Deviation correction control method of biological storage library and biological storage library

Abstract

The invention relates to the technical field of biomedical treatment, in particular to a correction control method of a biological storage library and the biological storage library, and aims to solve the problem of inaccurate straw grabbing precision in an existing storage library. In order to achieve the purpose, the deviation correction control method comprises the steps that under the condition that the straw disc rotates, a first sample image where the straw disc does not rotate and a second sample image where the straw disc rotates are obtained; acquiring first feature information according to the first sample image; obtaining second feature information according to the second sample image; determining a plurality of offset positions according to the first feature information and the second feature information; and the mechanical arm is controlled to move according to the multiple offset positions so as to grab the straw. The positions of the straw disc before rotation and after rotation are obtained respectively, then the deviation position is determined according to the position before rotation and the position after rotation, the mechanical arm grabs the straw according to the deviation position, and accurate grabbing of the mechanical arm is guaranteed.

Description

Technical Field

[0001] The present invention relates to the field of biomedical technology, and specifically provides a deviation correction control method for a biological storage repository and a biological storage repository. Background Art

[0002] Frozen straws are storage tubes for biological cells such as sperm, eggs and embryos. They are stored in a deep-cryogenic biological sample storage library. The low-temperature storage of liquid nitrogen can keep biological cells such as sperm, eggs and embryos active for a long time. The existing grasping of straws is based on fixed coordinates, and the robotic arm is controlled to pick up and place the straws in the library according to the three-dimensional coordinates initially measured. However, for straws stored on disks, since the disks move in irregular arcs in the storage library, if the robotic arm still uses the original method to locate the position, when the disk is offset, the robotic arm will not be able to grasp accurately.

[0003] Therefore, the present invention needs to provide a new technical solution to solve the above technical problems. Summary of the invention

[0004] The present invention aims to solve the above technical problem, that is, to solve the problem of inaccurate grasping accuracy of straws in a storage bin.

[0005] In a first aspect, the present invention provides a deviation correction control method for a biological storage library, the biological storage library comprising a tank body, a straw tray and a robotic arm, the straw tray being rotatably arranged in the tank body, a plurality of straw frames and a plurality of positioning frames being arranged on the straw tray, the plurality of positioning frames being arranged on both sides of the straw tray, straws being stored in the plurality of straw frames, and the robotic arm being arranged to be able to grasp the straws; the deviation correction control method comprises: when the straw tray rotates, obtaining a first sample image of the straw tray that has not rotated and a second sample image of the straw tray that has rotated; obtaining first feature information according to the first sample image; wherein the first feature information comprises a plurality of first initial calibration positions of the plurality of positioning frames; obtaining second feature information according to the second sample image; wherein the second feature information comprises a plurality of first target calibration positions of the plurality of positioning frames; determining a plurality of offset positions according to the first feature information and the second feature information; and controlling the robotic arm to move according to the plurality of offset positions to grasp the straws.

[0006] In a specific implementation of the correction control method for the above-mentioned biological storage library, the first initial calibration position and the first target calibration position refer to the pixel coordinates of the center point of the positioning frame; the steps of "obtaining first feature information based on the first sample image" and "obtaining second feature information based on the second sample image" specifically include: first inputting the sample image into a trained image model for comparison to obtain a preprocessed image with rough position information of the positioning frame; and using an image recognition algorithm to obtain the pixel coordinates of the center point of the positioning frame for the preprocessed image.

[0007] In a specific implementation of the above-mentioned correction control method for the biological storage library, the step of "using an image recognition algorithm to obtain the pixel coordinates of the center point of the positioning frame on the preprocessed image" specifically includes: first performing edge detection on the preprocessed image to identify the edges of the positioning frame; extracting the edges of the positioning frame to obtain a contour image of the positioning frame; and obtaining the pixel coordinates of the center point of the positioning frame based on the contour image of the positioning frame.

[0008] In a specific implementation of the correction control method for the above-mentioned biological storage library, the midpoint of the connecting line of the center points of adjacent positioning frames is the calibration point, and the first characteristic information also includes multiple second initial calibration positions of multiple calibration points; the step of "obtaining the first characteristic information according to the first sample image" further includes: determining the pixel coordinates corresponding to the calibration point according to the first initial calibration position of the adjacent positioning frame as the second initial calibration position; wherein, "the adjacent positioning frame" refers to the adjacent positioning frame along the rotation direction of the straw tray.

