A method and system for monitoring liquid difference compensation of pipette based on computer vision

Abstract

The present invention relates to the technical field of liquid differential compensation, and specifically discloses a pipette liquid differential compensation monitoring method and system based on computer vision, including: based on the initial image of the pipette tip and a preset image contour extraction model, obtaining a processed image of the pipette tip; based on the processed image of the pipette tip, obtaining all reference images and a liquid differential analysis matrix of the pipette tip, and based on all class pipetting parameters of all reference images of the pipette tip, obtaining a pipetting parameter analysis matrix of the pipette tip; based on the liquid differential analysis matrix of the pipette tip and the pipetting parameter analysis matrix, obtaining an adjustment value of the discharge speed of the processed image of the pipette tip, and based on the adjustment value of the discharge speed of the processed image of the pipette tip, obtaining a pipette liquid differential compensation result of the pipette. The present invention realizes liquid differential compensation for high-viscosity liquids and highly contaminated liquids in the discharge link of the pipetting process, and improves the accuracy of pipetting.

Description

Technical Field

[0001] The present invention relates to the technical field of liquid difference compensation, and in particular to a method and system for monitoring liquid difference compensation of a pipette based on computer vision. Background Art

[0002] At present, in the fields of life sciences, chemical analysis, drug development, etc., pipettes are commonly used equipment in laboratories for accurately transferring trace amounts of liquids. However, when dealing with high-viscosity liquids (such as certain polymer solutions, blood samples, etc.) or highly contaminated liquids (such as solutions containing a large number of particles, cell fragments or other impurities), the pipetting process will encounter a series of problems. In particular, in the discharge link of the pipetting process, the first problem encountered is liquid residue. High-viscosity liquids are prone to form residues inside the pipette tip, resulting in the appearance of liquid difference. The presence of residues makes the image of the pipette tip complicated and interferes with the judgment of the visual recognition system. Secondly, color changes and contamination will be encountered. Impurities in highly contaminated liquids may cause changes in the color of the surface of the pipette tip. These changes may mask the actual residual situation of the liquid and interfere with the judgment of the visual recognition system. Therefore, how to achieve liquid difference compensation for high-viscosity liquids and highly contaminated liquids in the discharge link of the pipetting process to improve the accuracy of pipetting has become a key difficulty in the field of pipette liquid difference compensation.

[0003] However, the existing computer vision-based liquid difference compensation monitoring method and system for pipettes only feed back the compensation value result of the liquid taking error through the feature value sequence in the image error recognition database, and then determine the actual amount of liquid that needs to be compensated for the liquid taking component according to the compensation value result, and do not consider how to realize the liquid difference compensation for high-viscosity liquids and highly contaminated liquids in the liquid discharge link of the pipetting process to improve the accuracy of the pipetting. For example, the patent publication number is "CN116091994A" and the patent name is "A monitoring method and device based on pipette visual recognition liquid difference compensation", which includes the following steps: obtaining a real-time image of the liquid taking component to be processed; restoring the real-time image of the liquid taking component to be processed to obtain a processed real-time image of the liquid taking component; inputting the processed real-time image of the liquid taking component into a preset image error recognition database for storage and calculation to determine whether there is a liquid taking error in the image of the liquid taking component; if there is, comparing it with the feature value sequence in the image error recognition database, and feeding back the compensation value result of the liquid taking error; determining the actual amount of liquid that needs to be compensated for the liquid taking component according to the compensation value result; and obtaining the image of the component in real time for liquid level monitoring. The above patent can improve the efficiency of liquid level recognition and compensation feedback of liquid-taking parts such as pipette tips, and monitor the changes in liquid level in real time. However, the patent only uses the characteristic value sequence in the image error recognition database to feedback the compensation value result of the liquid-taking error, and then determines the actual amount of liquid that needs to be compensated by the liquid-taking part based on the compensation value result, and does not consider how to achieve liquid difference compensation for high-viscosity liquids and highly contaminated liquids in the liquid discharge process to improve the accuracy of liquid transfer.

[0004] Therefore, the present invention proposes a method and system for monitoring liquid difference compensation of a pipette based on computer vision. Summary of the invention

[0005] The present invention provides a computer vision-based pipette liquid difference compensation monitoring method and system, which are used to obtain a standard contour area image and a first image pixel threshold of a pipette tip according to an initial image of the pipette tip and a preset image contour extraction model, thereby selecting a partial image that needs to be color corrected from the initial image of the pipette tip, and then obtaining a processed image of the pipette tip according to the standard contour area image of the pipette tip and the first image pixel threshold, thereby realizing color correction of the standard contour area image of the pipette tip, facilitating the subsequent determination of a first pixel area and a second pixel area, obtaining a first pixel area and a second pixel area of ​​the processed image of the pipette tip according to the processed image of the pipette tip, and accurately delineating the areas representing liquid residue and no liquid residue in the processed image of the pipette tip. According to the pixel point area of ​​the object, the liquid difference analysis matrix and the pipetting parameter analysis matrix of the pipette tip are obtained according to the first pixel area and the second pixel area of ​​the processed image of the pipette tip, and then the liquid difference compensation matrix of the pipette tip is constructed, and the construction of a matrix for analyzing the adjustment value of the discharge speed of the processed image of the pipette tip is realized. According to the liquid difference compensation matrix of the pipette tip, the adjustment value of the discharge speed of the processed image of the pipette tip is obtained, and the adjustment amount for adjusting or compensating the discharge speed in the pipetting process corresponding to the processed image of the pipette tip is accurately obtained. Finally, according to the adjustment value of the discharge speed of the processed image of the pipette tip, the pipette liquid difference compensation result of the pipette is obtained, and liquid difference compensation for high-viscosity liquid and high-contamination liquid is realized in the discharge link of the pipette process, thereby improving the accuracy of pipetting.

[0006] The present invention provides a computer vision-based pipette liquid difference compensation monitoring method, comprising:

[0007] S1: performing image processing on the original image of the pipette tip to obtain an initial image of the pipette tip;

[0008] S2: based on the initial image of the pipette tip and the preset image contour extraction model, obtaining a standard contour area image of the pipette tip and a first image pixel threshold, and based on the standard contour area image of the pipette tip and the first image pixel threshold, obtaining a processed image of the pipette tip;

[0009] S3: based on the processed image of the pipette tip, obtaining a first pixel area and a second pixel area of ​​the processed image of the pipette tip, and based on the first pixel area and the second pixel area of ​​the processed image of the pipette tip, obtaining all reference images and a liquid difference analysis matrix of the pipette tip, and based on all class pipetting parameters of all reference images of the pipette tip, obtaining a pipetting parameter analysis matrix of the pipette tip;

[0010] S4: Based on the liquid difference analysis matrix and the pipetting parameter analysis matrix of the pipette tip, the liquid difference compensation matrix of the pipette tip is obtained, and based on the liquid difference compensation matrix of the pipette tip, the adjustment value of the discharge speed of the processed image of the pipette tip is obtained, and based on the adjustment value of the discharge speed of the processed image of the pipette tip, the pipette liquid difference compensation result of the pipette is obtained.

[0011] Preferably, the pipette liquid difference compensation monitoring method based on computer vision, S1: performing image processing on the original image of the pipette tip to obtain the initial image of the pipette tip, including:

[0012] When the pipette completes half of the pipetting task, a real-time image of the pipette tip is acquired, and the acquired real-time image of the pipette tip is used as an original image of the pipette tip;

[0013] If the pixel difference between the pixel value of each pixel point in the original image of the pipette tip and the pixel values ​​of all adjacent pixel points of the pixel point is greater than the preset first pixel difference, the pixel value of the pixel point is adjusted to be the same as the average of the pixel values ​​of all adjacent pixel points of the pixel point to obtain the initial image of the pipette tip.

