Visual identification method and equipment for columnar container information

Abstract

The invention provides a visual identification method and equipment for columnar container information. The visual identification method for the columnar container information comprises the steps that the surface of a columnar container is sequentially photographed from N different circumferential angles, corresponding columnar container images are identified and acquired, and the columnar container image combination can reflect the full circumferential surface information of the columnar container; cutting off width images of two sides corresponding to preset dedistortion angles of the two sides in each columnar container image, and obtaining N dedistortion columnar container images; splicing the N deformity-removed cylindrical container images and removing the edge overlapping parts according to the edge overlapping parts in the adjacent deformity-removed cylindrical container images to form a cylindrical container two-dimensional expanded view; wherein N is an integer greater than or equal to 3. According to the invention, the two-dimensional expansion view of the cylindrical container comprises the whole circumferential surface information of the cylindrical container, the information is complete, layered information and / or bar code identity information in the two-dimensional expansion view of the cylindrical container can be conveniently identified, and the visual identification method and equipment for the information of the cylindrical container are high in automation degree and high in identification efficiency.

Description

Technical Field

[0001] The present application relates to the field of medical automation technology, and in particular to a method and device for visually recognizing columnar container information. Background Art

[0002] Blood collection tubes are currently widely used in the medical industry. Due to the different densities of blood components, blood is stratified in the blood collection tube. At least one barcode is attached to the surface of the blood collection tube. Usually, an image acquisition device needs to be facing the blood stratification or barcode of the blood collection tube to identify the stratification information and / or barcode identity information of the blood collection tube.

[0003] However, the existing image acquisition device can only scan the blood collection tube from a limited viewing angle, and the acquired blood collection tube image information is incomplete and has little information, which results in the image acquisition device being unable to obtain the complete layering information or barcode identity information of the blood collection tube. Summary of the invention

[0004] The present application provides a method and device for visually recognizing columnar container information, thereby improving the efficiency of columnar container recognition.

[0005] According to a first aspect of the present application, a method for visually identifying columnar container information is provided. The method for visually identifying columnar container information comprises:

[0006] Taking pictures of the surface of the columnar container from N different circumferential angles in sequence, identifying and acquiring corresponding columnar container images, wherein the combination of the columnar container images can reflect the full circumferential surface information of the columnar container;

[0007] Cut off the two side width images corresponding to the preset de-distortion angles on both sides of each of the cylindrical container images to obtain N de-distorted cylindrical container images;

[0008] According to the edge overlapped portions in the adjacent de-distorted cylindrical container images, the N de-distorted cylindrical container images are spliced ​​and the edge overlapped portions are removed to form a two-dimensional unfolded image of the cylindrical container;

[0009] Wherein N is an integer greater than or equal to 3.

[0010] Optionally, the photographing of the surface of the columnar container from N different circumferential angles in sequence includes, under the same conditions:

[0011] The image acquisition device takes a picture of the columnar container every time the image acquisition device and the columnar container rotate 360 / N degrees relative to each other; or,

[0012] N image acquisition devices take pictures of the columnar container from N different angles respectively.

[0013] Optionally, the same conditions include: the image acquisition devices are located at the same horizontal plane, are at the same distance from the columnar container, and have the same shooting parameters.

[0014] Optionally, N image acquisition devices are arranged circumferentially around the columnar container, and every two adjacent image acquisition devices are spaced at the same angle relative to the center of the columnar container; or,

[0015] The image acquisition device takes a picture of the surface of the columnar container while rotating the same angle around the columnar container as the center; or

[0016] When the columnar container is rotated circumferentially around its axis by the same angle, the image acquisition device takes pictures of the surface of the columnar container.

[0017] Optionally, a single-side preset dedistortion angle in the columnar container image is smaller than (90-180 / N)°.

[0018] Optionally, after forming the two-dimensional unfolded view of the columnar container, the method further includes:

[0019] A first liquid level layer in the two-dimensional development image of the columnar container is identified.

[0020] Optionally, after identifying the first liquid level layer in the two-dimensional unfolded view of the columnar container, the method further includes:

[0021] Determining whether the first liquid level stratification meets the verification requirements;

[0022] If the judgment result is that the first liquid level stratification meets the verification requirement, the first liquid level stratification is converted into a height in the space coordinate.

[0023] Optionally, the determining whether the first liquid level stratification meets the verification requirements includes:

[0024] Obtaining the confidence level of the first liquid level stratification;

[0025] Comparing the confidence level with a preset threshold;

[0026] If the confidence level is greater than the preset threshold, it is determined that the first liquid level stratification meets the verification requirement;

[0027] If the confidence level is less than the preset threshold, it is determined that the first liquid level stratification does not meet the verification requirement.

[0028] Optionally, the determining whether the first liquid level stratification meets the verification requirements includes:

[0029] Spectroscopic identification of the second liquid level layer in the columnar container;

[0030] comparing the first liquid level stratification and the second liquid level stratification;

[0031] If the first liquid level stratification falls within the second liquid level stratification interval, it is determined that the first liquid level stratification meets the verification requirements;

[0032] If the first liquid level stratification does not fall within the range of the second liquid level stratification interval, it is determined that the first liquid level stratification does not meet the verification requirements.

[0033] Optionally, if the judgment result is that the first liquid level stratification does not meet the verification requirement, prompting information that the verification requirement is not met;

[0034] User confirmation whether passed;

[0035] If the confirmation is passed, the first liquid level layer is converted into a height in the space coordinate, and the liquid is pipetted according to the height of the first liquid level layer;

[0036] If the confirmation fails, the data is recorded and the pipetting is performed manually.

[0037] Optionally, after converting the first liquid level layer into a height in a space coordinate, the method further includes:

[0038] Pipetting is performed according to the height of the first liquid level layer.

[0039] Optionally, after forming the two-dimensional unfolded view of the columnar container, the method further includes:

[0040] Saving the two-dimensional expansion diagram of the columnar container; and / or,

[0041] The two-dimensional unfolded diagram of the columnar container is three-dimensionally modeled to restore the three-dimensional modeled diagram of the columnar container, and the three-dimensional modeled diagram of the columnar container is saved.

[0042] Optionally, after forming the two-dimensional unfolded view of the columnar container, the method further includes:

[0043] The barcode image in the two-dimensional unfolded image of the columnar container is scanned and identified.

[0044] Optionally, after photographing the surface of the columnar container from N different circumferential angles in sequence, identifying and acquiring corresponding columnar container images, wherein the combination of columnar container images can reflect the full circumferential surface information of the columnar container, the method further includes:

[0045] Scanning and identifying a barcode image in the columnar container image;

[0046] If it cannot be recognized, parameterizing at least two of the barcode images to obtain the longest start parameter including the start end of the barcode and the longest end parameter including the end end of the barcode;

[0047] The longest start parameter and the longest end parameter are concatenated and verified, and the barcode identity information is identified after the verification rules are met.

[0048] Optionally, after the identifying and acquiring the corresponding columnar container image, the process further includes:

[0049] The columnar container image is corrected to be arranged in a vertical direction.

[0050] According to a second aspect of the present application, a visual recognition device for columnar container information is provided. The visual recognition device for columnar container information comprises:

[0051] A columnar container installation position, used to define the position of the columnar container;

[0052] An image acquisition device is arranged on a radial side of the columnar container; the image acquisition device is configured to take a picture of the columnar container under the same conditions when the columnar container is rotated circumferentially at the same angle with the columnar container as the center, or to take a picture of the columnar container under the same conditions when the columnar container is rotated circumferentially at the same angle around its axis.

[0053] a memory for storing a computer program; and

[0054] A processor is used to execute the computer program to implement the above-mentioned visual recognition method of columnar container information.

[0055] According to a third aspect of the present application, a visual recognition device for columnar container information is provided. The visual recognition device for columnar container information comprises:

[0056] A columnar container installation position, used to define the position of the columnar container;

[0057] N image acquisition devices, the N image acquisition devices are arranged circumferentially around the columnar container, and every two adjacent image acquisition devices are spaced at the same angle relative to the columnar container; the N image acquisition devices are located in the same horizontal plane and are spaced at the same distance from the columnar container, and have the same lens parameters; wherein N is an integer greater than or equal to 3;

[0058] a memory for storing a computer program; and

[0059] A processor is used to execute the computer program to implement the above-mentioned visual recognition method of columnar container information.

