Determination device, liquid level control system and determination method

Abstract

To determine a position of a liquid level of liquid stored in a container with further high accuracy.SOLUTION: A determination device includes a control unit that acquires a captured image obtained by capturing an image of a state where determination object liquid is stored in a prescribed container, extracts a vicinity area that is highly likely to contain an image representing a liquid surface of the liquid on the basis of features corresponding to color tones in the captured image, and determines a straight line representing the liquid surface of the liquid from the image of the vicinity area.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a determination device, a liquid level control system, and a determination method. [Background technology]

[0002] In a blood circulation system, a blood reservoir is used to store blood, and control is performed to maintain the position (height) of the liquid surface in the blood reservoir. For such control, it is necessary to determine the position of the liquid surface. Conventionally, the position of the liquid surface has been determined using ultrasound or capacitance technology. However, it is difficult to obtain continuous values ​​with these technologies. Therefore, a technology has been proposed that determines the position of the liquid surface using images obtained by photographing the blood reservoir, such as the technology disclosed in Patent Document 1. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 7038698 Summary of the Invention [Problem to be solved by the invention]

[0004] However, there is a demand for determining the position of the liquid level with higher accuracy, and this demand is not necessarily limited to determining the height of the liquid level in the blood reservoir. The present invention has been made in consideration of the above-mentioned circumstances, and provides a technique that makes it possible to determine the position of the liquid surface of a liquid stored in a container with higher accuracy. [Means for solving the problem]

[0005] One aspect of the present invention is a determination device that includes a control unit that acquires an image obtained by capturing an image of a liquid to be determined stored in a specified container, extracts a nearby area in the captured image that is likely to contain an image showing the liquid surface of the liquid based on features corresponding to the color tone, and determines a straight line showing the liquid surface of the liquid from the image of the nearby area.

[0006] In one aspect of the present invention, in the determination device, the control unit detects a plurality of straight lines from the image of the vicinity region, and determines a straight line indicating a liquid surface from among the plurality of straight lines.

[0007] One aspect of the present invention is the above-mentioned determination device, wherein the control unit determines the straight line indicating the liquid level from among the straight lines having an angle within a predetermined range that is expected to be the angle formed by the liquid level.

[0008] In one aspect of the present invention, in the determination device, the feature amount according to the hue is expressed using a value related to at least saturation or brightness.

[0009] In one aspect of the present invention, in the determination device, the control unit further determines the amount of the liquid according to a determination result of a straight line indicating the liquid level.

[0010] One aspect of the present invention is the above-mentioned judgment device, wherein the specified container has multiple types, and further includes a memory unit that stores a judgment model for determining the type of the container from the captured image, and the control unit judges the type of the container shown in the captured image based on the captured image and the judgment model.

[0011] One aspect of the present invention is the above-mentioned determination device, wherein the memory unit further stores information indicating a processing area, which is an area on the image to be processed for each type of container, and the control unit performs processing based on the processing area corresponding to the determined type of container.

[0012] In one aspect of the present invention, in the determination device, the control unit determines whether or not an error has occurred in the captured image based on a predetermined error criterion.

[0013] One aspect of the present invention is a liquid level control system that includes a determination device that acquires an image obtained by capturing an image of a liquid to be determined accumulated in a specified container, extracts a nearby area in the captured image that is likely to contain an image showing the liquid level of the liquid based on features corresponding to the color tone, and determines a straight line showing the liquid level from the image of the nearby area, and a pump control device that controls the operation of a pump that pumps the liquid in the container to the outside based on the straight line showing the liquid level obtained by the determination device.

[0014] One aspect of the present invention is the above-mentioned liquid level control system, further comprising a display device that displays the captured image, and the display device displays a straight line indicating the liquid level determined by the determination device superimposed on the captured image.

[0015] One aspect of the present invention is the above-mentioned liquid level control system, wherein the control unit further determines the amount of liquid based on the determination result of the straight line indicating the liquid level of the liquid, and the display device further displays information indicating the amount of liquid determined by the determination device.