[0014] Preferably, the computer vision-based pipette liquid difference compensation monitoring method obtains a standard contour area image of the pipette tip and a first image pixel threshold based on an initial image of the pipette tip and a preset image contour extraction model, including:

[0015] Based on the initial image of the pipette tip and the preset image contour extraction model, a contour area image of the pipette tip is obtained, and the pixel position with the maximum pixel value in the contour area image of the pipette tip is used as a fixed pixel point of the contour area image of the pipette tip;

[0016] When the fixed pixel point of the contour area image of the pipette tip is aligned with the preset intersection point of the preset segmentation template, the contour area image of the pipette tip and the preset segmentation template are overlapped and placed to form a rotation angle, which is used as the segmentation angle to obtain all kinds of segmentation angles, and based on the contour area image of the part of the pipette tip contained in each segmentation blank block of the preset segmentation template in the segmentation method corresponding to each segmentation angle, the preparatory segmentation sub-blocks of the contour area image of the pipette tip in the segmentation method corresponding to the corresponding segmentation angle are obtained;

[0017] When there is no blank area in each of the prepared segmented sub-blocks of the contour area image of the pipette tip in the segmentation method corresponding to each segmentation angle, the corresponding prepared segmented sub-block of the contour area image of the pipette tip in the segmentation method corresponding to the corresponding segmentation angle is determined as a complete segmented sub-block of the contour area image of the pipette tip in the segmentation method corresponding to the corresponding segmentation angle;

[0018] Determine the segmentation angle corresponding to the segmentation method with the largest number of complete segmentation sub-blocks of the contour area image of the pipette tip among the segmentation methods corresponding to all the segmentation angles, and take it as the standard segmentation angle;

[0019] When there is a blank area in each prepared segmented sub-block of the image of the contour area of ​​the pipette tip in the segmentation method corresponding to the standard segmentation angle, the blank area of ​​the corresponding prepared segmented sub-block is filled to obtain the filled segmented sub-block of the image of the contour area of ​​the pipette tip in the segmentation method corresponding to the standard segmentation angle;

[0020] The image composed of the padded segmentation sub-blocks and the complete segmentation sub-blocks of the contour area image of the pipette tip in the segmentation method corresponding to the standard segmentation angle is regarded as the standard contour area image of the pipette tip, and the average of the pixel values ​​of all pixel points in the standard contour area image of the pipette tip is regarded as the first image pixel threshold.

[0021] Preferably, the computer vision-based pipette liquid difference compensation monitoring method obtains a processed image of the pipette tip based on a standard contour area image of the pipette tip and a first image pixel threshold, comprising:

[0022] Based on the pixel value of each pixel point in the standard contour area image of the pipette tip and the first image pixel threshold, the judgment value of each pixel point in the standard contour area image of the pipette tip is obtained, that is:

[0023] ;

[0024] in, is the judgment value of the currently calculated pixel point in the standard contour area image of the pipette tip, is the pixel value of the currently calculated pixel point in the standard contour area image of the pipette tip, is the mean of the pixel values ​​of all adjacent pixels of the currently calculated pixel in the standard contour area image of the pipette tip, is the first image pixel threshold, is the maximum pixel value of all pixels in the standard contour area image of the pipette tip, is the minimum pixel value among all pixels in the standard contour area image of the pipette tip, is the standard splitting angle, is the number of pixels that can be accommodated in the segmentation blank block of the preset segmentation template. is the maximum pixel value among all pixels in the initial image of the pipette tip, is the minimum pixel value among all pixels in the initial image of the pipette tip, is the natural logarithm, and the value of the natural constant e is 2.718;

[0025] If the judgment value of each pixel point in the standard contour area image of the pipette tip is greater than the preset judgment value threshold, the pixel value of the pixel point in the standard contour area image of the pipette tip is adjusted to be the same as the first image pixel threshold to obtain a processed image of the pipette tip.

[0026] Preferably, the computer vision-based pipette liquid difference compensation monitoring method obtains a first pixel area and a second pixel area of ​​the processed image of the pipette tip based on the processed image of the pipette tip, including:

[0027] The average of the pixel values ​​of all the pixels in the processed image of the pipette tip is used as the second image pixel threshold;

[0028] Determine whether a pixel difference between a pixel value of each pixel point in the processed image of the pipette tip and a pixel threshold of the second image is greater than a preset second pixel difference; if so, regard the pixel point in the processed image of the pipette tip as a first pixel point in the processed image of the pipette tip; otherwise, regard the pixel point in the processed image of the pipette tip as a second pixel point in the processed image of the pipette tip;

[0029] An image area composed of all first pixel points of the processed image of the pipette tip is regarded as the first pixel area of ​​the processed image of the pipette tip, and an image area composed of all second pixel points of the processed image of the pipette tip is regarded as the second pixel area of ​​the processed image of the pipette tip.

[0030] Preferably, the computer vision-based pipette liquid difference compensation monitoring method obtains all reference images and liquid difference analysis matrices of the pipette tip based on the first pixel area and the second pixel area of ​​the processed image of the pipette tip, including:

[0031] Among all the processed images of historical pipette tips in the database, the processed images of historical pipette tips, for which the difference between the number of pixels in the first pixel area and the number of pixels in the first pixel area of ​​the processed image of the pipette tip is less than a preset first difference, and the difference between the number of pixels in the second pixel area and the number of pixels in the second pixel area of ​​the processed image of the pipette tip is less than a preset second difference, are used as reference images of the pipette tips, to obtain all reference images of the pipette tips;

[0032] All reference images of the pipette tip are ordinal-defined starting from 1 in the order of pixel mean values ​​from large to small, to obtain ordinal definition results of all reference images of the pipette tip;

[0033] The liquid difference of all reference images of the pipette tip is obtained, and based on the liquid difference of all reference images of the pipette tip and the ordinal definition result, the liquid difference analysis matrix of the pipette tip is obtained, which is:

[0034] ;

[0035] in, is the liquid differential analysis matrix of the pipette tip, is the number of pixels in the first pixel area of ​​the reference image with the ordinal number 1 of the pipette tip, is the number of pixels in the second pixel area of ​​the reference image with the ordinal number 1 of the pipette tip, is the value of the liquid difference of the reference image with the ordinal number 1 of the pipette tip, is the number of pixels in the first pixel area of ​​the reference image with the ordinal number 2 of the pipette tip, is the number of pixels in the second pixel area of ​​the reference image with the ordinal number 2 of the pipette tip, is the value of the liquid difference of the reference image with the ordinal number 2 of the pipette tip, The ordinal number of the pipette tip is The number of pixels in the first pixel area of ​​the reference image, The ordinal number of the pipette tip is The number of pixels in the second pixel area of ​​the reference image, The ordinal number of the pipette tip is The value of the liquid difference of the reference image.

