To do.

According to a fourth aspect of the present invention, in the composite server device according to the third aspect, the image quality determination by the capillary image discriminating / evaluating unit of the capillary blood vessel evaluation server is performed by detecting the light source reflection of the capillary image and the inclination of the capillary image It is performed by at least one or more calculation parameters of detection and focus detection of a capillary image.

The present invention according to claim 5 is the composite server device according to any one of claims 1 to 4, wherein the web interface server is accessed from an information terminal, and the information terminal is used. A user information storage unit for storing a person's ID and password, a capillary image storage unit for storing an image of a capillary test subject, and a user relationship management database having a subject information storage unit for storing the subject information. It is a thing.

According to a sixth aspect of the present invention, in the composite server device according to the fifth aspect, the subject information storage unit of the user relationship management database stores the non-personal information of the subject.

The present invention according to claim 7 causes one computer to function as a capillary blood vessel evaluation receiving unit that receives a capillary blood vessel evaluation and an evaluation result web screen configuration unit that displays a capillary blood vessel image evaluation result. , Function as a capillary image calculation unit having a capillary image discrimination evaluation unit for determining whether the capillary image uploaded to the capillary evaluation reception unit is correct and a capillary analysis processing unit for performing analysis processing of the capillary itself, In addition, the capillary computer image discriminating / evaluating unit of the other computer further includes at least one calculation parameter of light source reflection detection of the capillary blood vessel image, inclination detection of the capillary blood vessel image and focus detection of the capillary blood vessel image. Is a program for functioning as an image quality determination based on the above.

  When the composite server device of the present invention is used, the client server system has a practical advantage that remote capillary evaluation can be performed widely and easily using a general-purpose personal computer, a smart phone, and a tablet terminal.

  In addition, when uploading a capillary image using a general-purpose information terminal such as a personal computer or a smart phone, if a general photograph such as a landscape photograph or a person photograph that is not a capillary image is used by mistake, These false images are automatically eliminated and the evaluation can only be performed if the correct capillary image is uploaded.

  Further, according to the present invention, in response to variations in the quality of capillary images uploaded from a general-purpose personal computer, a smart phone, etc., quality evaluation of the image itself that has been evaluated along with the evaluation of capillaries Therefore, there is a practical advantage that the reliability of the evaluation result of the capillary blood vessels can be grasped.

  In the composite server device of the present invention, since the capillary blood vessel evaluation server that performs the evaluation calculation of the blood capillaries is configured separately from the web interface server for providing the cloud service, security of the capillary blood vessel evaluation server is ensured. In addition, the information on the user of the system and the information on the subject are not stored in the capillary evaluation server, and the information can be stored only in the web interface server. For the evaluation server, cybersecurity restrictions and risks can be avoided.

  In addition, the web interface server has an advantage that the design and specification can be arbitrarily designed as being separated from the core capillary evaluation server. In addition, as described above, user information and subject information are stored in the web interface server. In this system, this information includes information that does not include personal information such as name, address, and telephone number (non-personal information). This can also solve the problem of leakage of personal information on the cloud side.

1 is an overall view showing an outline of a cloud server type capillary remote evaluation system according to an embodiment of the present invention. It is a functional block diagram of the composite server apparatus concerning this invention. It is a conceptual diagram which shows an example of a capillary blood vessel observation apparatus. It is a functional block diagram which shows the structure of an information terminal. It is a flowchart which shows an example of the process in the composite server apparatus which concerns on this invention. It is a figure which shows an example of the evaluation result web screen by an evaluation result web screen structure part. It is a flowchart which shows an example of the process by a capillary vessel image discrimination | determination evaluation part. It is a flowchart which shows an example of the determination process of light source reflection. 6 is a flowchart illustrating an example of a focus match rate process. It is a flowchart which shows an example of the inclination determination process of a capillary vessel. It is a figure which shows the storage condition of the image folder in an information terminal. It is a light source reflection image before processing. It is a light source reflection image after processing. It is a related image of the focus matching rate before processing. It is a related image of the focus matching rate after processing. It is a related image of the inclination of the capillary blood vessel before a process. It is a relation image of the inclination of the capillary after processing. It is a functional block diagram which shows an example of the computer used by this invention.

  Next, although embodiment of this invention is described according to drawing, this invention is not limited to this embodiment.