[0016] One aspect of the present invention is a determination method comprising the steps of: acquiring an image obtained by capturing an image of a liquid to be determined stored in a specified container; extracting a nearby region in the captured image that is likely to contain an image showing the liquid surface based on features corresponding to the color tone; and determining a straight line showing the liquid surface from the image of the nearby region. [Effects of the Invention]

[0017] According to the present invention, it is possible to determine the position of the liquid surface of a liquid stored in a container with higher accuracy. [Brief explanation of the drawings]

[0018] [Figure 1] 1 is a schematic block diagram showing the system configuration of a liquid level control system 100 of the present invention. [Figure 2] 2 is a schematic block diagram showing a specific example of the functional configuration of a determination device 50. FIG. [Figure 3] 10 is a flowchart showing a specific example of the flow of processing by the determination device 50. [Figure 4] FIG. 1 is a diagram showing an area to be analyzed in a captured image. [Figure 5] FIG. 5 is a diagram showing color feature amounts of the image shown in FIG. 4. [Figure 6] This is a specific example of a grayscale image that is drawn by calculating the value of the feature S / V for each pixel of the target image and using that value as the pixel value. [Figure 7] FIG. 10 is a diagram showing the results of clustering the feature S / V into three categories. [Figure 8] This is a specific example of a binary image in which each of the two divided regions is displayed in binary. [Figure 9] FIG. 10 is a diagram showing a specific example of an image showing a neighboring region. [Figure 10] This is an image obtained by extracting a nearby area from the target image. [Figure 11] FIG. 10 is a diagram showing an outline of a liquid surface candidate acquisition process. [Figure 12] 10A and 10B are diagrams showing a plurality of liquid level candidates obtained by liquid level candidate acquisition processing. [Figure 13] FIG. 2 is a diagram illustrating an outline of an example of the hardware configuration of an information processing device 90 applied to the present embodiment. [Figure 14] FIG. 10 is a diagram showing a modified example of the liquid level control system 100. [Figure 15] 1 is a diagram showing an example of a display in the liquid level control system 100. FIG. [Figure 16] FIG. 10 is a diagram showing an example of an error display in the liquid level control system 100. DETAILED DESCRIPTION OF THE INVENTION

[0019] 1 is a schematic block diagram showing the system configuration of a liquid level control system 100 of the present invention. The liquid level control system 100 includes a blood storage tank 10, a holding device 20, an imaging device 30, a pump 40, a determination device 50, and a pump control device 60. The imaging device 30 and the determination device 50 communicate to transmit an image captured by the imaging device 30 to the determination device 50. The determination device 50 and the pump control device 60 communicate to transmit information indicating the determination result of the determination device 50 to the pump control device 60. The pump 40 and the pump control device 60 communicate to transmit a control signal from the pump control device 60 to the pump 40. These communications may be performed by wired communication or wireless communication.

[0020] The blood reservoir 10 is a specific example of a container that is the target of the liquid level determination process. The blood reservoir 10 stores blood transported from a living body (e.g., a human body) that is the target of treatment via a blood removal circuit or other means. The blood reservoir 10 is made of a light-transmitting material and has a structure that allows the liquid level of the liquid (blood) stored inside to be seen from the outside.

[0021] One end of the holding device 20 is fixed to a part of the blood reservoir 10. The other end of the holding device 20 is fixed to the imaging device 30. The holding device 20 has a predetermined level of rigidity or more. With this configuration, the holding device 20 keeps the relative positional relationship between the blood reservoir 10 and the imaging device 30 and the orientation of the imaging device 30 constant.

[0022] The imaging device 30 generates image data by capturing at least visible light. The imaging device 30 captures color images such as RGB. The imaging device 30 is held by the holding fixture 20 so as to capture at least an image of the area where the liquid level of the blood reservoir 10 is expected to be present.