[0060] The beneficial effects of this application include:

[0061] In the present application, the surface of the columnar container is photographed in sequence from N different circumferential angles to identify and obtain the corresponding columnar container image, and the combination of columnar container images can reflect the full circumferential surface information of the columnar container; the width images on both sides corresponding to the preset de-distortion angles on both sides of each columnar container image are cut off to obtain N de-distorted columnar container images, and the N de-distorted columnar container images are spliced ​​and the edge overlap portions are removed according to the edge overlap portions in adjacent de-distorted columnar container images to form a two-dimensional unfolded view of the columnar container. The two-dimensional unfolded view of the columnar container contains the full circumferential surface information of the columnar container, and the information is complete, which facilitates the identification of the layered information and / or barcode identity information in the two-dimensional unfolded view of the columnar container. The visual recognition method and equipment for columnar container information have a high degree of automation and high recognition efficiency. The two-dimensional unfolded view of the columnar container contains the full circumferential surface information of the columnar container, which facilitates the two-dimensional unfolding of the columnar container. Figure 3 Dimensional modeling can restore the three-dimensional modeling diagram of the columnar container to facilitate archiving and tracing.

[0062] By comparing the first liquid surface stratification with the second liquid surface stratification, it is determined whether the first liquid surface stratification meets the verification requirements, and the accuracy of the first liquid surface stratification can be verified. The accuracy and precision of the stratification information recognition are high.

[0063] It should be understood that the foregoing general description and the following detailed description are exemplary only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 It is a schematic diagram of a method for visually identifying columnar container information provided by an embodiment of the present application.

[0065] Figure 2 It is a schematic diagram of a method for visually identifying columnar container information provided by an embodiment of the present application.

[0066] Figure 3 It is a schematic diagram of the structure of a columnar container provided in one embodiment of the present application.

[0067] Figure 4 It is a two-dimensional expansion diagram of a columnar container provided in one embodiment of the present application.

[0068] Figure 5 It is a structural schematic diagram of a visual recognition device for columnar container information provided by an embodiment of the present application.

[0069] Figure 6 It is a structural schematic diagram of a visual recognition device for columnar container information provided by an embodiment of the present application.

[0070] Figure 7 This is a schematic diagram of the connection between a processor, an image acquisition device and a memory provided in one embodiment of the present application.

[0071] Figure 8 It is a structural schematic diagram of a spectral recognition device for columnar container information provided by an embodiment of the present application.

[0072] Reference numerals:

[0073] 10-columnar container; 11-columnar container mounting position; 12-window; 120-plasma layer; 122-leukocyte layer; 124-erythrocyte layer; 14-label; 20-two-dimensional unfolded view of columnar container; 50-image acquisition device; 60-processor; 70-memory; 80-light source; 90-collection device; 96-grating.

[0074] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application. DETAILED DESCRIPTION

[0075] In order to better understand the technical solution of the present application, the embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0076] According to the first embodiment of the present application, a method for visually identifying columnar container information is provided. Figure 3 The columnar container 10 is a cylindrical container, which can store consumables or samples. The liquid can be composed of components with different densities, so that the liquid surface is layered in the columnar container 10. A label can be attached to the circumferential surface of the columnar container 10, and a barcode can be provided on the label. More preferably, the columnar container 10 can be a blood collection tube, and blood is stored in the blood collection tube. Due to the different densities of blood components, the blood in the blood collection tube is layered, which are respectively a plasma layer 120 and a blood cell layer from top to bottom, and the blood cell layer is respectively a white blood cell layer 122 and a red blood cell layer 124 from top to bottom. At least one layer of label 14 is attached to the surface of the blood collection tube. A barcode can be provided on the label 14. For example, the barcode can be attached to the circumferential surface of the columnar container 10 vertically or obliquely relative to the vertical direction. If the label 14 is not attached to the entire circumferential surface of the columnar container 10, a window 12 without label obstruction will be generated.

[0077] See also Figure 1 , Figure 3 and Figure 4 The visual recognition method of columnar container information includes steps S1-S3.

[0078] Step S1: photograph the surface of the columnar container 10 from N different circumferential angles in sequence, identify and obtain corresponding columnar container images, wherein the combination of columnar container images can reflect the full circumferential surface information of the columnar container 10. Wherein, N is an integer greater than or equal to 3.

[0079] In this step, see Figure 5 or Figure 6 The image acquisition device 50 can rotate circumferentially relative to the columnar container 10 or be arranged around the columnar container 10 at uniform intervals in the circumference. The image acquisition device 50 can take pictures of the columnar container 10 from N different angles.

[0080] The image acquisition device 50 takes a photo of the columnar container 10 once, and the corresponding columnar container image identified and acquired may include a circumferential surface image corresponding to a central angle of 180 degrees of the columnar container.

[0081] By photographing the columnar container 10 from N different angles, N images can be obtained. The corresponding N columnar container images can be identified and obtained from the N images. The combination of the N columnar container images can reflect the full circumference information of the columnar container.

[0082] The image acquisition device 50 has a wide field of view of the image taken by the image acquisition device 50 on the surface of the columnar container 10, including not only the columnar container image but also the background image. The method can identify the corresponding columnar container image, and by cutting out the background image and other non-columnar container images in the image taken by the image acquisition device 50, the corresponding columnar container image can be retained.

[0083] N can be 3, 4, 5, 6 or any other value greater than 3. It is understandable that, according to industrial requirements, the value of N will not be infinite. Preferably, according to industrial requirements, N can be less than or equal to 12.

[0084] There are many specific ways to take pictures of the surface of the columnar container from N different circumferential angles in sequence.

[0085] For an alternative example, see Figure 5 , the step S1 of “taking photos of the surface of the columnar container 10 from N different circumferential angles in sequence” may include the following steps:

[0086] The image acquisition device 50 takes a picture of the columnar container 10 under the same conditions every time the image acquisition device 50 and the columnar container 10 rotate 360 / N degrees relative to each other.

[0087] In this example, see Figure 3 The image acquisition device 50 has a large field of view of the image taken by the image acquisition device 50 on the surface of the columnar container 10, including not only the columnar container image but also the background image. The method can identify the corresponding columnar container image, and by cutting out the background image and other non-columnar container images in the image taken by the image acquisition device 50, the corresponding columnar container image can be retained.

[0088] The image acquisition device 50 takes a photo of the columnar container 10 once, and the corresponding columnar container image identified and acquired may include a circumferential surface image corresponding to a central angle of 180 degrees of the columnar container.

[0089] N may be 3, 4, 5, 6 or any other value greater than 3. Preferably, according to industrial requirements, N may be less than or equal to 12.

[0090] For example Figure 5 As shown, N may be 3, and the image acquisition device 50 takes a picture of the columnar container 10 every time it rotates 120 degrees relative to the columnar container 10 .

[0091] For another example, N may be 4, and the image acquisition device 50 takes a picture of the columnar container 10 every time it rotates 90 degrees relative to the columnar container 10 .

[0092] The same conditions may specifically include: each time the columnar container 10 is photographed, the image acquisition device 50 is at the same horizontal plane, and is at the same distance from the columnar container 10, and the shooting parameters are the same. The shooting parameters may specifically include data that affect the shooting and imaging, such as focal length. In this way, the image acquisition device 50 photographs the surface of the columnar container from N different circumferential angles in sequence, and the parameters such as the scaled size of the columnar container image identified and acquired are exactly the same, which facilitates the subsequent cropping and splicing of the columnar container image.

[0093] There are many specific ways for the image acquisition device 50 to rotate circumferentially relative to the columnar container 10 .

[0094] For example Figure 5 As shown, the columnar container 10 can be configured to be circumferentially rotated around its axis. When the columnar container 10 rotates around its axis at the same angle, the image acquisition device 50 takes a picture of the surface of the columnar container 10 under the same conditions. The same conditions may include: each time the columnar container 10 rotates around its axis at the same angle, the image acquisition device 50 is in the same horizontal plane, the image acquisition device 50 is spaced the same distance from the columnar container 10, and the shooting parameters of the image acquisition device 50 are the same. In this way, each time the columnar container 10 rotates around its axis at the same angle, the image acquisition device 50 takes a picture of the surface of the columnar container 10 under the same conditions, so the parameters such as the zoom size of the corresponding columnar container image identified and acquired are exactly the same, which is convenient for subsequent cropping and splicing of the columnar container image.