[0036] Preferably, the computer vision-based pipette liquid difference compensation monitoring method obtains the pipette parameter analysis matrix of the pipette tip based on all types of pipette parameters of all reference images of the pipette tip, including:

[0037] Acquire all pipetting parameters of all reference images of pipette tips, wherein all pipetting parameters include aspiration speed, discharge speed, and immersion depth of the tip;

[0038] Based on all the pipetting parameters of all the reference images of the pipette tip, the pipetting parameter analysis matrix of the pipette tip is obtained, which is:

[0039] ;

[0040] in, is the pipetting parameter analysis matrix of the pipette tip, is the value of the liquid aspiration speed of the reference image with the ordinal number 1 of the pipette tip, is the value of the liquid aspiration speed of the reference image with the ordinal number 2 of the pipette tip, The ordinal number of the pipette tip is The value of the liquid absorption speed of the reference image, is the value of the liquid discharge speed of the reference image with the ordinal number 1 of the pipette tip, is the value of the liquid discharge speed of the reference image with the ordinal number 2 of the pipette tip, The ordinal number of the pipette tip is The value of the drainage velocity of the reference image, is the value of the immersion depth of the pipette tip in the reference image with the ordinal number 1, is the value of the immersion depth of the pipette tip in the reference image with ordinal number 2, The ordinal number of the pipette tip is The value of the tip immersion depth of the reference image.

[0041] Preferably, the computer vision-based pipette liquid difference compensation monitoring method obtains the liquid difference compensation matrix of the pipette tip based on the liquid difference analysis matrix of the pipette tip and the pipetting parameter analysis matrix, including:

[0042] The product of the liquid difference analysis matrix of the pipette tip and the pipetting parameter analysis matrix is ​​used as the liquid difference compensation matrix of the pipette tip.

[0043] Preferably, the pipette liquid difference compensation monitoring method based on computer vision obtains the adjustment value of the discharge speed of the processed image of the pipette tip based on the liquid difference compensation matrix of the pipette tip, and obtains the pipette liquid difference compensation result of the pipette based on the adjustment value of the discharge speed of the processed image of the pipette tip, including:

[0044] Obtaining the liquid discharge speed of the processed image of the pipette tip, and obtaining the ordinal definition results of all reference images of the pipette tip;

[0045] Among all the reference images of the pipette tip, the reference image of the pipette tip whose ordinal definition result is closest to the spectral radius of the liquid difference compensation matrix of the pipette tip is regarded as the adjustment image of the pipette tip;

[0046] taking the difference between the value of the liquid discharge speed of the adjustment image of the pipette tip and the value of the liquid discharge speed of the processing image of the pipette tip as the adjustment value of the liquid discharge speed of the processing image of the pipette tip;

[0047] Based on the adjustment value of the discharge speed of the processed image of the pipette tip, a pipette liquid difference compensation result of the pipette is obtained.

[0048] The present invention provides a pipette liquid difference compensation monitoring system based on computer vision, which is used to execute any one of the pipette liquid difference compensation monitoring methods based on computer vision in embodiments 1 to 9, comprising:

[0049] A first processing module, used for performing image processing on the original image of the pipette tip to obtain an initial image of the pipette tip;

[0050] A second processing module is used to obtain a standard contour area image of the pipette tip and a first image pixel threshold based on the initial image of the pipette tip and a preset image contour extraction model, and obtain a processed image of the pipette tip based on the standard contour area image of the pipette tip and the first image pixel threshold;

[0051] a matrix construction module, for obtaining a first pixel region and a second pixel region of the processed image of the pipette tip based on the processed image of the pipette tip, obtaining all reference images and a liquid difference analysis matrix of the pipette tip based on the first pixel region and the second pixel region of the processed image of the pipette tip, and obtaining a pipetting parameter analysis matrix of the pipette tip based on all class pipetting parameters of all reference images of the pipette tip;

[0052] The liquid difference compensation module is used to obtain the liquid difference compensation matrix of the pipette tip based on the liquid difference analysis matrix and the pipetting parameter analysis matrix of the pipette tip, and obtain the adjustment value of the discharge speed of the processed image of the pipette tip based on the liquid difference compensation matrix of the pipette tip, and obtain the pipette liquid difference compensation result of the pipette based on the adjustment value of the discharge speed of the processed image of the pipette tip.

[0053] The beneficial effects of the present invention compared with the prior art are as follows: based on the initial image of the pipette tip and the preset image contour extraction model, a standard contour area image of the pipette tip and a first image pixel threshold are obtained, so that a partial image that needs to be color corrected is selected from the initial image of the pipette tip, and then based on the standard contour area image of the pipette tip and the first image pixel threshold, a processed image of the pipette tip is obtained, so that color correction of the standard contour area image of the pipette tip is achieved, which facilitates the subsequent determination of the first pixel area and the second pixel area, and based on the processed image of the pipette tip, the first pixel area and the second pixel area of ​​the processed image of the pipette tip are obtained, and the pixel point areas representing liquid residue and no liquid residue in the processed image of the pipette tip are accurately delineated. domain, according to the first pixel area and the second pixel area of ​​the processed image of the pipette tip, the liquid difference analysis matrix and the pipetting parameter analysis matrix of the pipette tip are obtained, and then the liquid difference compensation matrix of the pipette tip is constructed, and the construction of a matrix for analyzing the adjustment value of the discharge speed of the processed image of the pipette tip is realized, and the adjustment value of the discharge speed of the processed image of the pipette tip is obtained according to the liquid difference compensation matrix of the pipette tip, and the adjustment amount for adjusting or compensating the discharge speed in the pipetting process corresponding to the processed image of the pipette tip is accurately obtained, and finally the pipette liquid difference compensation result of the pipette is obtained according to the adjustment value of the discharge speed of the processed image of the pipette tip, which realizes the liquid difference compensation for high-viscosity liquid and high-contamination liquid in the discharge link of the pipette process, and improves the accuracy of pipetting.

[0054] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or be understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structure specifically pointed out in the written application document.

[0055] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0057] Figure 1 This is a flow chart of a method for monitoring liquid difference compensation of a pipette based on computer vision in an embodiment of the present invention;

[0058] Figure 2 Schematic diagram of a pipette liquid difference compensation monitoring system based on computer vision in an embodiment of the present invention. DETAILED DESCRIPTION

[0059] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention. Embodiment 1:

[0060] The present invention provides a method for monitoring liquid difference compensation of a pipette based on computer vision, referring to Figure 1 ,include:

[0061] S1: performing image processing on the original image of the pipette tip to obtain an initial image of the pipette tip;

[0062] S2: based on the initial image of the pipette tip and the preset image contour extraction model, obtaining a standard contour area image of the pipette tip and a first image pixel threshold, and based on the standard contour area image of the pipette tip and the first image pixel threshold, obtaining a processed image of the pipette tip;

[0063] S3: based on the processed image of the pipette tip, obtaining a first pixel area and a second pixel area of ​​the processed image of the pipette tip, and based on the first pixel area and the second pixel area of ​​the processed image of the pipette tip, obtaining all reference images and a liquid difference analysis matrix of the pipette tip, and based on all class pipetting parameters of all reference images of the pipette tip, obtaining a pipetting parameter analysis matrix of the pipette tip;

[0064] S4: Based on the liquid difference analysis matrix and the pipetting parameter analysis matrix of the pipette tip, the liquid difference compensation matrix of the pipette tip is obtained, and based on the liquid difference compensation matrix of the pipette tip, the adjustment value of the discharge speed of the processed image of the pipette tip is obtained, and based on the adjustment value of the discharge speed of the processed image of the pipette tip, the pipette liquid difference compensation result of the pipette is obtained.

[0065] In this embodiment, the pipette tip is a conical hollow plastic object that fits on the end of the pipette.