[0023] The pump 40 pumps the blood stored in the blood reservoir 10 to the outside of the blood reservoir 10. The pump 40 pumps the blood to an instrument such as an oxygenator. The operation of the pump 40 is controlled by a pump control device 60.

[0024] The determination device 50 determines the level of the liquid (blood) in the blood reservoir 10 based on the image data obtained by the imaging device 30. The determination device 50 passes information indicating the determination result to the pump control device 60.

[0025] The pump control device 60 determines the amount of liquid in the blood reservoir 10 based on the information on the liquid level obtained by the determination device 50, and controls the operation of the pump 40 based on the amount of liquid.

[0026] 2 is a schematic block diagram showing a specific example of the functional configuration of the determination device 50. The determination device 50 is configured using information equipment such as a single-board computer, a smartphone, a tablet, a personal computer, or a dedicated device. The determination device 50 includes a communication unit 51, a storage unit 52, and a control unit 53.

[0027] The communication unit 51 is a communication device. The communication unit 51 may be configured using, for example, a communication cable interface or wireless communication. The communication unit 51 communicates data with other devices according to the control of the control unit 53. For example, the communication unit 51 communicates with the imaging device 30 and the pump control device 60. The communication unit 51 may be a device that performs wireless communication or a device that performs wired communication. The communication unit 51 that communicates with the imaging device 30 and the communication unit 51 that communicates with the pump control device 60 may be configured using different hardware, or may be configured using common hardware.

[0028] The storage unit 52 is configured using a storage device such as a magnetic hard disk drive, a semiconductor storage device, etc. The storage unit 52 stores data used by the control unit 53.

[0029] The control unit 53 is configured using a processor such as a CPU (Central Processing Unit) and a memory (main storage device). The processor executes a program, causing the control unit 53 to function as an error determination unit 531, a nearby region extraction unit 532, a liquid level candidate acquisition unit 533, a liquid level determination unit 534, and a correction processing unit 535. Note that all or part of the functions of the control unit 53 may be realized using hardware such as an ASIC (Application Specific Integrated Circuit), a PLD (Programmable Logic Device), or an FPGA (Field Programmable Gate Array). The above program may be recorded on a computer-readable recording medium. Examples of computer-readable recording media include portable media such as a flexible disk, a magneto-optical disk, a ROM, a CD-ROM, and a semiconductor storage device (e.g., a solid-state drive (SSD)), as well as storage devices such as a hard disk or semiconductor storage device built into a computer system. The above program may be transmitted via a telecommunications line.

[0030] The control unit 53 may execute, for example, an application installed in its own device (determination device 50). A specific example of such an application is an application provided to the determination device 50 as a dedicated application for the liquid level control system 100. The application may be installed in the determination device 50 in advance, or may be downloaded each time a determination process is executed. The control unit 15 operates according to the program of the application being executed.

[0031] The error determination unit 531 performs error determination processing based on a predetermined error criterion. The error determination unit 531 may determine that an error has occurred, for example, when an image to be processed (hereinafter referred to as a "target image") cannot be acquired from the imaging device 30, when a predetermined pattern cannot be detected from the target image, or when pixels of a predetermined color tone cannot be detected from the target image. The region to be processed to detect the predetermined pattern (hereinafter referred to as an "error determination region") may be determined in advance.

[0032] The predetermined pattern may be, for example, a pattern showing the shape of a part of the whole or part of the blood reservoir 10 that is originally intended to appear in the target image, or a predetermined pattern (for example, a barcode or two-dimensional barcode image) attached or formed on the blood reservoir 10, or may be formed in other ways. The predetermined color tone may be, for example, the color tone of a part of the whole or part of the blood reservoir 10 that is originally intended to appear in the target image, or the color tone of a liquid (for example, blood) that is intended to be stored in the blood reservoir 10. The color tone may be represented, for example, by a hue value, a combination of hue and brightness values, or other values ​​that indicate color.