[0009] In a specific implementation of the correction control method for the above-mentioned biological storage library, the second characteristic information also includes multiple second target calibration positions of the multiple calibration points; the step of "obtaining second characteristic information based on the second sample image" further includes: determining the pixel coordinates corresponding to the calibration point based on the second initial calibration position adjacent to the positioning frame as the second target calibration position.

[0010] In a specific implementation of the correction control method for the above-mentioned biological storage library, the multiple offset positions include multiple first offset positions, and the step of "determining multiple offset positions based on the first characteristic information and the second characteristic information" specifically includes: calculating the difference between the pixel coordinates of the corresponding center point of the positioning frame based on the first initial calibration position and the first target calibration position as the first offset position.

[0011] In a specific implementation of the correction control method for the above-mentioned biological storage library, the multiple offset positions also include multiple second offset positions, and the step of "determining multiple offset positions based on the first feature information and the second feature information" specifically includes: calculating the difference between the corresponding pixel coordinates based on the second initial calibration position and the second target calibration position as the second offset position.

[0012] In a specific implementation of the correction control method for the above-mentioned biological storage library, the correction control method also includes: before the straw tray rotates, obtaining the straw to be grasped and the corresponding initial position; controlling the robotic arm to move to the initial position; the step of "controlling the robotic arm to move according to the multiple offset positions to grasp the straw" specifically includes: matching the corresponding offset position according to the straw to be grasped; controlling the robotic arm to move according to the corresponding offset position to grasp the straw.

[0013] In a specific implementation of the correction control method for the above-mentioned biological storage library, the correction control method also includes: converting the pixel coordinates of the corresponding offset position into corresponding world coordinates and sending them to the robotic arm; the step of "controlling the robotic arm to move according to the corresponding offset position to grab the wheat straw" specifically includes: controlling the robotic arm to move according to the world coordinates of the corresponding offset position to grab the wheat straw.

[0014] In a second aspect, the present invention also provides a biological storage library, comprising a controller, wherein the controller is configured to execute the deviation correction control method as described above.

[0015] In the case of adopting the above technical scheme, the deviation correction control method of the present invention obtains a first sample image of the straw disk that has not rotated and a second sample image of the straw disk that has rotated when the straw disk rotates; obtains first feature information based on the first sample image; wherein the first feature information includes multiple first initial calibration positions of the multiple positioning frames; obtains second feature information based on the second sample image; wherein the second feature information includes multiple first target calibration positions of the multiple positioning frames; determines multiple offset positions based on the first feature information and the second feature information; and controls the robot arm to move according to the multiple offset positions to grab the straws. The present invention obtains the positions of the straw disk before and after rotation respectively, and then determines the offset position based on the position before and after rotation, so that the robot arm grabs the straws according to the offset position, thereby ensuring the accurate grabbing of the robot arm.

[0016] Furthermore, by first using a deep learning algorithm to coarsely locate the sample image, and then using an image recognition algorithm to finely locate it, in order to obtain the position of each positioning frame, by combining the deep learning algorithm with the image recognition algorithm, compared with the existing method, the position of the positioning frame can be accurately located, the overall positioning efficiency can be improved, and the recognition time and recognition cost can be shortened. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings, in which:

[0018] Figure 1 It is a schematic diagram of the installation structure of the straw tray and the rotating disk of the present invention;

[0019] Figure 2 It is a structural schematic diagram of the straw tray of the present invention;

[0020] Figure 3 It is a schematic diagram of the calibration points between the positioning frames No. 1 and No. 8 in the straw tray of the present invention;

[0021] Figure 4 It is a flow chart of the main steps of the deviation correction control method of the biological storage bank of the present invention;

[0022] Figure 5 is a detailed step flow chart of step S2 and step S3 of the present invention;

[0023] Figure 6 is a detailed step flow chart of step S22 of the present invention;

[0024] Figure 7 is a detailed step flow chart of step S5 of the present invention;

[0025] Among them, 1. straw tray; 11. straw frame; 12. positioning frame; 2. rotating disk. DETAILED DESCRIPTION

[0026] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not used to limit the scope of protection of the present invention. Those skilled in the art can make adjustments to them as needed to adapt to specific application scenarios.

[0027] It should be noted that in the description of the present invention, the terms "inside", "top" and other terms indicating directions or positional relationships are based on the directions or positional relationships shown in the drawings, which are only for the convenience of description, and do not indicate or imply that the relevant devices or elements must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, ordinal numbers such as "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0028] In addition, it should be noted that in the description of the present invention, unless otherwise clearly specified and limited, the terms "connected" and "connection" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0029] In addition, it should be noted that in the description of the present invention, although the various steps of the control method of the present invention are described in this application in a specific order, these orders are not restrictive. Without departing from the basic principles of the present invention, those skilled in the art can perform the steps in a different order.