[0066] In this embodiment, the original image of the pipette tip is a real-time image of the pipette tip taken when the pipette completes half of the pipetting task volume (for example, when the pipetting task volume is 100µL, completing the pipetting of 50µL is considered to have completed half of the pipetting task volume, and the pipetting task volume in this application document refers to the discharge process).

[0067] In this embodiment, the image processing is an image processing process of obtaining an initial image of the pipette tip based on the original image of the pipette tip.

[0068] In this embodiment, the initial image of the pipette tip is an image obtained by performing image processing on the original image of the pipette tip, and can truly reflect the actual details of the pipette tip.

[0069] In this embodiment, the preset image contour extraction model is composed of a large number of pre-collected initial images of pipette tips as model input, and the manually labeled contour area image of the corresponding pipette tip (the contour area image of the pipette tip is a part selected from the initial image of the pipette tip) as model output. The trained neural network model can be input with the initial image of the pipette tip, and can output the contour area image of the corresponding pipette tip.

[0070] In this embodiment, the standard contour region image of the pipette tip is a partial image selected from the initial image of the pipette tip and requiring color correction.

[0071] In this embodiment, the first image pixel threshold is an image pixel threshold obtained based on the initial image of the pipette tip and a preset image contour extraction model and used to determine the processed image of the pipette tip.

[0072] In this embodiment, the processed image of the pipette tip is a processed image obtained by color correction of the standard contour area image of the pipette tip based on the standard contour area image of the pipette tip and the first image pixel threshold.

[0073] In this embodiment, the first pixel region is a pixel region representing liquid residue in the processed image of the pipette tip or the processed image of the historical pipette tip.

[0074] In this embodiment, the second pixel region is a pixel region representing no liquid residue in the processed image of the pipette tip or the processed image of the historical pipette tip.

[0075] In this embodiment, the liquid difference analysis matrix is ​​a matrix obtained based on the first pixel area and the second pixel area of ​​the processed image of the pipette tip, and can be used to analyze the correspondence between the pixel point range of the liquid residue in the processed image of the pipette tip and the liquid difference amount.

[0076] In this embodiment, all types of pipetting parameters of the reference image of the pipette tip are all types of pipetting parameters in the pipetting process corresponding to the reference image of the pipette tip.

[0077] In this embodiment, the pipetting parameter analysis matrix is ​​a matrix obtained based on all types of pipetting parameters of all reference images of the pipette tip, and can be used to construct a liquid difference compensation matrix for the pipette tip.

[0078] In this embodiment, the liquid difference compensation matrix of the pipette tip is a matrix obtained based on the liquid difference analysis matrix and the pipetting parameter analysis matrix of the pipette tip, and can be used to analyze the adjustment value of the discharge speed of the processed image of the pipette tip.

[0079] In this embodiment, the adjustment value of the discharge speed of the processed image of the pipette tip is used to adjust or compensate the adjustment amount of the discharge speed during the pipetting process corresponding to the processed image of the pipette tip. If the adjustment value is a positive number, it indicates an increase, and if the adjustment value is a negative number, it indicates a decrease.

[0080] In this embodiment, the pipette liquid difference compensation result of the pipette is the pipetting result of the pipette obtained after adjusting the liquid discharge speed during the pipetting process corresponding to the processed image of the pipette tip and continuing to complete the entire pipetting task.

[0081] The beneficial effects of the above technology are as follows: based on the initial image of the pipette tip and the preset image contour extraction model, a standard contour area image and a first image pixel threshold of the pipette tip are obtained, so that a partial image that needs to be color corrected is selected from the initial image of the pipette tip, and then based on the standard contour area image of the pipette tip and the first image pixel threshold, a processed image of the pipette tip is obtained, so that color correction of the standard contour area image of the pipette tip is performed, which is convenient for the subsequent determination of the first pixel area and the second pixel area, and based on the processed image of the pipette tip, the first pixel area and the second pixel area of ​​the processed image of the pipette tip are obtained, and the pixel point areas representing liquid residue and no liquid residue in the processed image of the pipette tip are accurately delineated, and based on The first pixel area and the second pixel area of ​​the processed image of the pipette tip are used to obtain the liquid difference analysis matrix and the pipetting parameter analysis matrix of the pipette tip, and then the liquid difference compensation matrix of the pipette tip is constructed, so as to realize the construction of a matrix which can be used to analyze the adjustment value of the discharge speed of the processed image of the pipette tip. According to the liquid difference compensation matrix of the pipette tip, the adjustment value of the discharge speed of the processed image of the pipette tip is obtained, and the adjustment amount used to adjust or compensate the discharge speed in the pipetting process corresponding to the processed image of the pipette tip is accurately obtained. Finally, according to the adjustment value of the discharge speed of the processed image of the pipette tip, the pipette liquid difference compensation result of the pipette is obtained, so as to realize the liquid difference compensation for high-viscosity liquid and high-contamination liquid in the discharge link of the pipette process, and improve the accuracy of pipetting. Embodiment 2:

[0082] On the basis of Example 1, a pipette liquid difference compensation monitoring method based on computer vision, S1: performing image processing on the original image of the pipette tip to obtain the initial image of the pipette tip, including:

[0083] When the pipette completes half of the pipetting task, a real-time image of the pipette tip is acquired, and the acquired real-time image of the pipette tip is used as an original image of the pipette tip;

[0084] If the pixel difference between the pixel value of each pixel point in the original image of the pipette tip and the pixel values ​​of all adjacent pixel points of the pixel point is greater than the preset first pixel difference, the pixel value of the pixel point is adjusted to be the same as the average of the pixel values ​​of all adjacent pixel points of the pixel point to obtain the initial image of the pipette tip.

[0085] In this embodiment, the preset first pixel difference value is a preset pixel difference value used to obtain an initial image of the pipette tip.

[0086] The beneficial effects of the above technology are: performing image processing on the original image of the pipette tip to obtain the initial image of the pipette tip, filtering out noise points in the original image of the pipette tip, and facilitating the subsequent acquisition of the standard contour area image of the pipette tip. Embodiment 3:

[0087] On the basis of Example 1, a pipette liquid difference compensation monitoring method based on computer vision obtains a standard contour area image of the pipette tip and a first image pixel threshold based on an initial image of the pipette tip and a preset image contour extraction model, including:

[0088] Based on the initial image of the pipette tip and the preset image contour extraction model, a contour area image of the pipette tip is obtained, and the pixel position with the maximum pixel value in the contour area image of the pipette tip is used as a fixed pixel point of the contour area image of the pipette tip;

[0089] When the fixed pixel point of the contour area image of the pipette tip is aligned with the preset intersection point of the preset segmentation template, the contour area image of the pipette tip and the preset segmentation template are overlapped and placed to form a rotation angle, which is used as the segmentation angle to obtain all kinds of segmentation angles, and based on the contour area image of the part of the pipette tip contained in each segmentation blank block of the preset segmentation template in the segmentation method corresponding to each segmentation angle, the preparatory segmentation sub-blocks of the contour area image of the pipette tip in the segmentation method corresponding to the corresponding segmentation angle are obtained;

[0090] When there is no blank area in each of the prepared segmented sub-blocks of the contour area image of the pipette tip in the segmentation method corresponding to each segmentation angle, the corresponding prepared segmented sub-block of the contour area image of the pipette tip in the segmentation method corresponding to the corresponding segmentation angle is determined as a complete segmented sub-block of the contour area image of the pipette tip in the segmentation method corresponding to the corresponding segmentation angle;

[0091] Determine the segmentation angle corresponding to the segmentation method with the largest number of complete segmentation sub-blocks of the contour area image of the pipette tip among the segmentation methods corresponding to all the segmentation angles, and take it as the standard segmentation angle;

[0092] When there is a blank area in each prepared segmented sub-block of the image of the contour area of ​​the pipette tip in the segmentation method corresponding to the standard segmentation angle, the blank area of ​​the corresponding prepared segmented sub-block is filled to obtain the filled segmented sub-block of the image of the contour area of ​​the pipette tip in the segmentation method corresponding to the standard segmentation angle;

[0093] The image composed of the padded segmentation sub-blocks and the complete segmentation sub-blocks of the contour area image of the pipette tip in the segmentation method corresponding to the standard segmentation angle is regarded as the standard contour area image of the pipette tip, and the average of the pixel values ​​of all pixel points in the standard contour area image of the pipette tip is regarded as the first image pixel threshold.