[0033] If the error determination unit 531 determines that an error has occurred, it performs error processing. In the error processing, the error determination unit 531 performs processing to notify nearby people, an administrator, or the like that an error has occurred, for example, using an output device (not shown). For example, the error processing may include sounding an error from an audio output device provided in the determination device 50 or another device, lighting a lamp, or displaying an image indicating the error on an image output device.

[0034] The nearby region extraction unit 532 executes a process (nearby region extraction process) to extract a nearby region from the target image. The nearby region is a region of the target image that is likely to include an image showing the liquid surface. In other words, the nearby region is a region of the target image that is near the liquid surface. In this embodiment, the nearby region corresponds to a region that is likely to include the blood liquid surface.

[0035] The liquid surface candidate acquisition unit 533 performs a process of acquiring pixel regions that are candidates for the liquid surface from the image of the nearby region (liquid surface candidate acquisition process). The liquid surface candidate acquisition unit 533 acquires one or more pixel regions that satisfy predetermined conditions as liquid surface candidates.

[0036] The liquid level determination unit 534 performs a process of determining the liquid level (liquid level determination process). The liquid level determination unit 534 determines, from among the liquid level candidates acquired by the liquid level candidate acquisition unit 533, the one with the highest likelihood as a liquid level as a pixel region of the liquid level.

[0037] The correction processing unit 535 performs a correction process on the determined liquid level. The correction processing unit 535 may perform the correction process on the newly determined liquid level, for example, based on information on the pixel area of ​​the liquid level previously obtained by the liquid level determination process. As a specific example of such a correction process, a Kalman filter may be applied.

[0038] FIG. 3 is a flowchart showing a specific example of the processing flow of the determination device 50. First, the error determination unit 531 determines whether an error has occurred (step S101). If an error has occurred (step S101-YES), the error determination unit 531 executes error processing (step S102). On the other hand, if no error has occurred (step S101-NO), the nearby area extraction unit 532 executes nearby area extraction processing (step S103). Next, the liquid level candidate acquisition unit 533 acquires a liquid level candidate (step S104). Next, the liquid level determination unit 534 determines the liquid level from multiple liquid level candidates (step S105). Next, the correction processing unit 535 performs correction processing on the liquid level (step S106). The determination device 50 repeatedly executes the processing of steps S101 to S106 at a predetermined cycle. The determination device 50 outputs information indicating the determined liquid level to another device (for example, the pump control device 60). Based on the determined liquid level, the determination device 50 may determine the amount of liquid (e.g., blood) accumulated in the blood reservoir 10. In this case, the determination device 50 may output the amount of liquid (blood) to the pump control device 60 as the determination result.

[0039] FIG. 4 is a diagram illustrating an outline of the neighboring region extraction process. The image shown in FIG. 4 is an image showing a portion of the blood reservoir 10 captured by the imaging device 30. A predetermined portion of the captured image (hereinafter referred to as the "processing target region") may be used as the processing target. The processing target region may be predetermined on the image, for example. For example, if the container (blood reservoir 10), the holding device 20, the angle of view of the imaging device 30, etc. are predetermined, it is determined which part of the container will be captured in which part of the captured image. Therefore, it is possible to define the processing target region on the image in advance based on such properties. The horizontal and vertical axes of FIG. 4 indicate positions on the image along the respective axes (e.g., pixel coordinates). The image shown in FIG. 4 includes a first region 71, a second region 72, and a third region 73.

[0040] The first region 71 is an upper region of the blood reservoir 10 that has not come into contact with blood. The second region 72 is a region near the center of the blood reservoir 10 that has come into contact with blood in the past but does not currently contain blood. Due to the influence of the material of the blood storage portion provided inside the blood reservoir 10, although no blood is currently stored, some blood from past contact remains, resulting in the blood color remaining. The third region 73 is a lower region of the blood reservoir 10 that contains blood. Therefore, the third region 73 shows the color of blood. A liquid level exists between the second region 72 and the third region 73. In the example of Figure 4, it can be seen that the liquid level exists around 250 on the vertical axis.