[0095] For example Figure 5As shown, the image acquisition device 50 can be configured to rotate with the columnar container 10 as the center. When the image acquisition device 50 rotates at the same angle around the columnar container as the center, the image acquisition device 50 takes a picture of the surface of the columnar container 10 under the same conditions. The same conditions may include: each time the image acquisition device 50 rotates at the same angle around the columnar container as the center, the image acquisition device 50 is in the same horizontal plane, the image acquisition device 50 and the columnar container 10 are spaced at the same distance, and the shooting parameters of the image acquisition device 50 are the same. In this way, each time the image acquisition device 50 rotates at the same angle around the columnar container as the center, the image acquisition device 50 takes a picture of the surface of the columnar container 10 under the same conditions, so the parameters such as the zoom size of the corresponding columnar container image identified and acquired are exactly the same, which is convenient for subsequent cropping and splicing of the columnar container image.

[0096] For an alternative example, see Figure 6 , the step S1 of “taking photos of the surface of the columnar container 10 from N different circumferential angles in sequence” may include the following steps:

[0097] Under the same conditions, N image acquisition devices 50 take pictures of the columnar container 10 from N different angles respectively.

[0098] In this example, N image acquisition devices 50 may be disposed circumferentially around the columnar container, and every two adjacent image acquisition devices 50 may be spaced at the same angle relative to the center of the columnar container 10 .

[0099] The image acquisition device 50 has a wide field of view of the image taken by the image acquisition device 50 on the surface of the columnar container 10, including not only the columnar container image but also the background image. The method can identify the corresponding columnar container image, and by cutting out the background image and other non-columnar container images in the image taken by the image acquisition device 50, the corresponding columnar container image can be retained.

[0100] The image acquisition device 50 takes a photo of the columnar container 10 once, and the corresponding columnar container image identified and acquired may include a circumferential surface image corresponding to a central angle of 180 degrees of the columnar container.

[0101] N may be 3, 4, 5, 6 or any other value greater than 3. Preferably, according to industrial requirements, N may be less than or equal to 12.

[0102] For example Figure 6 As shown, N may be 3, and three image acquisition devices 50 may be disposed circumferentially around the columnar container, and each two adjacent image acquisition devices 50 are spaced 120 degrees apart relative to the center of the columnar container 10 .

[0103] For another example, N may be 4, and four image acquisition devices 50 may be disposed circumferentially around the columnar container, and every two adjacent image acquisition devices 50 are spaced 90 degrees apart relative to the center of the columnar container 10 .

[0104] It can be understood that the angle between two adjacent image acquisition devices 50 and the center of the columnar container 10 is the angle between the lines connecting the two adjacent image acquisition devices 50 and the center of the columnar container 10 .

[0105] The same conditions may specifically include: the N image acquisition devices 50 are at the same level, are at the same distance from the columnar container 10, and have the same shooting parameters, etc. The shooting parameters may specifically include data that affect the shooting and imaging, such as focal length.

[0106] In this arrangement, the N image acquisition devices 50 are only at different circumferential angles relative to the columnar container 10, so that the surface of the columnar container can be photographed in sequence from N different circumferential angles. Since the N image acquisition devices 50 photograph the surface of the columnar container 10 from different circumferential angles under the same conditions, the parameters such as the zoom size of the corresponding columnar container images identified and acquired are completely the same, which facilitates the subsequent cropping and splicing of the N columnar container images.

[0107] Preferably, the step of identifying and acquiring the corresponding columnar container image in step S1 may further include the following step: correcting the columnar container image to be arranged in a vertical direction.

[0108] Specifically, see Figure 3 , the angle at which the columnar container image is tilted relative to the vertical direction Z can be calculated. Specifically, by identifying the length of the long contour line of the columnar container image, the angle between the long contour line length and the vertical direction Z is calculated, and the angle is the angle at which the columnar container image is tilted relative to the vertical direction Z. For example, as shown in the figure, after correction, the axial direction of the columnar container 10 is in the same direction as the vertical direction Z.

[0109] Step S2: cropping the two width images corresponding to the preset de-distortion angles on both sides of each cylindrical container image to obtain N de-distorted cylindrical container images.

[0110] In this step, see Figure 3 Since the columnar container 10 is a cylindrical container, the circumferential surface of the columnar container 10 facing the image acquisition device 50 is an arc surface, which causes the two sides of the corresponding columnar container image identified and acquired by the image acquisition device 50 after taking a photo of the columnar container surface to be deformed and deformed, resulting in large distortion. Therefore, this method needs to cut off the two side width images corresponding to the preset de-distortion angles on both sides of each columnar container image, so as to obtain a de-distorted columnar container image with less distortion in the middle.

[0111] The preset unilateral distortion removal angle in the columnar container image 50 can be specifically set according to the unilateral deformation and distortion of the columnar container image. Figure 3 , the single-side preset dedistortion angle in the columnar container image 50 is less than (90-180 / N)°.

[0112] If the value of N is 3, the single-side preset de-distortion angle in the columnar container image 50 is less than 30. For example, the preset de-distortion angle can be set to 20 degrees, 25 degrees, 30 degrees or other angles. For example, the preset de-distortion angle can be set to 20 degrees, and the two-side width images corresponding to the central angle of 20 degrees on both sides of each columnar container image are cut off, and the de-distorted columnar container image corresponding to the central angle of 140 degrees in the middle of the columnar container image can be retained. For another example, the preset de-distortion angle can be set to 30 degrees, and the two-side width images corresponding to the central angle of 30 degrees on both sides of each columnar container image are cut off, and the de-distorted columnar container image corresponding to the central angle of 120 degrees in the middle of the columnar container image can be retained.

[0113] If the value of N is 4, the preset de-distortion angle of one side in the columnar container image 50 is less than 45 degrees. For example, the preset de-distortion angle can be set to 20 degrees, 25 degrees, 30 degrees, 35 degrees, 40 degrees or other angles.

[0114] Of course, it is understandable that N may also be other values ​​greater than or equal to 3 and less than or equal to 12, and the single-side preset de-distortion angle in the columnar container image 50 is set accordingly.

[0115] Step S3: Please refer to Figure 4 According to the edge overlap in adjacent dedistorted cylindrical container images, N dedistorted cylindrical container images are spliced ​​and the edge overlap is removed to form a two-dimensional unfolded cylindrical container. Figure 4 .

[0116] In this step, there will be edge overlaps between the adjacent parts of two adjacent de-distorted cylindrical container images. According to the edge overlaps in the two adjacent de-distorted cylindrical container images, N de-distorted cylindrical container images can be spliced ​​together. After splicing the two adjacent de-distorted cylindrical container images together, one of the edge overlaps is removed to form a two-dimensional unfolded cylindrical container. Figure 4 In this way, N dedistorted cylindrical container images can be spliced ​​and the overlapping edges can be removed to form a two-dimensional unfolded cylindrical container. Figure 4 .

[0117] As a preference, see Figure 4 , the columnar container formed in step S3 is expanded two-dimensionally Figure 4 , and then the following steps may be included:

[0118] Save the 2D expansion of the columnar container Figure 4 ; and / or

[0119] 2D expansion of cylindrical containers Figure 4 Perform three-dimensional modeling, restore and save the three-dimensional modeling diagram of the columnar container.

[0120] In this step, the columnar container is expanded two-dimensionally Figure 4 Save the file to solve the problem of incomplete single photo window and incomplete information. Figure 4 Roll it up and restore it to a cylindrical shape, and you can unfold the cylindrical container in two dimensions Figure 4 Perform three-dimensional modeling to restore the three-dimensional modeling diagram of the columnar container.

[0121] In this embodiment, please refer to Figure 2 and Figure 4 , step S4 may be further included after step S3.

[0122] Step S4:

[0123] Identify 2D expansion of columnar containers Figure 4 The first liquid level layer in.

[0124] In this step, see Figure 4 Due to the stratification of the liquid, the colors of the liquids in different layers of the columnar container 10 are different. By two-dimensionally unfolding the columnar container Figure 4 By performing basic image processing, such as binarization or sharpening, the boundary line or boundary portion between two adjacent layers of liquid in the columnar container 10 can be identified through the window 12, so that the two-dimensional expansion of the columnar container can be identified. Figure 4 The first liquid level layer in.