[0094] In this embodiment, the preset segmentation template is a pre-set template for segmenting the contour area image of the pipette tip, and the preset segmentation template is a segmentation template composed of a large number of segmentation blank blocks, each of which is a square area containing 16 pixels.

[0095] In this embodiment, the preset intersection point is a point pre-selected in a preset segmentation template.

[0096] In this embodiment, the prepared segmentation sub-block is an image composed of a contour area image of a part of the pipette tip contained in each segmentation blank block of a preset segmentation template in the segmentation method corresponding to each segmentation angle (a contour area image of a part of the pipette tip must exist in the prepared segmentation sub-block) and a blank area (there may not be a blank area in the prepared segmentation sub-block).

[0097] In this embodiment, padding is to adjust the pixel values ​​of all the pixels in the blank area of ​​the prepared segmented sub-block to be the same as the average value of the pixel values ​​of all the pixels in the non-blank area of ​​the corresponding prepared segmented sub-block.

[0098] The beneficial effects of the above technology are as follows: the present invention provides in detail a specific method for obtaining a standard contour area image of a pipette tip and a first image pixel threshold based on an initial image of the pipette tip and a preset image contour extraction model, and accurately selects a partial image that needs to be color corrected from the initial image of the pipette tip, thereby facilitating the subsequent acquisition of a processed image of the pipette tip. Embodiment 4:

[0099] On the basis of Example 3, a pipette liquid difference compensation monitoring method based on computer vision obtains a processed image of the pipette tip based on a standard contour area image of the pipette tip and a first image pixel threshold, including:

[0100] Based on the pixel value of each pixel point in the standard contour area image of the pipette tip and the first image pixel threshold, the judgment value of each pixel point in the standard contour area image of the pipette tip is obtained, that is:

[0101] ;

[0102] in, is the judgment value of the currently calculated pixel point in the standard contour area image of the pipette tip, is the pixel value of the currently calculated pixel point in the standard contour area image of the pipette tip, is the mean of the pixel values ​​of all adjacent pixels of the currently calculated pixel in the standard contour area image of the pipette tip, is the first image pixel threshold, is the maximum pixel value of all pixels in the standard contour area image of the pipette tip, is the minimum pixel value among all pixels in the standard contour area image of the pipette tip, is the standard splitting angle, is the number of pixels that can be accommodated in the segmentation blank block of the preset segmentation template. is the maximum pixel value among all pixels in the initial image of the pipette tip, is the minimum pixel value among all pixels in the initial image of the pipette tip, is the natural logarithm, and the value of the natural constant e is 2.718;

[0103] If the judgment value of each pixel point in the standard contour area image of the pipette tip is greater than the preset judgment value threshold, the pixel value of the pixel point in the standard contour area image of the pipette tip is adjusted to be the same as the first image pixel threshold to obtain a processed image of the pipette tip.

[0104] In this embodiment, the preset judgment value threshold is a pre-set judgment value threshold used to obtain a processed image of a pipette tip.

[0105] The beneficial effects of the above technology are as follows: This embodiment gives in detail a specific method for obtaining a processed image of a pipette tip based on a standard contour area image of the pipette tip and a first image pixel threshold, accurately performs color correction on the standard contour area image of the pipette tip, and realizes correction of the liquid color caused by highly polluting substances in the liquid, which facilitates subsequent liquid difference compensation.

[0106] Embodiment 5:

[0107] On the basis of Example 1, a pipette liquid difference compensation monitoring method based on computer vision obtains a first pixel area and a second pixel area of ​​the processed image of the pipette tip based on the processed image of the pipette tip, including:

[0108] The average of the pixel values ​​of all pixels in the processed image of the pipette tip is used as the second image pixel threshold;

[0109] Determine whether a pixel difference between a pixel value of each pixel point in the processed image of the pipette tip and a pixel threshold of the second image is greater than a preset second pixel difference; if so, regard the pixel point in the processed image of the pipette tip as a first pixel point in the processed image of the pipette tip; otherwise, regard the pixel point in the processed image of the pipette tip as a second pixel point in the processed image of the pipette tip;

[0110] An image area composed of all first pixel points of the processed image of the pipette tip is regarded as the first pixel area of ​​the processed image of the pipette tip, and an image area composed of all second pixel points of the processed image of the pipette tip is regarded as the second pixel area of ​​the processed image of the pipette tip.

[0111] In this embodiment, the preset second pixel difference value is a pixel difference value that is preset and used to determine whether a pixel point in the processed image of the pipette tip is the first pixel point.

[0112] The beneficial effect of the above technology is: according to the processed image of the pipette tip, the second image pixel threshold and the preset second pixel difference, the first pixel area and the second pixel area of ​​the processed image of the pipette tip are obtained, providing a reliable data basis for subsequent liquid difference analysis and pipetting parameter adjustment.

[0113] Embodiment 6:

[0114] On the basis of Example 1, a pipette liquid difference compensation monitoring method based on computer vision obtains all reference images and liquid difference analysis matrices of the pipette tip based on the first pixel area and the second pixel area of ​​the processed image of the pipette tip, including:

[0115] Among all the processed images of historical pipette tips in the database, the processed images of historical pipette tips, for which the difference between the number of pixels in the first pixel area and the number of pixels in the first pixel area of ​​the processed image of the pipette tip is less than a preset first difference, and the difference between the number of pixels in the second pixel area and the number of pixels in the second pixel area of ​​the processed image of the pipette tip is less than a preset second difference, are used as reference images of the pipette tips, to obtain all reference images of the pipette tips;

[0116] All reference images of the pipette tip are ordinal-defined starting from 1 in the order of pixel mean values ​​from large to small, to obtain ordinal definition results of all reference images of the pipette tip;

[0117] The liquid difference of all reference images of the pipette tip is obtained, and based on the liquid difference of all reference images of the pipette tip and the ordinal definition result, the liquid difference analysis matrix of the pipette tip is obtained, which is:

[0118] ;

[0119] in, is the liquid differential analysis matrix of the pipette tip, is the number of pixels in the first pixel area of ​​the reference image with the ordinal number 1 of the pipette tip, is the number of pixels in the second pixel area of ​​the reference image with the ordinal number 1 of the pipette tip, is the value of the liquid difference of the reference image with the ordinal number 1 of the pipette tip, is the number of pixels in the first pixel area of ​​the reference image with the ordinal number 2 of the pipette tip, is the number of pixels in the second pixel area of ​​the reference image with the ordinal number 2 of the pipette tip, is the value of the liquid difference of the reference image with the ordinal number 2 of the pipette tip, The ordinal number of the pipette tip is The number of pixels in the first pixel area of ​​the reference image, The ordinal number of the pipette tip is The number of pixels in the second pixel area of ​​the reference image, The ordinal number of the pipette tip is The value of the liquid difference of the reference image.