[0041] FIG. 5 is a diagram showing color features of the image shown in FIG. 4. The horizontal axis in the graph of FIG. 5 indicates the position along the vertical axis in FIG. 4. The left side of the graph of FIG. 5 indicates the upper side of FIG. 4, and the right side of the graph of FIG. 5 indicates the lower side of FIG. 4. When the value on the horizontal axis in FIG. 5 is the same as the value on the vertical axis in FIG. 4, the same position is indicated. The value shown in the graph of FIG. 5 may be represented, for example, using the value of a pixel at a predetermined position (column) on the horizontal axis of the target image, or may be represented using a statistical value (e.g., average value) of the values ​​of multiple pixels lined up horizontally at each vertical axis value.

[0042] The vertical axis of FIG. 5(A) represents saturation (S) and lightness (V). That is, FIG. 5(A) shows the change in the average saturation and lightness values ​​along the vertical axis of the image (target image) shown in FIG. 4. FIG. 5(B) shows the change in the feature quantity related to color tone along the vertical axis of the target image. In the example of FIG. 5(B), S / V is used as a specific example of the feature quantity. As is clear from FIG. 5, in the first region 71, V is high and S is low, so S / V takes a very low value. In the second region 72, saturation increases and lightness decreases as one approaches the liquid surface. In the third region 73, saturation is almost entirely high and lightness decreases. As a result, the feature quantity S / V increases from the first region 71 to the second region 72 and the third region 73.

[0043] Fig. 6 shows a specific example of a grayscale image drawn by calculating the value of the feature S / V for each pixel of the target image and using that value as the pixel value. The pixel values ​​in Fig. 6 may be normalized values ​​of the feature S / V between 0 and 255, for example. The pixel values ​​increase (closer to 255, closer to white) from the first region 71 through the second region 72 to the third region 73.

[0044] FIG. 7 shows the results of clustering the feature S / V into three categories. In FIG. 7, the images are clustered into three categories: black (the area of ​​the lowest feature class), gray (the area of ​​the intermediate feature class), and white (the area of ​​the highest feature class). The color of blood is characterized by high saturation and low brightness. On the other hand, an empty blood reservoir is highly transparent, resulting in low saturation and high brightness. Therefore, for example, if the container is a blood reservoir and the liquid stored inside is blood, each of the first to third regions 71 to 73 has the following characteristics: The region of the blood reservoir that has not come into contact with blood (first region 71) has a low feature S / V; the region of the blood reservoir that has come into contact with blood but does not currently contain blood (second region 72) has the next lowest feature S / V; and the region containing blood (third region 73) has a high feature S / V. Therefore, by clustering the images into three categories based on the feature S / V as described above, the first to third regions 71 to 73 can be accurately detected from the image.

[0045] Such clustering may be performed using, for example, the K-Means algorithm with k=3. A predetermined region including a region with the highest feature value S / V may be extracted as the neighboring region. Instead of the feature value S / V, the pixel values ​​shown in FIG. 6 may be used. Specifically, the neighboring region may be determined by, for example, the following process.

[0046] First, the three clusters are divided into two clusters. Specifically, the gray area (area of ​​a class with intermediate feature amounts) and one of the remaining two areas whose feature amounts are closer to the gray area are combined into one area. The remaining two areas are the black area (area of ​​a class with the lowest feature amount) and the white area (area of ​​a class with the highest feature amount). The feature amounts compared at this time may be, for example, representative feature amounts for each area (such as representative values ​​used in the K-Means method), average feature amounts, or some other statistical value. The following explanation deals with an example in which the gray area and the black area are combined into one area.