[0125] After step S4, step S5 may be further included.

[0126] Step S5:

[0127] Determine whether the first liquid level stratification meets the inspection requirements;

[0128] If the judgment result is that the first liquid level layer meets the verification requirement, the first liquid level layer is converted into a height in the space coordinate. Further, after the first liquid level layer is converted into a height in the space coordinate, liquid transfer can be performed according to the height of the first liquid level layer.

[0129] If the judgment result is that the first liquid surface stratification does not meet the verification requirements, a prompt may be given indicating that the verification requirements are not met; the user confirms whether it passes; if it passes, the first liquid surface stratification is converted into a height in the spatial coordinate, and the liquid is transferred according to the height of the first liquid surface stratification; if it fails, the data is recorded and the liquid is transferred manually.

[0130] In this step, the processor can convert the first liquid level layer calculation into a height in the space coordinate according to the identified first liquid level layer and the preset calibration line, that is, the height of the first liquid level layer can be obtained.

[0131] The specific method steps for determining whether the first liquid level stratification meets the verification requirements can be set as needed.

[0132] In an optional example, determining whether the first liquid level stratification meets the verification requirements may specifically include the following steps:

[0133] Obtaining the confidence level of the first liquid level layer;

[0134] Compare the obtained confidence level with the preset threshold;

[0135] If the confidence level is greater than the preset threshold, it is determined that the first liquid level stratification meets the verification requirements;

[0136] If the confidence level is less than the preset threshold, it is determined that the first liquid level stratification does not meet the verification requirements.

[0137] In this example, the preset threshold value may be pre-stored in the memory. The confidence of the first liquid level stratification may be obtained by calculation or analysis through a program. For example, standard liquid level stratification information may be pre-written in the program, and the confidence of the first liquid level stratification may be obtained by comparing the first liquid level stratification with the standard liquid level stratification information.

[0138] In another optional example, determining whether the first liquid level stratification meets the verification requirements may specifically include the following steps:

[0139] Spectroscopic identification of the second liquid level layer in the columnar container;

[0140] comparing the first liquid surface stratification and the second liquid surface stratification;

[0141] If the first liquid level stratification falls within the second liquid level stratification interval, it is determined that the first liquid level stratification meets the verification requirements;

[0142] If the first liquid level stratification does not fall within the range of the second liquid level stratification interval, it is determined that the first liquid level stratification does not meet the verification requirements.

[0143] In this example, the second liquid level stratification is the liquid level stratification of the columnar container identified by the spectral recognition method. The first liquid level stratification is the liquid level stratification of the columnar container identified by the visual recognition method. It should be noted that the spectral recognition method and the visual recognition method are to identify and obtain an actual liquid level stratification of the columnar container, but due to the differences in accuracy and precision of the spectral recognition method and the visual recognition method, the liquid level stratification results identified by the spectral recognition method and the visual recognition method may be different or the same, so the first liquid level stratification and the second liquid level stratification are only used to describe and distinguish the liquid level stratification results of the columnar container obtained by different recognition methods.

[0144] See also Figure 8 The columnar container spectrum identification method may specifically include steps one to six.

[0145] Step 1: Turn on the light source 80 located on one side of the radial direction R of the columnar container 10, so that a wavelength of transmitted light or at least two different wavelengths of transmitted light pass horizontally through the inspection area of ​​the columnar container.

[0146] In this step, the columnar container 10 can be installed in the columnar container installation position 11. The columnar container installation position 11 is structured to be suitable for moving along the axial direction Z to drive the columnar container 10 installed thereon to move along the axial direction Z. The columnar container 10 has a plurality of test areas in sequence along the axial direction Z thereof.

[0147] A light source 80 is installed on one side of the radial direction R of the columnar container 10, and the light source 80 is located on one side of the radial direction R of the columnar container 10. The light source 80 is mechanically fixed. When the light source 80 is turned on, it can generate a wavelength of transmitted light or at least two different wavelengths of transmitted light, and the one wavelength of transmitted light or at least two different wavelengths of transmitted light generated by the light source 80 can pass horizontally through the columnar container to be inspected area.

[0148] A grating 96 may be provided on the top of the columnar container 10 . The grating 96 structure is configured to shield external light and prevent external light from entering the columnar container 10 .

[0149] It should be noted that, since the liquid surface and the liquid surface layers in the columnar container 10 extend horizontally, the transmitted light is configured to pass through the to-be-tested area horizontally to identify the liquid surface information of the to-be-tested area.

[0150] like Figure 8 As shown, if the columnar container 10 is arranged along the vertical direction Z, the horizontal direction is in the same direction as the radial direction R of the columnar container 10 .

[0151] If the columnar container 10 is arranged tilted relative to the vertical direction Z, the horizontal direction is not in the same direction as the radial direction R of the columnar container 10 .

[0152] Step 2: Collect the transmitted light passing through the to-be-tested area of ​​the columnar container from the other side of the columnar container 10 in the radial direction R.

[0153] In this step, a collection device 90 is installed on the other side of the columnar container 10 in the radial direction R. The collection device 90 is arranged horizontally aligned with the light source 80, and the collection device 90 can collect the transmitted light passing through the area to be inspected of the columnar container.

[0154] Step 3: Obtain the transmittance of the area to be inspected in the columnar container to the transmitted light.

[0155] In this step, when the light source 80 is turned on, the light source 80 generates a wavelength of transmitted light or at least two different wavelengths of transmitted light that pass horizontally through the columnar container inspection area from one side of the radial direction R of the columnar container 10. The transmitted light that passes through the columnar container inspection area can be collected by the collection device 90. By comparing the original light intensity of at least the transmitted light before transmission with the light intensity of the transmitted light collected by the collection device 90, the transmittance of the corresponding columnar container inspection area to the at least two different wavelengths of transmitted light, that is, the loss of light intensity, can be calculated.

[0156] Step 4: Compare the acquired transmittance with the standard transmittance database to identify the liquid level information of the area to be inspected in the columnar container.

[0157] In this step, the standard transmittance database may include the pre-collected transmittance of different layers of liquid in the columnar container 10 to one wavelength of transmitted light or at least two different wavelengths of transmitted light without a label or under the influence of at least one layer of label 14. By comparing the acquired transmittance with the standard transmittance database, the liquid level information of the columnar container to be inspected area can be identified. The liquid level information includes the type of liquid in the columnar container to be inspected area.

[0158] Step 5: Move the columnar container 10 axially so that the light source 80 is aligned with the next area to be inspected of the columnar container.

[0159] In this step, when the columnar container 10 is moved axially, the columnar container 10 and the area to be inspected thereon can move relative to the light source 80 and the collection device 90 along the axial direction Z, so that the light source 80 is aligned with the next area to be inspected of the columnar container; then, the liquid level information of the next area to be inspected of the columnar container can be identified by looping through step 1. Afterwards, when the columnar container 10 is moved axially, the columnar container 10 and the area to be inspected thereon can move relative to the light source 80 and the collection device 90 along the axial direction Z again, so that the light source 80 is aligned with the next area to be inspected of the columnar container.

[0160] Step 6: loop through the above steps 1 to 5 until the liquid level information of all the to-be-inspected areas of the columnar container is identified, and the second liquid level layer of the columnar container 10 is obtained.

[0161] In this step, the liquid level information of the next area to be inspected of the columnar container can be identified by repeating steps 1 to 5. This cycle can be repeated for multiple times to identify the liquid level information of all areas to be inspected of the columnar container and obtain the second liquid level layer of the columnar container 10.

[0162] Specifically, see Figure 8 When the light source 80 is turned on, it can generate a wavelength of transmitted light or at least two different wavelengths of transmitted light. The standard transmittance database may include the pre-collected transmittance of different layers of liquid in the columnar container 10 to a wavelength of transmitted light or at least two different wavelengths of transmitted light without a label or under the influence of at least one layer of label 14.