  As shown in FIG. 1, the remote evaluation system for capillary blood vessels according to this embodiment is a so-called cloud server system, and the server group (0006) of the capillary blood vessel evaluation center (0001) constituting the cloud. 0007), and information terminals (0002) to (0004) such as personal computers accessing the server group (0006) (0007) of the capillary blood vessel evaluation center (0001) via a communication network such as the Internet, and these information terminals A microscopic capillary observation device (0005) connected to each of (0002) to (0004).

  In the remote evaluation system, the information terminals (0002) to (0004) and the capillary blood vessel observation devices (0005) are, for example, health care centers for performing health examinations, hospitals, and medical institutions such as various local clinics as well as caregivers. It is installed extensively in various places such as establishments that provide services. In general, the information terminals (0002) to (0004) and the capillary blood vessel observation devices (0005) are operated as users by the staff at each facility described above, and patients, health checkup examinees at each facility, In such cases, care recipients and the like receive evaluation and determination of their own capillaries as subjects.

  This system is so-called multi-device compatible with the information terminals (0002) to (0004), and various devices such as a tablet terminal and a smart phone can be used in addition to the personal computer. Further, the cloud in the cloud server system according to the present embodiment is configured in the above-described capillary blood vessel evaluation center (0001), for example, in a hierarchy such as a cloud base layer, a cloud service layer and a cloud application layer above it, and will be described later. An evaluation service regarding such capillaries and their image quality is provided.

  The server device used in the present invention functions at least by computer hardware and a computer program executed thereon.

As shown in FIG. 18, the computer (1801) is connected to, for example, a CD-ROM drive (1802), an MPU (1803), the MPU (1803), and the CD-ROM drive (1802). A bus (1804), a ROM (1805) for storing the program and the like, and a RAM (1) for temporarily storing application program instructions and providing a temporary storage space are connected to the MPU (1803). 1806), a hard disk (1807) for storing application programs, system programs, and data. However, these configurations are examples, and may be realized by other hardware configurations.

  That is, the capillary blood vessel evaluation center (0001) performs a capillary blood vessel evaluation process with the web interface server (0006) accessed from the information terminals (0002) to (0004) via a communication network such as the Internet. It has a capillary blood vessel evaluation server (0007).

  In addition, the web interface server (0006) performs management of information on users and subjects related to evaluation of capillaries, secure access management, web screen editing configuration, and the like, as will be described later.

Specifically, as shown in FIG. 2, the web interface server (0006) is a web interface for providing a cloud service related to capillary blood vessel evaluation and the like with the information terminals (0002) to (0004). (2001) and a user relationship management database (2002) for managing various information related to the user who requests the evaluation of capillary blood vessels and the like.

The web interface (2001) includes a capillary blood vessel evaluation receiving unit (2003) for receiving login user ID information and password user information and performing login control, and HTML for web images of a subject's capillary blood vessels. It has a configuration unit (2004) and an evaluation result web screen configuration unit (2005) indicating an evaluation result.

  The web interface (2001) has a function of performing communication connection with the information terminals (0002) to (0004) by a web browser.

As described above, the capillary blood vessel evaluation acceptance unit (2003) is stored in the user information storage unit (2007) of the user relationship management database (2002) described above by access from the information terminals (0002) to (0004). It has a function of checking whether or not login is possible by checking the login ID and password of a user who has been accessed, and is realized by a communication device and login control software.

The web HTML configuration unit (2004) has a function for forming a file or the like by web HTML, and is usually realized by creation software, but may be formed by a dedicated circuit.

As described above, the evaluation result web screen configuration unit (2005) has a function of configuring the evaluation result web screen indicating the evaluation result, and is realized by creation software. Specifically, a CAS screen configuration as shown in FIG.

  The user relationship management database (2002) includes a user log section (2006) for storing a user's usage history and the like, and user information storage for storing user information such as a user's login ID, password, and affiliated organization. Unit (2007), a capillary image storage unit (2008) for storing a capillary image of the subject, and a subject information storage unit (2009) for storing information such as the sex and age of the subject.

  As described above, the user log unit (2006) has a function of storing a usage history and the like, and is realized by software such as a memory.

  The user information storage unit (2007) is usually realized by a recording medium such as a hard disk.

  The capillary blood vessel image accumulation unit (2008) has a function of appropriately storing a subject's capillary blood vessel image, and is realized by a memory or hardware (dedicated circuit).

The subject information storage unit (2009) stores the reference number, age, and gender of the subject, but does not include the subject's personal information such as the subject's name, address, and telephone number. However, personal information may be exceptionally included when security risks are completely eliminated. The storage unit (2009) is usually realized by a recording medium such as a hard disk.