[0047] The target image is divided into two regions by combining the black region (region of the class with the lowest feature amount) and the gray region (region of the class with the intermediate feature amount) into one region. FIG. 8 is a specific example of a binary image in which each of the two regions thus divided is displayed in binary. From the entire region including these two regions, a predetermined region including the boundary between the white region (region of the class with the relatively high feature amount) and the black region (region of the class with the relatively low feature amount) is extracted as the neighboring region. FIG. 9 is a diagram showing a specific example of an image showing the neighboring region. For example, as shown in FIG. 9, a region including a predetermined number of pixels (e.g., 5 pixels, 10 pixels, etc.) further to the left of the white region located at the leftmost position (upper direction) may be acquired as the neighboring region.

[0048] 10 is an image (hereinafter referred to as a "neighborhood area image") obtained by extracting a neighborhood area from a target image. The neighborhood area image may be used as a color image, like the target image, or as a grayscale image.

[0049] Next, the liquid level candidate acquisition process will be described. The liquid level candidate acquisition unit 533 acquires a liquid level candidate by processing the nearby area image. FIG. 11 is a diagram showing an outline of the liquid level candidate acquisition process. The liquid level candidate acquisition unit 533 extracts straight lines that meet predetermined criteria in the nearby area image. FIG. 11 shows an image in which the extracted straight lines are shown in white. Straight lines may be extracted using any algorithm. For example, edges may be extracted in the nearby area image, and straight lines may be extracted based on the extracted edges. The liquid level candidate acquisition unit 533 extracts, as liquid level candidates, straight lines with angles within a predetermined range that include the angle of the straight line assumed to be the liquid level from the extracted straight lines. In the example of FIG. 11, an angle range of 45 degrees forward and backward, including the horizontal, is defined as the angle of the straight line assumed to be the liquid level. The angle range may be any range as long as it includes the angle of the straight line assumed to be the liquid level. The liquid level candidate acquisition unit 533 may also perform a process of forming a longer straight line by connecting straight lines that have similar angles and whose ends are close to each other on the image. FIG. 12 is a diagram showing a plurality of liquid level candidates obtained by the liquid level candidate acquisition process.

[0050] The liquid level determination unit 534 selects a line indicating the actual liquid level from the multiple liquid level candidates obtained by the liquid level candidate acquisition process. The liquid level determination unit 534 may use any algorithm to select a line indicating the actual liquid level from the multiple liquid level candidates. For example, the liquid level determination unit 534 may calculate a predetermined feature (likelihood) indicating the likelihood of a liquid level for each liquid level candidate, and acquire the line with the highest likelihood based on the feature as the line indicating the actual liquid level. Such a feature may be obtained, for example, based on the liquid level determined at a previous point in time on the time axis, or based on the amount of edges on the line of the liquid level candidate, or based on the degree of distortion in the upper and lower regions of the line of the liquid level candidate.

[0051] The liquid level control system 100 configured in this manner determines a nearby area including the area where blood is pooling (third area 73) based on the captured image, and determines the liquid level in the nearby area. By performing such processing, the liquid level control system 100 can accurately determine the liquid level.

[0052] The liquid level determination unit 534 may determine the amount of liquid (blood) stored in the blood reservoir 10 according to the position of the liquid level, which is the determination result. If the angle of view of the container (blood reservoir 10), the holding device 20, the imaging device 30, etc. are predetermined, the amount of liquid in the container and the position of the liquid level in the captured image can be defined in a one-to-one correspondence. Therefore, based on such properties, the liquid level determination unit 534 can further determine the amount of liquid according to the position of the liquid level on the image. In this case, the storage unit 52 may previously store a table that associates the position of the liquid level with the amount of liquid.

[0053] Furthermore, the liquid level control system 100 is not only able to determine the liquid level when it reaches a specific amount, but is also able to continuously determine the position of the liquid level as long as the liquid level is present within the angle of view captured by the imaging device 30.