[0163] Considering that: if the transmitted light source 80 only generates one wavelength of transmitted light from the radial R side of the columnar container 10 horizontally through the columnar container inspection area, the transmittance of a certain layer of liquid in the columnar container 10 at a certain number of layers of labels to the transmitted light may be the same as the transmittance of another layer of liquid at other number of layers of labels to the transmitted light, which will lead to the inability to accurately identify the liquid level information of the columnar container inspection area. Therefore, when the light source 80 is turned on, it is preferred that at least two different wavelengths of transmitted light can be generated sequentially or simultaneously. By obtaining and comparing the transmittances of the columnar container inspection area to at least two different wavelengths of transmitted light, the transmittances of the columnar container inspection area to at least two different wavelengths of transmitted light can be mutually verified, and the confusion caused by the number of layers of labels attached to the columnar container on the inspection can be eliminated.

[0164] Correspondingly, the standard transmittance database may include pre-collected transmittances of different layers of liquid in the columnar container 10 to transmitted light of two different wavelengths without a label or under the influence of at least one layer of label 14 .

[0165] The columnar container 10 may be a blood collection tube, and the blood contained in the blood collection tube is a plasma layer 101 and a blood cell layer from top to bottom. The surface of the blood collection tube may be affixed with three layers of labels 14. The standard transmittance database may include the transmittance of the plasma layer 120 and the blood cell layer collected in advance to two different wavelengths of transmitted light without a label or under the influence of at least one layer of label 14.

[0166] In a specific implementation, please refer to Figure 8, the light source 80 can generate two different wavelengths of transmitted light, and the two different wavelengths of transmitted light can be a first wavelength of transmitted light and a second wavelength of transmitted light, respectively. The columnar container 10 uses a blood collection tube, and the blood contained in the blood collection tube is a plasma layer 101 and a blood cell layer from top to bottom. The surface of the blood collection tube can be affixed with three layers of labels 14. According to the position of the label 14, the surface of the blood collection tube can be unobstructed by the label, or covered with one layer of label 14, or covered with two layers of label 14, or covered with three layers of label 14. The standard transmittance database can correspond to the transmittance of different layers of liquid in the columnar container 10 collected in advance to the two wavelengths of transmitted light without a label or under the influence of at least one layer of label. The standard transmittance database may correspond to: the transmittance of the plasma layer to the first wavelength transmitted light in the absence of label influence, the transmittance of the plasma layer to the first wavelength transmitted light in the absence of label influence, the transmittance of the plasma layer to the first wavelength transmitted light in the absence of label influence, the transmittance of the plasma layer to the first wavelength transmitted light in the absence of label influence, the transmittance of the blood cell layer to the first wavelength transmitted light in the absence of label influence, the transmittance of the blood cell layer to the first wavelength transmitted light in the absence of label influence, the transmittance of the blood cell layer to the first wavelength transmitted light in the absence of label influence, the transmittance of the blood cell layer to the first wavelength transmitted light in the absence of label influence, the transmittance of the blood cell layer to the first wavelength transmitted light in the presence of one label, the transmittance of the blood cell layer to the first wavelength transmitted light in the presence of two labels, and the transmittance of the blood cell layer to the first wavelength transmitted light in the absence of label influence. The transmittance of the first wavelength of transmitted light, the transmittance of the plasma layer to the second wavelength of transmitted light without the influence of labels, the transmittance of the plasma layer to the second wavelength of transmitted light under the influence of one layer of labels, the transmittance of the plasma layer to the second wavelength of transmitted light under the influence of two layers of labels, the transmittance of the plasma layer to the second wavelength of transmitted light under the influence of three layers of labels, the transmittance of the blood cell layer to the second wavelength of transmitted light without the influence of labels, the transmittance of the blood cell layer to the second wavelength of transmitted light under the influence of one layer of labels, the transmittance of the blood cell layer to the second wavelength of transmitted light under the influence of two layers of labels, and the transmittance of the blood cell layer to the second wavelength of transmitted light under the influence of three layers of labels.

[0167] It can be understood that the specific wavelengths of the above two different wavelengths of transmitted light can be set as needed to meet the following requirements: when the transmittance of a certain layer of liquid in the columnar container 10 to one wavelength of transmitted light under the influence of the label 14 (or without a label) is the same as the transmittance of another layer of liquid in the columnar container 10 to one wavelength of transmitted light under the influence of the label 14 (or without a label), the transmittance of a certain layer of liquid in the columnar container 10 to another wavelength of transmitted light under the influence of the label 14 (or without a label) is different from or greatly different from the transmittance of another layer of liquid in the columnar container 10 to another wavelength of transmitted light under the influence of the label 14 (or without a label). In this way, the confusing effect of the number of label layers on the detection can be eliminated, so that the liquid level information of the above two layers of liquid can be accurately identified.

[0168] If the transmittance of a layer of liquid in the columnar container 10 to the first wavelength of transmitted light under the influence of the label 14 (or without a label) is the same as the transmittance of another layer of liquid in the columnar container 10 to the first wavelength of transmitted light under the influence of the label 14 (or without a label), then the transmittance data in the above two situations in the standard transmittance database are also the same. Therefore, the liquid level information of the columnar container to be inspected area may not be identified by only setting the first wavelength of transmitted light; at this time, the transmittance of a layer of liquid in the columnar container 10 to the second wavelength of transmitted light under the influence of the label 14 (or without a label) is different from the transmittance of another layer of liquid in the columnar container 10 to the second wavelength of transmitted light under the influence of the label 14 (or without a label). Therefore, by adding the second wavelength of transmitted light, the liquid level information of any area to be inspected in the above two layers of liquid in the columnar container can be accurately identified. The transmittances of the columnar container to the two different wavelengths of transmitted light can be mutually verified, eliminating the confusion caused by the number of layers of labels attached to the columnar container 10 on the detection, and the recognition and acquisition accuracy is high.

[0169] The axial movement direction of the columnar container 10 can be set as required, which can realize the recognition of the liquid level information of all the to-be-inspected areas of the columnar container so as to obtain the second liquid level stratification of the columnar container 10 .

[0170] For an alternative example, see Figure 8 , before step 1, the following steps may be included:

[0171] Aim the light source 80 at the lowermost area to be tested of the columnar container.

[0172] Step 5 may specifically include the following steps:

[0173] The columnar container 10 is moved axially downward so that the light source 80 is aligned with the next area to be inspected of the columnar container.

[0174] With such arrangement, when the columnar container 10 is moved axially downward, the columnar container 10 and the area to be inspected thereon can move downward along the axial direction Z relative to the light source 80 and the acquisition device 90, so that the light source 80 is aligned with the next area to be inspected of the columnar container; then, by looping through steps 1 to 5, the liquid level information of all the areas to be inspected from bottom to top on the columnar container 10 can be identified in turn, and the second liquid level stratification of the columnar container 10 can be obtained.

[0175] For an alternative example, see Figure 8 , before step 1, the following steps may be included:

[0176] Aim the light source 80 at the uppermost area to be tested of the columnar container.

[0177] Step 5 may specifically include the following steps:

[0178] The columnar container 10 is moved axially upward so that the light source 80 is aligned with the next area to be inspected of the columnar container.

[0179] With such arrangement, when the columnar container 10 is moved axially upward, the columnar container 10 and the area to be inspected thereon can move upward along the axial direction Z relative to the light source 80 and the acquisition device 90, so that the light source 80 is aligned with the next area to be inspected of the columnar container; then, by looping through steps 1 to 5, the liquid level information of all the areas to be inspected from top to bottom on the columnar container 10 can be identified in turn, and the second liquid level stratification of the columnar container 10 can be obtained.

[0180] Since the visual recognition method is greatly affected by external environmental factors (such as light, lines, scratches, etc.), the accuracy of the first liquid surface stratification obtained by the above-mentioned visual recognition method is lower than the accuracy of the second liquid surface stratification obtained by the spectral recognition method. Since the spectral recognition method is limited by the size of the transmitted light spot, the accuracy of the first liquid surface stratification obtained by the visual recognition method is higher than the accuracy of the second liquid surface stratification obtained by the spectral recognition method. Therefore, by comparing the first liquid surface stratification obtained by the visual recognition method with the second liquid surface stratification obtained by the spectral recognition method, it is determined that the first liquid surface stratification falls within the range of the second liquid surface stratification, and the accuracy of the first liquid surface stratification obtained by the visual recognition method can be verified.