[0120] In this embodiment, the database is a database composed of a large number of pre-collected historical pipette tip processing images (the pipette has completed the entire pipetting task).

[0121] In this embodiment, the preset first difference value is a first difference value preset to determine the reference image of the pipette tip.

[0122] In this embodiment, the preset second difference value is a second difference value preset to determine the reference image of the pipette tip.

[0123] In this embodiment, the liquid difference amount is the difference between the set transfer amount and the actual transfer amount in the transfer process corresponding to the reference image of the pipette tip (when the pipette completes half of the transfer task).

[0124] The beneficial effects of the above technology are: based on the first pixel area and the second pixel area of ​​the processed image of the pipette tip, all reference images of the pipette tip are obtained, and then based on the liquid difference amount of all reference images of the pipette tip, the liquid difference analysis matrix of the pipette tip is obtained, thereby realizing the construction of a matrix that can be used to analyze the correspondence between the pixel point range and the liquid difference amount of the liquid residue in the processed image of the pipette tip.

[0125] Embodiment 7:

[0126] On the basis of Example 6, the pipette liquid difference compensation monitoring method based on computer vision obtains the pipette parameter analysis matrix of the pipette tip based on all the pipette parameters of all the reference images of the pipette tip, including:

[0127] Acquire all pipetting parameters of all reference images of pipette tips, wherein all pipetting parameters include aspiration speed, discharge speed, and immersion depth of the tip;

[0128] Based on all the pipetting parameters of all the reference images of the pipette tip, the pipetting parameter analysis matrix of the pipette tip is obtained, which is:

[0129] ;

[0130] in, is the pipetting parameter analysis matrix of the pipette tip, is the value of the liquid aspiration speed of the reference image with the ordinal number 1 of the pipette tip, is the value of the liquid aspiration speed of the reference image with the ordinal number 2 of the pipette tip, The ordinal number of the pipette tip is The value of the liquid absorption speed of the reference image, is the value of the liquid discharge speed of the reference image with the ordinal number 1 of the pipette tip, is the value of the liquid discharge speed of the reference image with the ordinal number 2 of the pipette tip, The ordinal number of the pipette tip is The value of the drainage velocity of the reference image, is the value of the immersion depth of the pipette tip in the reference image with the ordinal number 1, is the value of the immersion depth of the pipette tip in the reference image with ordinal number 2, The ordinal number of the pipette tip is The value of the tip immersion depth of the reference image.

[0131] In this embodiment, the liquid aspiration speed is the speed of the pipette tip during the liquid aspiration process corresponding to the processed image of the pipette tip, and the unit of the liquid aspiration speed is µL / s.

[0132] In this embodiment, the liquid discharge speed is the speed of the pipette tip during the liquid discharge process corresponding to the processed image of the pipette tip, and the unit of the liquid suction speed is µL / s.

[0133] In this embodiment, the tip immersion depth is the length of the pipette tip immersed in the pipette process corresponding to the processed image of the pipette tip during the pipetting process, and the unit of the tip immersion depth is mm.

[0134] The beneficial effects of the above technology are as follows: This embodiment provides in detail a specific method for obtaining a pipetting parameter analysis matrix of a pipette tip based on all types of pipetting parameters of all reference images of the pipette tip, which facilitates the subsequent construction of a liquid difference compensation matrix of the pipette tip.

[0135] Embodiment 8:

[0136] On the basis of Example 1, a pipette liquid difference compensation monitoring method based on computer vision, based on the liquid difference analysis matrix of the pipette tip and the pipetting parameter analysis matrix, obtains the liquid difference compensation matrix of the pipette tip, including:

[0137] The product (matrix multiplication) of the liquid difference analysis matrix of the pipette tip and the pipetting parameter analysis matrix is ​​regarded as the liquid difference compensation matrix of the pipette tip.

[0138] The beneficial effects of the above technology are as follows: This embodiment provides in detail a specific method for constructing a matrix of adjustment values ​​of the liquid discharge speed that can be analyzed in the processed image of the pipette tip.

[0139] Embodiment 9:

[0140] On the basis of Example 1, a pipette liquid difference compensation monitoring method based on computer vision obtains an adjustment value of a liquid discharge speed of a processed image of a pipette tip based on a liquid difference compensation matrix of a pipette tip, and obtains a pipette liquid difference compensation result of a pipette based on the adjustment value of the liquid discharge speed of a processed image of a pipette tip, including:

[0141] Obtaining the liquid discharge speed of the processed image of the pipette tip, and obtaining the ordinal definition results of all reference images of the pipette tip;

[0142] Among all the reference images of the pipette tip, the reference image of the pipette tip whose ordinal definition result is closest to the spectral radius of the liquid difference compensation matrix of the pipette tip is regarded as the adjustment image of the pipette tip;

[0143] taking the difference between the value of the liquid discharge speed of the adjustment image of the pipette tip and the value of the liquid discharge speed of the processing image of the pipette tip as the adjustment value of the liquid discharge speed of the processing image of the pipette tip;

[0144] Based on the adjustment value of the discharge speed of the processed image of the pipette tip, a pipette liquid difference compensation result of the pipette is obtained.

[0145] In this embodiment, the spectral radius is the maximum value of the modulus of the eigenvalues ​​of the liquid difference compensation matrix of the pipette tip.

[0146] In this embodiment, based on the adjustment value of the liquid discharge speed of the processed image of the pipette tip, the pipette liquid difference compensation result of the pipette is obtained, that is:

[0147] The liquid discharge speed of the processed image of the pipette tip is adjusted, and the liquid discharge speed of the processed image of the pipette tip after the adjustment is the sum of the liquid discharge speed of the processed image of the pipette tip before the adjustment and the adjustment value of the liquid discharge speed of the processed image of the pipette tip.

[0148] The beneficial effects of the above technology are: according to the liquid difference compensation matrix of the pipette tip, the adjustment value of the discharge speed of the processed image of the pipette tip is obtained, and the adjustment amount used to adjust or compensate the discharge speed in the pipetting process corresponding to the processed image of the pipette tip is accurately obtained. Finally, according to the adjustment value of the discharge speed of the processed image of the pipette tip, the pipette liquid difference compensation result of the pipette is obtained, which realizes the liquid difference compensation for high-viscosity liquids and highly contaminated liquids in the discharge link of the pipette process, thereby improving the accuracy of pipetting.

[0149] Embodiment 10:

[0150] The present invention provides a pipette liquid difference compensation monitoring system based on computer vision, which is used to perform any one of the pipette liquid difference compensation monitoring methods based on computer vision in embodiments 1 to 9, with reference to Figure 2 ,include:

[0151] A first processing module, used for performing image processing on the original image of the pipette tip to obtain an initial image of the pipette tip;

[0152] A second processing module is used to obtain a standard contour area image of the pipette tip and a first image pixel threshold based on the initial image of the pipette tip and a preset image contour extraction model, and obtain a processed image of the pipette tip based on the standard contour area image of the pipette tip and the first image pixel threshold;

[0153] a matrix construction module, for obtaining a first pixel region and a second pixel region of the processed image of the pipette tip based on the processed image of the pipette tip, obtaining all reference images and a liquid difference analysis matrix of the pipette tip based on the first pixel region and the second pixel region of the processed image of the pipette tip, and obtaining a pipetting parameter analysis matrix of the pipette tip based on all class pipetting parameters of all reference images of the pipette tip;

[0154] The liquid difference compensation module is used to obtain the liquid difference compensation matrix of the pipette tip based on the liquid difference analysis matrix and the pipetting parameter analysis matrix of the pipette tip, and obtain the adjustment value of the discharge speed of the processed image of the pipette tip based on the liquid difference compensation matrix of the pipette tip, and obtain the pipette liquid difference compensation result of the pipette based on the adjustment value of the discharge speed of the processed image of the pipette tip.