  The capillary blood vessel evaluation server (0007) includes an evaluation target image receiving unit (2011) that receives a capillary image of a subject from the web HTML constructing unit (2004) of the web interface server (0006), and the web HTML construct A parameter receiving unit (2012) for receiving calculation parameters to be described later from the unit (2004), a capillary vessel image received by the evaluation target image receiving unit (2011), and a calculation parameter received by the parameter receiving unit (2012). Based on the capillary image calculation unit (2013) for performing an evaluation process described later based on the results, the result image processed by the capillary image calculation unit (2013), image quality determination information, and the like are evaluated by the web interface server (0006). Calculation result output unit (2014) for output to result web screen configuration unit (2005) A.

  As described above, the evaluation target image receiving unit (2011) has a function of receiving a subject's capillary blood vessel image, and is realized by a wired or wireless receiving device and a driver that drives the device.

  The parameter receiving unit (2012) has a function of receiving a calculation parameter from the web HTML configuration unit (2004) of the web interface server (0006), a driver that drives with a wired or wireless receiving device, etc. It is realized by.

Examples of the calculation parameter include age / sex correction, automatic brightness adjustment in a capillary blood vessel image, which will be described later, and threshold setting for reflection / inclination / focus determination in the same image.

These calculation parameters are factors that affect the evaluation of capillaries, which will be described later. In general, the total length of capillaries becomes shorter as men and women get older, and the total length of capillaries is shorter in women than in men. Therefore, whether or not to perform the correction of the age and gender, and because there is a variation in the brightness of the capillary images acquired by the capillary image observation device, the average value of the brightness is calculated to be a constant value. Whether or not to perform automatic brightness correction, or whether to set a threshold value for a good image in the reflection / tilt / focus determination process.

A capillary blood vessel image calculation unit (2013) performs a capillary blood vessel analysis processing unit (2015) that performs analysis processing of the capillary blood vessel itself by a reaction-diffusion equation, which will be described later, and performs image discrimination as to whether or not it is a capillary blood vessel image and quality determination of the capillary blood vessel image. Capillary vessel image discrimination / evaluation unit (2016) to perform, and as described above, a function to perform analysis processing of capillary blood vessel itself, capillary image discrimination and quality evaluation in the manner described later using calculation parameters These functions are usually realized by software, and the software is stored in a recording medium such as a ROM in the server (0007).

For the evaluation analysis of the capillary blood vessel itself in the capillary blood vessel analysis processing unit (2015), for example, a method similar to that described in Japanese Patent Laid-Open No. 2006-202442 described above by the applicant of the present application is used.

That is, as shown in FIG. 3, the microscopic capillary observation device (0005) on the terminal side used by the user includes a finger rest 2 on which the finger F is placed and a nail on the finger F on the finger rest 2. An optical system unit 3 including a plurality of lenses for focusing on an observation target such as the floor portion F1, an illumination unit 4 for irradiating the observation target with visible light, infrared rays, and the like, a CCD, a CMOS, etc. for photographing the observation target A photographing unit 5 including the photographing element, a photographing control unit 6 such as a shutter for controlling the photographing element in the photographing unit 5, an image output interface 7 for outputting digital data of the photographed image, and digital of the photographed image. A transmission system 8 such as USB, NTSC, wireless, or the Internet for transmitting data is provided.

In the capillary blood vessel analysis processing unit (2015), the capillary blood vessel observation device (0005) is used to photograph the blood vessel by irradiating the body part with visible light or infrared light, and the photographed image is optically captured. For the enlarged image, the local state of the brightness of the enlarged image is automatically determined and a directional element based on the local threshold parameter is given using a reaction-diffusion equation described later in which the solution state changes with time. This is a capillary image processing method in which the shape of a capillary vessel is clarified by binarization, and the shape of the capillary vessel in this clarified image is digitized. Then, as described above, the diffusion reaction in the later-described equation performs image processing with directionality using the variable parameter a that varies depending on the previous calculation processing result.

The reaction diffusion equation is
Ε, b, c and d are all constant parameters, Du and Dv are diffusion parameters, and a is a variable parameter (threshold parameter) which varies depending on the result of the previous calculation process. And the digitization is the calculation of the length, thickness, and number of the longitudinal capillary vessels by clarifying the longitudinal capillary vessels by binarization by reaction diffusion in the lateral direction, or Is the calculation of the presence or absence of horizontal capillaries by binarization by reaction diffusion in the vertical and horizontal directions, and the length, thickness, and number of the capillaries when the capillaries are present, or This is a calculation of twisting and bending of capillaries thinned by binarization and skeletonization processing by reaction diffusion in the vertical and horizontal directions.