[0181] Table 1:

[0182] category Vision spectrum Accuracy 99％ 100％ Accuracy 0.1mm 2mm

[0183] As shown in Table 1, since the visual recognition method is greatly affected by external environmental factors, the accuracy of the first liquid surface layer obtained by the visual recognition method is 99%. The spectral recognition method is less affected by external environmental factors, and the accuracy of the second liquid surface layer obtained by the spectral recognition method is 100%. The accuracy of the first liquid surface layer obtained by the visual recognition method is 0.1mm. Since the spectral recognition method is limited by the size of the transmitted light spot, the accuracy of the second liquid surface layer obtained by the spectral recognition method is 2mm. It can be seen that the accuracy of the first liquid surface stratification obtained by the visual recognition method is lower than the accuracy of the second liquid surface stratification obtained by the spectral recognition method, and the accuracy of the first liquid surface stratification obtained by the visual recognition method is higher than the accuracy of the second liquid surface stratification obtained by the spectral recognition method. By comparing the first liquid surface stratification obtained by the visual recognition method with the second liquid surface stratification obtained by the spectral recognition method, and judging whether the first liquid surface stratification falls within the range of the second liquid surface stratification, the accuracy of the first liquid surface stratification obtained by the visual recognition method can be verified, which makes up for the low accuracy of the first liquid surface stratification obtained by the visual recognition method.

[0184] In this embodiment, the columnar container is formed in step S3. Figure 4 , and then may include step S4'.

[0185] Step S4':

[0186] Scan and identify columnar containers and expand them in two dimensions Figure 4 The barcode image in .

[0187] In this embodiment, please refer to Figure 2 and Figure 3 , after step S1, step S2' may be further included.

[0188] Step S2':

[0189] Scan and identify the barcode image in the columnar container image;

[0190] If it cannot be recognized, at least two barcode images are processed parameter-wise to obtain the longest start parameter including the start end of the barcode and the longest end parameter including the end end of the barcode;

[0191] The longest start parameter and the longest end parameter are concatenated and verified, and the barcode identity information is identified after the verification rules are met.

[0192] In this step, the "scanning and identifying the barcode image in the columnar container image" in step S2' may specifically include: scanning and identifying the barcode image in the columnar container image, and determining whether the barcode identity information can be directly identified.

[0193] The "parametric processing of at least two barcode images to obtain the longest start parameter including the start end of the barcode and the longest end parameter including the end end of the barcode" in step S2' may specifically include the following steps:

[0194] Scanning the barcode image to form a plurality of virtual vertical lines, and converting the barcode images at positions corresponding to the plurality of virtual vertical lines into a plurality of line parameters respectively;

[0195] Checking a plurality of line parameters to identify a plurality of first parameters including a start end of the barcode and a plurality of second parameters including an end end of the barcode, wherein the plurality of first parameters may have different lengths, and the plurality of second parameters may have different lengths;

[0196] Compare multiple first parameters and obtain the longest starting parameter; compare multiple second parameters and obtain the longest starting parameter. In this way, the longest starting parameter and the longest ending parameter can be obtained.

[0197] Specifically, a barcode is a graphic identifier that arranges multiple black bars and blanks of different widths according to certain coding rules to express a set of information. A complete barcode can be composed of multiple black and white lines, and the line parameters can correspond to the black and white color information, line width information, and arrangement order information of all lines in the image 12 at the corresponding position of the vertical line 14. For example, the line parameters can be 212343..., which represents black 2 white 1 black 2 white 3 black 4 white 3...

[0198] There are specific rules for specific barcode formats, which are not consistent between different barcode formats. The barcode format can be: start + data + check + end character. By checking the matching line parameters, multiple first parameters including the start end of the barcode and multiple second parameters belonging to the end part including the end end of the barcode can be identified. Among them, multiple first parameters can have different lengths, and multiple second parameters can have different lengths.

[0199] For example, the starting part of the standard line parameters of a barcode is: black 2 white 1 black 1 white 4 black 1 white 2, and the ending part is: black 2 white 3 black 3 white 1 black 1 white 1 black 2. Compare the line parameters with the starting part and the ending part of the standard line parameters respectively. If the line parameters are black 2 white 1 black 1 white 4 black 1 white 2..., then they cover the starting part of the barcode standard line parameters, and the standard parameters can be considered as the first parameters. If the line parameters are... black 2 white 3 black 3 white 1 black 1 white 1 black 2, then they cover the ending part of the barcode standard line parameters, and the standard parameters can be considered as the second parameters.

[0200] More preferably, the parameterized processing of at least two barcode images in step S2' to obtain the longest start parameter including the start end of the barcode and the longest end parameter including the end end of the barcode may specifically include the following steps:

[0201] Process each barcode image in a parameterized manner to obtain a single longest starting parameter including a barcode start end and / or a single longest ending parameter including a barcode end end in each barcode image;

[0202] The longest start parameters of at least two barcode images are compared to obtain the longest start parameter; the longest end parameters of at least two barcode images are compared to obtain the longest end parameter.

[0203] It is understandable that each barcode image may contain both one end of the barcode and the other end of the barcode, or may not contain both. When a barcode image does not contain the start end of the barcode, the single longest start parameter obtained by parameterizing each barcode image may be zero or not be included in the comparison. When an image does not contain the end end of the barcode, the single longest end parameter obtained by parameterizing each image may be zero or not be included in the comparison.

[0204] In this step, the specific barcode format has specific rules, which are inconsistent between different barcode formats. The barcode format can be: start + data + check + end character. Check the longest start parameter and the longest end parameter after splicing. If the check meets the check rules, the barcode identity information is identified; if the check does not meet the check rules, a check failure message can be sent.

[0205] More preferably, since the longest start parameter and the longest end parameter contain the complete parameters of the barcode after being spliced, and there are repeated parameters in the middle part, the repeated parameters need to be removed after the longest start parameter and the longest end parameter are spliced. Since the spliced ​​deduplication parameters obtained after splicing the longest start parameter and the longest end parameter and removing the repeated parameters may still have abnormal situations such as overlapping parts or partial parameter loss, the spliced ​​deduplication parameters need to be verified.

[0206] Correspondingly, in step S2', the longest start parameter and the longest end parameter are concatenated and verified, and the barcode identity information is identified after the verification rule is met, which may specifically include:

[0207] The longest start parameter and the longest end parameter are concatenated and repeated parameters are removed to form concatenated deduplicated parameters;

[0208] Verify the splicing deduplication parameters and identify the barcode identity information if they meet the verification rules; if the verification does not meet the verification rules, a verification failure message can be sent.

[0209] Second embodiment:

[0210] According to the same inventive concept, a second embodiment provides a visual recognition device for columnar container information. Figure 5 The visual recognition device for columnar container information includes a columnar container mounting position 11, an image acquisition device 50, a memory 70 and a processor 60.

[0211] The columnar container mounting position 11 is used to limit the position of the columnar container 10. The columnar container 10 can be installed in the columnar container mounting position 11. The columnar container 10 is a cylindrical container, which can store consumables or samples. The liquid can be composed of components with different densities, thus forming layers in the columnar container 10. A barcode can be affixed to the circumferential surface of the columnar container 10. For example, the columnar container 10 can be a blood collection tube, which contains blood, and the blood in the blood collection tube is layered due to the different densities of the blood components; at least one barcode is affixed to the surface of the blood collection tube, and the barcode can be affixed to the circumferential surface of the columnar container 10 vertically or obliquely relative to the vertical direction.

[0212] Any two radial sides of the columnar container 10 are respectively one radial side and the other radial side. An image acquisition device 50 is disposed on one radial side of the columnar container 10. The image acquisition device 50 can take a picture of the columnar container 10 from one radial side. For example, the image acquisition device 50 can be a camera, a video camera, a scanner or other image acquisition devices 50.

[0213] The image acquisition device 50 may be configured to be rotatable relative to the columnar container 10. The image acquisition device 50 takes a picture of the columnar container every time it rotates 360 / N degrees relative to the columnar container 10. In this way, by the image acquisition device 50 rotating relative to the columnar container 10, the surface of the columnar container can be photographed from N different circumferential angles in sequence, and the corresponding columnar container images can be identified and acquired, wherein the combination of columnar container images can reflect the full circumferential surface information of the columnar container.