[0155] The beneficial effects of the above technology are as follows: based on the initial image of the pipette tip and the preset image contour extraction model, a standard contour area image and a first image pixel threshold of the pipette tip are obtained, so that a partial image that needs to be color corrected is selected from the initial image of the pipette tip, and then based on the standard contour area image of the pipette tip and the first image pixel threshold, a processed image of the pipette tip is obtained, so that color correction of the standard contour area image of the pipette tip is performed, which is convenient for the subsequent determination of the first pixel area and the second pixel area, and based on the processed image of the pipette tip, the first pixel area and the second pixel area of ​​the processed image of the pipette tip are obtained, and the pixel point areas representing liquid residue and no liquid residue in the processed image of the pipette tip are accurately delineated, and based on The first pixel area and the second pixel area of ​​the processed image of the pipette tip are used to obtain the liquid difference analysis matrix and the pipetting parameter analysis matrix of the pipette tip, and then the liquid difference compensation matrix of the pipette tip is constructed, so as to realize the construction of a matrix which can be used to analyze the adjustment value of the discharge speed of the processed image of the pipette tip. According to the liquid difference compensation matrix of the pipette tip, the adjustment value of the discharge speed of the processed image of the pipette tip is obtained, and the adjustment amount used to adjust or compensate the discharge speed in the pipetting process corresponding to the processed image of the pipette tip is accurately obtained. Finally, according to the adjustment value of the discharge speed of the processed image of the pipette tip, the pipette liquid difference compensation result of the pipette is obtained, so as to realize the liquid difference compensation for high-viscosity liquid and high-contamination liquid in the discharge link of the pipette process, and improve the accuracy of pipetting.

[0156] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention, and the present invention is also intended to include these changes and modifications.

Claims

1. A method for monitoring liquid difference compensation of a pipette based on computer vision, characterized in that: include: S1: performing image processing on the original image of the pipette tip to obtain an initial image of the pipette tip; S2: based on the initial image of the pipette tip and the preset image contour extraction model, obtaining a standard contour area image of the pipette tip and a first image pixel threshold, and based on the standard contour area image of the pipette tip and the first image pixel threshold, obtaining a processed image of the pipette tip; S3: based on the processed image of the pipette tip, obtaining a first pixel area and a second pixel area of ​​the processed image of the pipette tip, and based on the first pixel area and the second pixel area of ​​the processed image of the pipette tip, obtaining all reference images and a liquid difference analysis matrix of the pipette tip, and based on all class pipetting parameters of all reference images of the pipette tip, obtaining a pipetting parameter analysis matrix of the pipette tip; S4: Based on the liquid difference analysis matrix and the pipetting parameter analysis matrix of the pipette tip, the liquid difference compensation matrix of the pipette tip is obtained, and based on the liquid difference compensation matrix of the pipette tip, the adjustment value of the discharge speed of the processed image of the pipette tip is obtained, and based on the adjustment value of the discharge speed of the processed image of the pipette tip, the pipette liquid difference compensation result of the pipette is obtained.

2. The computer vision-based pipette liquid difference compensation monitoring method according to claim 1, characterized in that: S1: performing image processing on the original image of the pipette tip to obtain the initial image of the pipette tip, including: When the pipette completes half of the pipetting task, a real-time image of the pipette tip is acquired, and the acquired real-time image of the pipette tip is used as an original image of the pipette tip; If the pixel difference between the pixel value of each pixel point in the original image of the pipette tip and the pixel values ​​of all adjacent pixel points of the pixel point is greater than the preset first pixel difference, the pixel value of the pixel point is adjusted to be the same as the average of the pixel values ​​of all adjacent pixel points of the pixel point to obtain the initial image of the pipette tip.

3. The computer vision-based pipette liquid difference compensation monitoring method according to claim 1, characterized in that: Based on the initial image of the pipette tip and the preset image contour extraction model, a standard contour area image of the pipette tip and a first image pixel threshold are obtained, including: Based on the initial image of the pipette tip and the preset image contour extraction model, a contour area image of the pipette tip is obtained, and the pixel position with the maximum pixel value in the contour area image of the pipette tip is used as a fixed pixel point of the contour area image of the pipette tip; When the fixed pixel point of the contour area image of the pipette tip is aligned with the preset intersection point of the preset segmentation template, the contour area image of the pipette tip and the preset segmentation template are overlapped and placed to form a rotation angle, which is used as the segmentation angle to obtain all kinds of segmentation angles, and based on the contour area image of the part of the pipette tip contained in each segmentation blank block of the preset segmentation template in the segmentation method corresponding to each segmentation angle, the preparatory segmentation sub-blocks of the contour area image of the pipette tip in the segmentation method corresponding to the corresponding segmentation angle are obtained; When there is no blank area in each of the prepared segmented sub-blocks of the contour area image of the pipette tip in the segmentation method corresponding to each segmentation angle, the corresponding prepared segmented sub-block of the contour area image of the pipette tip in the segmentation method corresponding to the corresponding segmentation angle is determined as a complete segmented sub-block of the contour area image of the pipette tip in the segmentation method corresponding to the corresponding segmentation angle; Determine the segmentation angle corresponding to the segmentation method with the largest number of complete segmentation sub-blocks of the contour area image of the pipette tip among the segmentation methods corresponding to all the segmentation angles, and take it as the standard segmentation angle; When there is a blank area in each prepared segmented sub-block of the image of the contour area of ​​the pipette tip in the segmentation method corresponding to the standard segmentation angle, the blank area of ​​the corresponding prepared segmented sub-block is filled to obtain the filled segmented sub-block of the image of the contour area of ​​the pipette tip in the segmentation method corresponding to the standard segmentation angle; The image composed of the padded segmentation sub-blocks and the complete segmentation sub-blocks of the contour area image of the pipette tip in the segmentation method corresponding to the standard segmentation angle is regarded as the standard contour area image of the pipette tip, and the average of the pixel values ​​of all pixel points in the standard contour area image of the pipette tip is regarded as the first image pixel threshold.

4. The computer vision-based pipette liquid difference compensation monitoring method according to claim 3, characterized in that: Based on the standard contour area image of the pipette tip and the first image pixel threshold, a processed image of the pipette tip is obtained, including: Based on the pixel value of each pixel point in the standard contour area image of the pipette tip and the first image pixel threshold, the judgment value of each pixel point in the standard contour area image of the pipette tip is obtained, that is: ； in, is the judgment value of the currently calculated pixel point in the standard contour area image of the pipette tip, is the pixel value of the currently calculated pixel point in the standard contour area image of the pipette tip, is the mean of the pixel values ​​of all adjacent pixels of the currently calculated pixel in the standard contour area image of the pipette tip, is the first image pixel threshold, is the maximum pixel value of all pixels in the standard contour area image of the pipette tip, is the minimum pixel value among all pixels in the standard contour area image of the pipette tip, is the standard splitting angle, is the number of pixels that can be accommodated in the segmentation blank block of the preset segmentation template. is the maximum pixel value among all pixels in the initial image of the pipette tip, is the minimum pixel value among all pixels in the initial image of the pipette tip, is the natural logarithm, and the value of the natural constant e is 2.718; If the judgment value of each pixel point in the standard contour area image of the pipette tip is greater than the preset judgment value threshold, the pixel value of the pixel point in the standard contour area image of the pipette tip is adjusted to be the same as the first image pixel threshold to obtain a processed image of the pipette tip.