[0054] 13 is a diagram showing an outline of an example of the hardware configuration of an information processing device 90 applied to this embodiment. The information processing device 90 includes a processor 91, a main memory device 92, a communication interface 93, an auxiliary memory device 94, an input / output interface 95, and an internal bus 96. The processor 91, the main memory device 92, the communication interface 93, the auxiliary memory device 94, and the input / output interface 95 are communicably connected to each other via the internal bus 96. The information processing device 90 may be applied to, for example, the determination device 50. In this case, for example, the communication unit 51 may be configured using the communication interface 93. For example, the memory unit 52 may be configured using the auxiliary memory device 94. Furthermore, the control unit 53 may be configured using the processor 91 and the main memory device 92.

[0055] (Variation) In this embodiment, the determination device 50 and the pump control device 60 are configured as separate devices, but they may also be configured as an integrated device. The determination device 50 is connected to the imaging device 30 and the pump control device 60 via a network, and may be implemented using a server device or a cloud.

[0056] FIG. 14 is a diagram showing a modified example of the liquid level control system 100. The liquid level control system 100 may further include an image display device 80. The image display device 80 is configured using a device for displaying images such as a liquid crystal display. A terminal device equipped with a display (for example, a personal computer, tablet, smartphone, etc.) may be used as the image display device 80. The determination device 50 outputs to the image display device 80 the image data itself captured by the imaging device 30, or image data in which a determination result or the like is superimposed on the image data. The image display device 80 displays the image data output from the determination device 50. The determination device 50 and the image display device 80 may be connected by wired communication or wireless communication.

[0057] FIG. 15 is a diagram showing an example of a display in the liquid level control system 100. In the image shown in FIG. 14 , a frame image 81, a liquid level image 82, and a liquid volume image 83 are further displayed in the image captured by the imaging device 30. The frame image 81 is an image showing the processing target area. Note that the frame image 81 is a specific example of an image showing the processing target area and does not necessarily have to be a frame. Furthermore, the frame image 81 does not necessarily have to be displayed; for example, the user may be able to select whether or not to display it. The liquid level image 82 is an image showing the liquid level determined by the determination device 50. For example, the liquid level image 82 may be displayed as a bar-shaped image of a predetermined color, as a frame surrounding the liquid level portion, or as a figure such as a rectangle showing the liquid level and liquid portion. The liquid volume image 83 is an image showing the amount of liquid. The liquid volume image 83 may be displayed, for example, as a numerical value showing the amount of liquid, as words showing the relative amount of liquid (high, normal, low, etc.), or as a figure showing the relative amount of liquid.

[0058] FIG. 16 is a diagram showing an example of an error display in the liquid level control system 100. In FIG. 15, an obstructing object is reflected in front of the blood storage tank 10, and the blood storage tank 10 is not correctly captured, resulting in an error. When an error is detected, a character string or image indicating the error may be displayed. In the example of FIG. 15, a character string indicating the error is displayed in the upper left corner, but this is not a limitation. Furthermore, when an error is detected, some or all of the frame image 81, the liquid level image 82, and the liquid volume image 83 may not be displayed. When an error occurs, it is possible that an incorrect judgment has been made. Therefore, by intentionally not displaying these images to prevent people from making decisions based on incorrect judgment results, it is possible to prevent accidents. In the example of FIG. 15, the liquid level image 82 and the liquid volume image 83 are not displayed.

[0059] There may be multiple types of blood reservoirs 10. In this case, the storage unit 52 may store in advance a table associating the position of the liquid level with the amount of liquid for each blood reservoir 10. This configuration allows the liquid level determination unit 534 to determine the amount of liquid more accurately. The storage unit 52 may also store information defining a processing target region and an error determination region for each blood reservoir 10. This configuration allows the error determination unit 531 and the neighboring region extraction unit 532 to make more accurate determinations.

[0060] The storage unit 52 may also store a determination model for determining the type of the blood reservoir 10 from a captured image. A specific example of such a determination model may be information that associates information on a barcode or two-dimensional barcode attached to the blood reservoir 10 (or information obtained by decoding the barcode or two-dimensional barcode) with information indicating the type. In this case, the type may be determined based on the barcode or two-dimensional barcode obtained from the captured image.