[0214] The image acquisition device 50 takes a photo of the columnar container 10 once, and the corresponding columnar container image identified and acquired may include a circumferential surface image corresponding to a central angle of 180 degrees of the columnar container.

[0215] N can be 3, 4, 5, 6 or any other value greater than 3. It is understandable that, according to industrial requirements, the value of N will not be infinite. Preferably, according to industrial requirements, N can be less than or equal to 12.

[0216] For example, N may be 3, and the image acquisition device 50 may acquire 3 photos in total; each image acquisition device 50 takes a photo of the columnar container 10 each time it rotates 120 degrees relative to the columnar container 10 .

[0217] For another example, N may be 4, and the image acquisition device 50 may acquire 4 photos in total; each image acquisition device 50 rotates 90 degrees relative to the columnar container 10 each time and takes a photo of the columnar container 10 once.

[0218] There are many specific ways for the image acquisition device 50 to rotate relative to the columnar container 10 .

[0219] For an alternative example, see Figure 5, the columnar container mounting position 11 is configured to be rotatable in a circumferential direction. When the columnar container mounting position 11 rotates in a circumferential direction, the columnar container 10 mounted thereon can be driven to rotate in a circumferential direction around its axis. For example, the columnar container mounting position 11 can be connected to a motor or other driving mechanism for being driven to rotate so as to drive the columnar container 10 to rotate in a circumferential direction around its axis. When the columnar container 10 rotates in a circumferential direction around its axis at the same angle, the image acquisition device 50 takes a picture of the surface of the columnar container 10 under the same conditions. Among them, the same conditions may include: each time the columnar container 10 rotates in a circumferential direction around its axis at the same angle, the image acquisition device 50 is in the same horizontal plane, the image acquisition device 50 is spaced the same distance from the columnar container 10, and the shooting parameters of the image acquisition device 50 are the same. With such arrangement, each time the columnar container 10 rotates circumferentially around its axis by the same angle, the image acquisition device 50 takes a picture of the surface of the columnar container 10 under the same conditions, so that the parameters such as the zoom size of the corresponding columnar container image identified and acquired are exactly the same, which facilitates the subsequent cropping and splicing of the columnar container image.

[0220] For an alternative example, see Figure 5 , the image acquisition device 50 may be configured to rotate with the column container 10 as the center. For example, the image acquisition device 50 may be connected to a motor or other driving mechanism, and is used to be driven to rotate with the column container 10 as the center. When the image acquisition device 50 rotates circumferentially at the same angle with the column container as the center, the image acquisition device 50 takes pictures of the surface of the column container 10 under the same conditions. Among them, the same conditions may include: each time the image acquisition device 50 rotates circumferentially at the same angle with the column container as the center, the image acquisition device 50 is in the same horizontal plane, the image acquisition device 50 and the column container 10 are spaced at the same distance, and the shooting parameters of the image acquisition device 50 are the same. In this way, each time the image acquisition device 50 rotates circumferentially at the same angle with the column container as the center, the image acquisition device 50 takes pictures of the surface of the column container 10 under the same conditions, so the parameters such as the zoom size of the corresponding column container image identified and acquired are exactly the same, which is convenient for subsequent cropping and splicing of the column container image.

[0221] See also Figure 6, a computer program is stored in the memory 70. The processor 60 is connected to the image acquisition device 50 and the memory 70 respectively. The processor 60 is used to execute the computer program stored in the memory 70. For example, the processor 60 can be a device master. The computer program is used to control the processor 60 to operate to support the implementation of the visual recognition method of the columnar container information in the first embodiment, and can achieve the same technical effect, which is not described in detail here. The technician can design computer programs and instructions according to the scheme disclosed in the present disclosure. How the computer program controls the processor 60 to operate is well known in the art, so it is not described in detail here. For example, the memory 70 may include ROM (read-only memory), RAM (random access memory), non-volatile memory such as a hard disk, etc. For example, the processor 60 can be a central processing unit CPU, a graphics processing unit GPU, a microprocessor MCU, etc., for executing a computer program, and the computer program can be written in an instruction set of an architecture such as x86, Arm, RISC, MIPS, SSE, etc.

[0222] Third embodiment:

[0223] According to the same inventive concept, a third embodiment provides a visual recognition device for columnar container information. Figure 6 The visual recognition device for columnar container information includes a columnar container installation position 11, N image acquisition devices 50, a memory 70 and a processor 60.

[0224] The columnar container mounting position 11 is used to limit the position of the columnar container 10. The columnar container 10 is a cylindrical container, which can store consumables or samples. The liquid can be composed of components with different densities, thus forming layers in the columnar container 10. A barcode can be attached to the circumferential surface of the columnar container 10. For example, the columnar container 10 can be a blood collection tube, which contains blood, and the blood in the blood collection tube is layered due to the different densities of the blood components; at least one barcode is attached to the surface of the blood collection tube, and the barcode can be attached to the circumferential surface of the columnar container 10 vertically or obliquely relative to the vertical direction.

[0225] The columnar container 10 is installed in the columnar container installation position 11. N image acquisition devices 50 are arranged around the columnar container 10 in a circumferential direction, and every two adjacent image acquisition devices 50 are spaced at the same angle relative to the columnar container 10, where N is an integer greater than or equal to 3. Under the same conditions, the N image acquisition devices 50 can take pictures of the columnar container 10 from N different angles. For example, the image acquisition device 50 can be a camera, a video camera, a scanner, or other image acquisition devices.

[0226] It can be understood that the angle between two adjacent image acquisition devices 50 and the center of the columnar container 10 is the angle between the lines connecting the two adjacent image acquisition devices 50 and the center of the columnar container 10 .

[0227] The image acquisition device 50 takes a photo of the columnar container 10 once, and the corresponding columnar container image identified and acquired may include a circumferential surface image corresponding to a central angle of 180 degrees of the columnar container.

[0228] The above-mentioned same conditions may specifically include: N image acquisition devices 50 are located at the same horizontal plane, and are spaced at the same distance from the columnar container 10, and have the same shooting parameters. In this way, the N image acquisition devices 50 are only at different circumferential angles relative to the columnar container 10, so that the surface of the columnar container can be photographed from N different circumferential angles in sequence. Since the N image acquisition devices 50 photograph the surface of the columnar container 10 from different circumferential angles under the same conditions, the parameters such as the zoom size of the corresponding columnar container images identified and acquired are exactly the same, which facilitates the subsequent cropping and splicing of the N columnar container images.

[0229] N can be 3, 4, 5, 6 or any other value greater than 3. It is understandable that, according to industrial requirements, the value of N will not be infinite. Preferably, according to industrial requirements, N can be less than or equal to 12.

[0230] For example Figure 6 As shown, N may be 3, and three image acquisition devices 50 may be disposed circumferentially around the columnar container, and each two adjacent image acquisition devices 50 are spaced 120 degrees apart relative to the center of the columnar container 10 .

[0231] For another example, N may be 4, and four image acquisition devices 50 may be disposed circumferentially around the columnar container, and every two adjacent image acquisition devices 50 are spaced 90 degrees apart relative to the center of the columnar container 10 .

[0232] See also Figure 7, a computer program is stored in the memory 70. The processor 60 is connected to the image acquisition device 50 and the memory 70 respectively. The processor 60 is used to execute the computer program stored in the memory 70. For example, the processor 60 can be a device master. The computer program is used to control the processor 60 to operate to support the implementation of the visual recognition method of the columnar container information in the first embodiment, and can achieve the same technical effect, which is not repeated here. The technician can design computer programs and instructions according to the scheme disclosed in the present disclosure. How the computer program controls the processor to operate is well known in the art, so it is not described in detail here. For example, the memory 70 may include ROM (read-only memory), RAM (random access memory), non-volatile memory such as a hard disk, etc. For example, the processor 60 can be a central processing unit CPU, a graphics processing unit GPU, a microprocessor MCU, etc., for executing a computer program, and the computer program can be written in an instruction set of an architecture such as x86, Arm, RISC, MIPS, SSE, etc.