The capillary blood vessel discriminating / evaluating unit (2016) discriminates a capillary blood vessel image based on the aforementioned calculation parameters and evaluates the quality.

  As shown in FIG. 4, the information terminals (0002) to (0004) include a communication unit (4001) that accesses the capillary blood vessel evaluation center (0001) via a communication network such as the Internet, and the capillary blood vessel observation device. A connector section (4002) for connection to (0005), a capillary image storage section (4003) for storing the capillary image together with the ID of the subject, and a capillary image display section (4004) for displaying the capillary image And an input unit (4005) for inputting the user's ID and password, a capillary image acquisition unit (4006) for acquiring a capillary image from the capillary image storage unit (4003) and the connector unit (4002), and the like. Have.

  The communication unit (4001) is realized by a communication module or a modem, the connector unit (4002) is a connector of a hardware interface, a connection terminal or the like, and the capillary blood vessel image storage unit (4003) is a recording medium such as a hard disk. It is realized by software that records.

  In addition, the capillary blood vessel image display unit (4004) is usually a display, and in addition to the capillary blood vessel image, the evaluation of the capillary blood vessels viewed by accessing the capillary blood vessel evaluation center (0001), the image quality determination result, and the like. Is displayed.

  In addition, the input unit (4005) is a keyboard, a touch panel, or the like, and the capillary blood vessel image acquisition unit (4006) is usually realized by software.

  Next, as shown in FIG. 5, a description will be given of a related operation by the web interface server (0006) and the capillary blood vessel evaluation server (0007) in the aforementioned capillary blood vessel evaluation center (0001). First, the web interface server (0006) Then, login control from users using the information terminals (0002) to (0004) is performed (S5001), and user confirmation and user history storage are performed using a predetermined login ID and password (S5002). Next, the subject information from the information terminals (0002) to (0004) of the user and the capillary images of the subject are received (S5003), and these are stored in the user relationship management database (2002) (S5004). Thereafter, the capillary blood vessel image is transferred together with the calculation parameters to the capillary blood vessel evaluation server (0007) (S5005). The calculation parameters include, for example, whether or not to correct according to the age and sex of the subject (ON / OFF), ON / OFF of the automatic brightness adjustment function in the capillary image, and further, the light source in the capillary image described later For example, setting of a threshold value in determination of reflection, inclination, and focus can be given.

  The capillary blood vessel evaluation server (0007) receives the capillary blood vessel image and calculation parameters (S5006), and first determines whether or not the capillary blood vessel image is authentic by a method described later (S5007). ).

That is, as shown in FIG. 11, since the information terminals (0002) to (0004) are not dedicated terminals in the cloud-type remote blood vessel evaluation system, the image folder (1101) includes the capillary blood vessels. In addition to the capillary image (1102) acquired by the observation device (0005), a capillary image such as a landscape photograph (1103), an astronomical image (1104), a CG image (1105), a test image (1106), an illustration (1107), etc. In many cases, non-images are included in random, and if these images are sent by mistake, they are not automatically evaluated and are judged as errors, and the error results are transferred to the web interface server (0006). .

  On the other hand, when it is determined in the image determining step (S5007) that the image is a genuine capillary blood vessel image, a light source reflection determination (S5009), a focus match rate determination (S5010), and a reaction diffusion equation, which will be described later. Binarization (S5011) and capillary inclination determination (S5012) are performed, and the length, bend, number, total length, and the like are calculated as the evaluation of the capillary (S5013). Then, the light source reflection determination (S5009), the focus match rate determination (S5010), the binarization by the reaction diffusion equation (S5011), and the capillary inclination determination (S5012) are performed on the capillary image described above. The overall quality judgment, the evaluation of the capillary itself, and the processed capillary image are transferred to the web interface server (0006) (S5014). The transfer is also performed as an error screen when the above-described capillary image error determination is performed. Finally, the web interface server (0006) receives such various results and evaluation determinations, and the web screen is configured (S5015), which is saved as a history of calculation results (S1516). . The web screen is browsed and displayed on the information terminals (0002) to (0004). Note that the order in the processing as described above is not limited. For example, the determination of the light source reflection (S5009), the determination of the focus match rate (S5010), and the like can be arbitrarily performed.