[0061] As another specific example of such a determination model, for example, pattern information indicating the outline of the appearance of the blood reservoir 10 may be used for each type of the blood reservoir 10. In this case, the type may be determined by performing pattern matching using the pattern information.

[0062] As another specific example of such a determination model, a trained model obtained by performing a learning process using images of the blood storage tanks 10 of various types, which have been created in advance as training data, and label information indicating the types may be used. In this case, the type may be determined by performing a determination process using the captured image and the trained model.

[0063] Although an embodiment of the present invention has been described in detail above with reference to the drawings, the specific configuration is not limited to this embodiment, and includes designs within the scope of the gist of the present invention. [Explanation of symbols]

[0064] 100...liquid level control system, 10...blood reservoir, 20...holding device, 30...imaging device, 40...pump, 50...determination device, 51...communication unit, 52...storage unit, 53...control unit, 531...error determination unit, 532...neighborhood area extraction unit, 533...liquid level candidate acquisition unit, 534...liquid level determination unit, 535...correction processing unit, 60...pump control device, 71...first area, 72...second area, 73...third area

Claims

1. A determination device comprising: a control unit that acquires an image obtained by capturing an image of a liquid to be determined stored in a specified container, extracts a nearby area in the captured image that is likely to contain an image showing the liquid surface based on features corresponding to the color tone, and determines a straight line showing the liquid surface from the image of the nearby area.

2. The determination device according to claim 1 , wherein the control unit detects straight lines from the image of the vicinity region and determines a straight line indicating a liquid surface from among the straight lines.

3. The determination device according to claim 2 , wherein the control unit determines the straight line indicating the liquid level from among the straight lines having angles within a predetermined range assumed to be angles formed by the liquid level.

4. The determination device according to claim 1 , wherein the feature quantity according to the color tone is expressed using a value related to at least saturation or brightness.

5. The determination device according to claim 1 , wherein the control unit performs a correction process on a determination result of the straight line indicating the liquid level.

6. The determination device according to claim 1 , wherein the control unit further determines the amount of the liquid depending on a determination result of the straight line indicating the liquid level.

7. There are multiple types of the predetermined container, a storage unit that stores a determination model for determining the type of the container from the captured image; The determination device according to claim 1 , wherein the control unit determines the type of the container shown in the captured image based on the captured image and the determination model.

8. The storage unit further stores information indicating a processing area, which is an area on the image to be processed, for each type of container; The determination device according to claim 7 , wherein the control unit performs processing based on a processing region corresponding to the determined type of container.

9. The determination device according to claim 1 , wherein the control unit determines whether an error occurs in the captured image based on a predetermined error criterion.

10. a control unit that acquires an image obtained by capturing an image of a state in which a liquid to be determined is accumulated in a predetermined container, extracts a nearby region that is an area that is likely to include an image showing the liquid surface based on a feature amount corresponding to a color tone in the captured image, and determines a straight line showing the liquid surface from the image of the nearby region; a pump control device that controls the operation of a pump that pumps the liquid in the container to the outside based on the straight line indicating the liquid level obtained by the determination device; A liquid level control system comprising:

11. a display device for displaying the captured image; The liquid level control system according to claim 10 , wherein the display device displays a straight line indicating the liquid level determined by the determination device superimposed on the captured image.

12. The control unit further determines the amount of the liquid according to a determination result of the straight line indicating the liquid level, The liquid level control system according to claim 11 , wherein the display device further displays information indicating the amount of the liquid determined by the determination device.

13. acquiring a captured image obtained by capturing an image of a state in which a liquid to be determined is accumulated in a predetermined container; extracting a nearby region that is likely to include an image showing the liquid surface based on a feature amount corresponding to a color tone in the captured image; determining a straight line indicating the liquid surface from the image of the nearby region; A determination method having the following.

Images