[0233] In the present application, the surface of the columnar container is photographed in sequence from N different circumferential angles to identify and obtain the corresponding columnar container image, and the combination of columnar container images can reflect the full circumferential surface information of the columnar container; the width images on both sides corresponding to the preset de-distortion angles on both sides of each columnar container image are cut off to obtain N de-distorted columnar container images, and the N de-distorted columnar container images are spliced ​​and the edge overlap portions are removed according to the edge overlap portions in adjacent de-distorted columnar container images to form a two-dimensional unfolded view of the columnar container. The two-dimensional unfolded view of the columnar container contains the full circumferential surface information of the columnar container, and the information is complete, which facilitates the identification of the layered information and / or barcode identity information in the two-dimensional unfolded view of the columnar container. The visual recognition method and equipment for columnar container information have a high degree of automation and high recognition efficiency. The two-dimensional unfolded view of the columnar container contains the full circumferential surface information of the columnar container, which facilitates the two-dimensional unfolding of the columnar container. Figure 3 Dimensional modeling can restore the three-dimensional modeling diagram of the columnar container to facilitate archiving and tracing.

[0234] By comparing the first liquid surface stratification with the second liquid surface stratification, it is determined whether the first liquid surface stratification meets the verification requirements, and the accuracy of the first liquid surface stratification can be verified. The accuracy and precision of the stratification information recognition are high.

[0235] It should be clear that the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0236] The terms used in the embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the present application. The singular forms "a", "said" and "the" used in the embodiments of the present application and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings.

[0237] It should be understood that the term "and / or" used in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0238] It should be understood that in the description of this application, unless otherwise clearly specified and limited, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance, and are not used to describe a specific order or sequence.

[0239] The word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.

[0240] It should be noted that, in this article, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the statement "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be noted that the scope of the method and device in the embodiment of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0241] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a computer software product, which is stored in a storage medium (such as ROM / RAM, disk, CD), including a number of instructions for a terminal (which can be a mobile phone, computer, control device, or network equipment, etc.) to execute the methods described in each embodiment of the present application.

[0242] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

Claims

1. A method for visually identifying columnar container information, characterized in that: include: Taking pictures of the surface of the columnar container from N different circumferential angles in sequence, identifying and acquiring corresponding columnar container images, wherein the combination of the columnar container images can reflect the full circumferential surface information of the columnar container; Cut off the two side width images corresponding to the preset de-distortion angles on both sides of each of the cylindrical container images to obtain N de-distorted cylindrical container images; According to the edge overlapped portions in the adjacent de-distorted cylindrical container images, the N de-distorted cylindrical container images are spliced ​​and the edge overlapped portions are removed to form a two-dimensional unfolded image of the cylindrical container; Wherein N is an integer greater than or equal to 3.

2. The visual recognition method of columnar container information according to claim 1, characterized in that: The photographing of the surface of the columnar container from N different circumferential angles in sequence includes: The image acquisition device takes a picture of the columnar container every time the image acquisition device and the columnar container rotate 360 / N degrees relative to each other; or, N image acquisition devices take pictures of the columnar container from N different angles respectively.

3. The visual recognition method of columnar container information according to claim 2, characterized in that: The same conditions include: the image acquisition device is on the same horizontal plane, is at the same distance from the columnar container, and has the same shooting parameters.

4. The visual recognition method of columnar container information according to claim 2, characterized in that: N image acquisition devices are arranged circumferentially around the cylindrical container, and every two adjacent image acquisition devices are spaced at the same angle relative to the center of the cylindrical container; or, The image acquisition device takes a picture of the surface of the columnar container while rotating the same angle around the columnar container as the center; or When the columnar container is rotated circumferentially around its axis by the same angle, the image acquisition device takes pictures of the surface of the columnar container.

5. The visual recognition method of columnar container information according to claim 1, characterized in that: The single-side preset dedistortion angle in the columnar container image is less than (90-180 / N)°.

6. The visual recognition method of columnar container information according to claim 1, characterized in that: After forming the two-dimensional unfolded view of the columnar container, the method further includes: A first liquid level layer in the two-dimensional development image of the columnar container is identified.

7. The visual recognition method of columnar container information according to claim 6, characterized in that: After identifying the first liquid level layer in the two-dimensional development diagram of the columnar container, the method further includes: Determining whether the first liquid level stratification meets the verification requirements; If the judgment result is that the first liquid level stratification meets the verification requirement, the first liquid level stratification is converted into a height in the space coordinate.

8. The visual recognition method of columnar container information according to claim 6, characterized in that: The determining whether the first liquid level stratification meets the verification requirements includes: Obtaining the confidence level of the first liquid level stratification; Comparing the confidence level with a preset threshold; If the confidence level is greater than the preset threshold, it is determined that the first liquid level stratification meets the verification requirement; If the confidence level is less than the preset threshold, it is determined that the first liquid level stratification does not meet the verification requirement.

9. The visual recognition method of columnar container information according to claim 7, characterized in that: The determining whether the first liquid level stratification meets the verification requirements includes: Spectroscopic identification of the second liquid level layer in the columnar container; comparing the first liquid level stratification and the second liquid level stratification; If the first liquid level stratification falls within the second liquid level stratification interval, it is determined that the first liquid level stratification meets the verification requirements; If the first liquid level stratification does not fall within the range of the second liquid level stratification interval, it is determined that the first liquid level stratification does not meet the verification requirements.

10. The visual recognition method for columnar container information according to any one of claims 7 to 9, characterized in that: If the judgment result is that the first liquid level stratification does not meet the verification requirement, prompting information that does not meet the verification requirement; Whether the user confirms the result; If the confirmation is passed, the first liquid level layer is converted into a height in the space coordinate, and the liquid is pipetted according to the height of the first liquid level layer; If the confirmation fails, the data is recorded and the pipetting is performed manually.

11. The visual recognition method of columnar container information according to claim 7, characterized in that: After converting the first liquid level layer into the height in the space coordinate, the method further includes: Pipetting is performed according to the height of the first liquid level layer.

12. The visual recognition method of columnar container information according to claim 1, characterized in that: After forming the two-dimensional unfolded view of the columnar container, the method further includes: Saving the two-dimensional expansion diagram of the columnar container; and / or, The two-dimensional unfolded diagram of the columnar container is three-dimensionally modeled to restore the three-dimensional modeled diagram of the columnar container, and the three-dimensional modeled diagram of the columnar container is saved.

13. The visual recognition method of columnar container information according to claim 1, characterized in that: After forming the two-dimensional unfolded view of the columnar container, the method further includes: The barcode image in the two-dimensional unfolded image of the columnar container is scanned and identified.

14. The visual recognition method of columnar container information according to claim 1, characterized in that: After photographing the surface of the columnar container from N different circumferential angles in sequence, identifying and acquiring corresponding columnar container images, wherein the combination of columnar container images can reflect the full circumferential surface information of the columnar container, the following further includes: Scanning and identifying a barcode image in the columnar container image; If it cannot be recognized, parameterizing at least two of the barcode images to obtain the longest start parameter including the start end of the barcode and the longest end parameter including the end end of the barcode; The longest start parameter and the longest end parameter are concatenated and verified, and the barcode identity information is identified after the verification rules are met.

15. The visual recognition method of columnar container information according to claim 1, characterized in that: After the identification and acquisition of the corresponding columnar container image, the method further includes: The columnar container image is corrected to be arranged in a vertical direction.

16. A visual recognition device for columnar container information, characterized in that: include: A columnar container installation position, used to define the position of the columnar container; An image acquisition device is arranged on a radial side of the columnar container; the image acquisition device is configured to take a picture of the columnar container under the same conditions when the columnar container is rotated circumferentially at the same angle with the columnar container as the center, or to take a picture of the columnar container under the same conditions when the columnar container is rotated circumferentially at the same angle around its axis. A memory, the memory being used to store a computer program; and A processor, wherein the processor is configured to execute the computer program to implement the method for visually recognizing columnar container information according to any one of claims 1 to 15.

17. A visual recognition device for columnar container information, characterized in that: include: A columnar container installation position, used to define the position of the columnar container; N image acquisition devices, the N image acquisition devices are arranged circumferentially around the columnar container, and every two adjacent image acquisition devices are spaced at the same angle relative to the columnar container; the N image acquisition devices are located in the same horizontal plane and are spaced at the same distance from the columnar container, and have the same lens parameters; wherein N is an integer greater than or equal to 3; a memory for storing a computer program; and A processor, wherein the processor is configured to execute the computer program to implement the method for visually recognizing columnar container information according to any one of claims 1 to 15.

Images