  As shown in FIG. 6, as the web evaluation screen, for example, the gender, age, total length of capillaries, evaluation points of capillaries, and reliability as image quality judgment (maximum rating is 5 stars, By the deduction process, the lower evaluation from 4 stars to 1 star is displayed.

  As shown in FIG. 7, an example of automatic capillary image discrimination processing in the capillary blood vessel image discrimination evaluation unit (2016) of the capillary blood vessel image calculation unit (2013) of the capillary blood vessel evaluation server (0007) described above will be described automatically. As a large flow of discrimination, first, auto white balance (AWB) processing is performed to standardize the color difference for each image (S7001), and then automatic brightness correction processing is performed to standardize the brightness difference for each image ( (S7002) Finally, the capillary image is determined based on the determination parameter (S7003). Note that, in these discriminations, the discrimination parameters may be set manually or based on a deep learning method.

  More specifically, first, as standardization of the color difference (S7001), an RGB image is converted into YUV (S7001A), a background part is extracted from the luminance distribution (S7001B), and the background part is assumed to be achromatic. (S7001C), the achromatic target value is determined, and the RGB value for each pixel is corrected by multiplication (S7001D). After that, as standardization of the brightness difference (S7002), the RGB image is converted to grayscale (S7002A), the average luminance is calculated from the grayscale image (S7002B), and the RGB pixel luminance is calculated based on a predetermined luminance target value. Addition / subtraction calculation correction is performed (S7002C). Then, as a final discrimination process (S7003), the RGB image is converted into HSV (S7003A), the HSV average value and standard deviation for each pixel are obtained (S7003B), and the HSV average value and standard deviation are within a predetermined range. It is determined whether or not it is within the range (S7003C). At this time, if it is finally within the predetermined range, it is determined as a capillary blood vessel image, and if it is not within the predetermined range, it is determined not to be a capillary blood vessel image.

  As shown in FIGS. 8, 12, and 13, an example of the light source reflection determination process will be described. First, an RGB image is converted into a gray scale (S8001), and a reflection portion is extracted from the luminance distribution (S8002). ), The area of the reflecting portion is determined as a threshold (S8003), and it is detected whether or not the area is within a predetermined range (S8004). If the area is within the predetermined range, no deduction processing is performed for reliability evaluation. (S8005) If it exceeds the predetermined range, it affects the accuracy of the analysis / evaluation of capillaries, so deduction processing is performed depending on the accuracy (S8006), and the image extracted in the reflection portion extraction step (S8002) Based on this, a composite image (for reference) of the reflection part is created (S8007).

By the image processing of FIG. 8 described above, the white light reflection images of FIGS. 12A to 12C are converted into the images of FIGS. 13A to 13C, and FIGS. Among them, the image shown in FIG. 13 (a) has the widest black range indicating the reflection portion, so that the reflection range is wide, so that the determination process greatly reduces the points, and the images shown in FIGS. 13 (b) and 13 (c) show the black reflection. Since there are fewer parts than (a), the deductions are also reduced.

  As shown in FIG. 9, FIG. 14 and FIG. 15, an example of the focus matching rate determination process will be described. First, an RGB image is converted to gray scale (S9001), and then a contour is extracted by a contour extraction filter. (S9002) Local binarization is performed (S9003). Thereafter, the focus match amount or the focus match rate is calculated (S9004), the calculated values are determined as threshold values (S9005), and whether or not the result is within a predetermined range (S9006) is reached. If not, the deduction process is not performed (S9007). If the predetermined range is not reached, the deduction process is performed according to the degree (S9008), and based on the image extracted in the binarization step (S9003). A focus composite image (for reference) is created (S9009).

  By the image processing of FIG. 9 described above, the images of FIGS. 14A to 14C are converted into the images of the focus matching ratios of FIGS. 15A to 15C, and FIGS. 15 (a) and 15 (b) have a large range of white spots showing a focus match, so there are few or no deduction points, whereas FIG. 15 (c) shows a very wide range of white spots. Therefore, the evaluation is greatly reduced, and the evaluation of the capillaries based on such an image has low reliability.

  As shown in FIG. 10, FIG. 16, and FIG. 17, an example of the capillary image inclination determination processing by the capillary blood vessel evaluation server (0007) described above will be described. Binarization processing using an equation is performed (S1001), then tilt calculation processing for each capillary is performed (S1002), and it is finally determined whether or not the tilt is within an allowable range (S1003).