5. The computer vision-based pipette liquid difference compensation monitoring method according to claim 1, characterized in that: Based on the processed image of the pipette tip, obtaining a first pixel region and a second pixel region of the processed image of the pipette tip includes: The average of the pixel values ​​of all pixels in the processed image of the pipette tip is used as the second image pixel threshold; Determine whether a pixel difference between a pixel value of each pixel point in the processed image of the pipette tip and a pixel threshold of the second image is greater than a preset second pixel difference; if so, regard the pixel point in the processed image of the pipette tip as a first pixel point in the processed image of the pipette tip; otherwise, regard the pixel point in the processed image of the pipette tip as a second pixel point in the processed image of the pipette tip; An image area composed of all first pixel points of the processed image of the pipette tip is regarded as the first pixel area of ​​the processed image of the pipette tip, and an image area composed of all second pixel points of the processed image of the pipette tip is regarded as the second pixel area of ​​the processed image of the pipette tip.

6. The computer vision-based pipette liquid difference compensation monitoring method according to claim 1, characterized in that: Based on the first pixel area and the second pixel area of ​​the processed image of the pipette tip, all reference images and a liquid difference analysis matrix of the pipette tip are obtained, including: Among all the processed images of historical pipette tips in the database, the processed images of historical pipette tips, for which the difference between the number of pixels in the first pixel area and the number of pixels in the first pixel area of ​​the processed image of the pipette tip is less than a preset first difference, and the difference between the number of pixels in the second pixel area and the number of pixels in the second pixel area of ​​the processed image of the pipette tip is less than a preset second difference, are used as reference images of the pipette tips, to obtain all reference images of the pipette tips; All reference images of the pipette tip are ordinal-defined starting from 1 in the order of pixel mean values ​​from large to small, to obtain ordinal definition results of all reference images of the pipette tip; The liquid difference of all reference images of the pipette tip is obtained, and based on the liquid difference of all reference images of the pipette tip and the ordinal definition result, the liquid difference analysis matrix of the pipette tip is obtained, which is: ； in, is the liquid differential analysis matrix of the pipette tip, is the number of pixels in the first pixel area of ​​the reference image with the ordinal number 1 of the pipette tip, is the number of pixels in the second pixel area of ​​the reference image with the ordinal number 1 of the pipette tip, is the value of the liquid difference of the reference image with the ordinal number 1 of the pipette tip, is the number of pixels in the first pixel area of ​​the reference image with the ordinal number 2 of the pipette tip, is the number of pixels in the second pixel area of ​​the reference image with the ordinal number 2 of the pipette tip, is the value of the liquid difference of the reference image with the ordinal number 2 of the pipette tip, The ordinal number of the pipette tip is The number of pixels in the first pixel area of ​​the reference image, The ordinal number of the pipette tip is The number of pixels in the second pixel area of ​​the reference image, The ordinal number of the pipette tip is The value of the liquid difference of the reference image.

7. The computer vision-based pipette liquid difference compensation monitoring method according to claim 6, characterized in that: Based on all the pipetting parameters of all the reference images of the pipette tips, the pipetting parameter analysis matrix of the pipette tips is obtained, including: Acquire all pipetting parameters of all reference images of pipette tips, wherein all pipetting parameters include aspiration speed, discharge speed, and immersion depth of the tip; Based on all the pipetting parameters of all the reference images of the pipette tip, the pipetting parameter analysis matrix of the pipette tip is obtained, which is: ； in, is the pipetting parameter analysis matrix of the pipette tip, is the value of the liquid aspiration speed of the reference image with the ordinal number 1 of the pipette tip, is the value of the liquid aspiration speed of the reference image with the ordinal number 2 of the pipette tip, The ordinal number of the pipette tip is The value of the liquid absorption speed of the reference image, is the value of the liquid discharge speed of the reference image with the ordinal number 1 of the pipette tip, is the value of the liquid discharge speed of the reference image with the ordinal number 2 of the pipette tip, The ordinal number of the pipette tip is The value of the drainage velocity of the reference image, is the value of the immersion depth of the pipette tip in the reference image with the ordinal number 1, is the value of the immersion depth of the pipette tip in the reference image with ordinal number 2, The ordinal number of the pipette tip is The value of the tip immersion depth of the reference image.

8. The computer vision-based pipette liquid difference compensation monitoring method according to claim 1, characterized in that: Based on the liquid difference analysis matrix and the pipetting parameter analysis matrix of the pipette tip, the liquid difference compensation matrix of the pipette tip is obtained, including: The product of the liquid difference analysis matrix of the pipette tip and the pipetting parameter analysis matrix is ​​used as the liquid difference compensation matrix of the pipette tip.

9. The computer vision-based pipette liquid difference compensation monitoring method according to claim 1, characterized in that: Based on the liquid difference compensation matrix of the pipette tip, an adjustment value of the liquid discharge speed of the processed image of the pipette tip is obtained, and based on the adjustment value of the liquid discharge speed of the processed image of the pipette tip, a pipette liquid difference compensation result of the pipette is obtained, including: Obtaining the liquid discharge speed of the processed image of the pipette tip, and obtaining the ordinal definition results of all reference images of the pipette tip; Among all the reference images of the pipette tip, the reference image of the pipette tip whose ordinal definition result is closest to the spectral radius of the liquid difference compensation matrix of the pipette tip is regarded as the adjustment image of the pipette tip; taking the difference between the value of the liquid discharge speed of the adjustment image of the pipette tip and the value of the liquid discharge speed of the processing image of the pipette tip as the adjustment value of the liquid discharge speed of the processing image of the pipette tip; Based on the adjustment value of the discharge speed of the processed image of the pipette tip, a pipette liquid difference compensation result of the pipette is obtained.

10. A pipette liquid difference compensation monitoring system based on computer vision, characterized in that: A method for monitoring liquid difference compensation of a pipette based on computer vision for executing any one of claims 1 to 9, comprising: A first processing module, used for performing image processing on the original image of the pipette tip to obtain an initial image of the pipette tip; A second processing module is used to obtain a standard contour area image of the pipette tip and a first image pixel threshold based on the initial image of the pipette tip and a preset image contour extraction model, and obtain a processed image of the pipette tip based on the standard contour area image of the pipette tip and the first image pixel threshold; a matrix construction module, for obtaining a first pixel region and a second pixel region of the processed image of the pipette tip based on the processed image of the pipette tip, obtaining all reference images and a liquid difference analysis matrix of the pipette tip based on the first pixel region and the second pixel region of the processed image of the pipette tip, and obtaining a pipetting parameter analysis matrix of the pipette tip based on all class pipetting parameters of all reference images of the pipette tip; The liquid difference compensation module is used to obtain the liquid difference compensation matrix of the pipette tip based on the liquid difference analysis matrix and the pipetting parameter analysis matrix of the pipette tip, and obtain the adjustment value of the discharge speed of the processed image of the pipette tip based on the liquid difference compensation matrix of the pipette tip, and obtain the pipette liquid difference compensation result of the pipette based on the adjustment value of the discharge speed of the processed image of the pipette tip.

Images