  More specifically, as the reaction diffusion binarization process (S1001), first, an RGB image is converted into a gray scale (S1001A), and then binarization is performed using a reaction diffusion equation (S1001B). (S1001C), and then, each capillary region in the image is specified (S1001D). Thereafter, as the tilt calculation process (S1002), the top of each capillary is detected (S1002A), the top of the head is deleted (S1002B), and the remaining straight blood vessels are labeled. An individual blood vessel is identified (S1002C), and then the individual blood vessel is surrounded by a rectangular frame, and the diagonal angle is calculated as an inclination value for each blood vessel (S1002D). Next, as the determination of the allowable range of inclination (S1003), first, a median value for each blood vessel is obtained (S1003A), and the median value is determined as an inclination threshold value (S1003B) to detect whether the inclination is within the allowable range. However, if it is within the allowable range, no deduction process is performed (S1003D). If the inclination is out of the allowable range, a deduction is evaluated according to the degree (S1003E).

  16 (a) and 16 (b) are converted into the tilt images of FIGS. 17 (a) and 17 (b), and FIGS. 17 (a) and 17 (b) are partially shown. However, since there are many capillary vessels with relatively little inclination as a whole, all of the images are judged to have no deduction or only a small deduction.

  According to the present invention, it is possible to perform appropriate evaluation of capillaries remotely, so that it can be widely used in the medical field, health care field, nursing field, and the like.

(0001) Capillary Evaluation Center (0002) to (0004) Information Terminal (0005) Capillary Observation Device (0006) Web Interface Server (0007) Capillary Evaluation Server (2001) Web Interface (2002) User Relationship Management Database ( 2003) Capillary evaluation acceptance unit (2004) Web HTML configuration unit (2005) Evaluation result web screen configuration unit (2013) Capillary vessel image calculation unit (2015) Capillary vessel analysis processing unit (2016) Capillary vessel image discrimination evaluation unit

Claims (7)

  A capillary blood vessel evaluation receiving unit that receives a capillary blood vessel evaluation from an information terminal, a web interface server having an evaluation result web screen configuration unit that displays a capillary blood vessel evaluation result, and a capillary image uploaded to the capillary blood vessel evaluation receiving unit A composite server device including a capillary blood vessel evaluation server including a capillary blood vessel image calculation unit having a capillary blood vessel image determination / evaluation unit for performing determination processing and a capillary blood vessel analysis processing unit for performing analysis processing of the capillary blood vessel itself.   The capillary image discrimination evaluation unit of the capillary blood vessel evaluation server determines whether the capillary image is correct or not by standardizing the color difference of the image by auto white balance processing, standardizing the brightness difference of the image by automatic luminance correction processing, and the HSV average value and The composite server device according to claim 1, wherein the composite server device is performed by determining a predetermined range of the standard deviation.   3. The composite server device according to claim 1, wherein the capillary blood vessel image discrimination evaluation unit of the capillary blood vessel evaluation server also performs image quality determination of the capillary blood vessel image.   The image quality determination by the capillary image discriminating / evaluating unit of the capillary blood vessel evaluation server is performed by at least one calculation parameter among the light source reflection detection of the capillary blood vessel image, the inclination detection of the capillary blood vessel image, and the focus detection of the capillary blood vessel image. The composite server device according to claim 3.   A web interface server is accessed from an information terminal, a user information storage unit that stores an ID or password of a user of the information terminal, a capillary image storage unit that stores an image of a capillary test subject, and the subject The composite server apparatus as described in any one of Claims 1-4 provided with the user relationship management database which has the test subject information storage part which memorize | stores the information of.   6. The composite server device according to claim 5, wherein the subject information storage unit of the user relationship management database stores non-personal information of the subject. One computer,
A capillary blood vessel evaluation receiving unit for receiving a capillary blood vessel evaluation;
Function as an evaluation result web screen component that displays the evaluation result of the capillary image,
The other computer
Function as a capillary image calculation unit having a capillary image discriminating / evaluating unit for determining whether the capillary image uploaded to the capillary vessel evaluation receiving unit is correct and a capillary vessel analyzing unit for analyzing the capillary itself,
And the other computer is
The capillary image discriminating / evaluating unit further performs image quality determination based on at least one calculation parameter among the light source reflection detection of the capillary blood vessel image, the inclination detection of the capillary blood vessel image, and the focus detection of the capillary blood vessel image. Program to function as.