[0030] See first Figure 1 and Figure 2 The biological storage repository includes a tank body (not shown in the figure), a straw tray and a robotic arm (not shown in the figure). The straw tray is rotatably arranged in the tank body. A plurality of straw frames and a plurality of positioning frames are arranged on the straw tray. The plurality of positioning frames are arranged on both sides of the straw tray. Wheat straws (not shown in the figure) are stored in the plurality of straw frames. An opening (not shown in the figure) is arranged on the top of the tank body. The robotic arm is fixed on a cover plate (not shown in the figure) on the top of the tank body. The robotic arm is arranged to be able to extend into the opening to grab the straws. A rotating disk is arranged in the tank body. A plurality of straw trays are evenly arranged at the circumferential position of the rotating disk. The rotating disk rotates to drive the straw tray to rotate so that the straw tray rotates to the bottom of the opening so that the robotic arm can grab the straws. The positioning frames are symmetrically arranged on the straw tray. In actual application, the number of positioning frames is related to the space reserved around the straw tray. The more positioning frames are arranged, the higher the accuracy of obtaining the offset position of the straw tray. For example, Figure 2 As shown, there are eight positioning frames, which are distributed on both sides and the bottom of the straw tray and are symmetrically arranged as a whole. In addition, the positioning frame is arranged in a rectangular shape. Of course, those skilled in the art can also arrange it in other shapes besides a rectangular shape as needed, and the present invention does not impose any limitation on this.

[0031] Then see Figure 4 , which is a flow chart of the main steps of the deviation correction control method of the biological storage bank of the present invention. The deviation correction control method of the present invention comprises:

[0032] S1, when the straw tray rotates, obtaining a first sample image in which the straw tray does not rotate and a second sample image in which the straw tray rotates;

[0033] S2, acquiring first feature information according to the first sample image; wherein the first feature information includes a plurality of first initial calibration positions of the plurality of positioning frames;

[0034] S3, acquiring second feature information according to the second sample image; wherein the second feature information includes a plurality of first target calibration positions of a plurality of positioning frames;

[0035] S4, determining a plurality of offset positions according to the first feature information and the second feature information;

[0036] S5, controlling the robot arm to move according to a plurality of offset positions to grab the straw.

[0037] In step S1, in some embodiments, sample images of the straw tray that is not rotated and rotated are obtained by a camera. The camera is set on the top of the tank body, which is located above the opening so as to be able to shoot the straw tray, and the camera is a 2D camera. In addition, since the straw tray is stored in a deep low temperature environment, this will cause abnormal conditions such as frost on its surface. In order to ensure that the positioning frame of the camera can be recognized when shooting and avoid blurring due to frost and affecting subsequent operations, it is preferred to set the positioning frame to a brighter color, such as red, etc.

[0038] like Figure 5 As shown, in step S2 and step S3, the first initial calibration position and the first target calibration position refer to the pixel coordinates (X, Y) of the center point of the positioning frame; wherein the steps of "obtaining first feature information according to the first sample image" and "obtaining second feature information according to the second sample image" specifically include:

[0039] S21, first input the sample image into the trained image model for comparison, and obtain a pre-processed image with rough position information of the positioning frame;

[0040] S22, using an image recognition algorithm to obtain the pixel coordinates of the center point of the positioning frame on the preprocessed image.

[0041] In step S21, in some embodiments, the image model is an image model obtained by training through a deep learning algorithm based on sufficient sample images collected in advance, and the obtained sample images are input into the image model for comparison to obtain a preprocessed image. The sample images described above are sample images of straws stored in different positions, different abnormal states of the straws, and different positions of positioning frames on each type of straw tray, which are collected in advance. There are many types of straw trays and positioning frames for storing straws in the prior art. For each type of straw tray, straws and positioning frames in different positions and straws in different abnormal states are photographed to obtain sample images. Here, different positions refer to different positions of the straws and positioning frames on the straw tray in each sample image. Different abnormal states of the straws refer to abnormal states such as frost, fog, reflection, and obstruction. The number of straws in each sample image is not limited by this application and can be set arbitrarily. The placement position of the straws can also be set arbitrarily. Regarding the preset number of sample images recorded above, each type of straw tray needs to take a preset number of images, and the preset number of images can be 100, 200, 500, etc. The specific value of the preset number of images is flexibly set according to the recognition accuracy and recognition difficulty of the trained image model.

[0042] Furthermore, the coarse position information of the positioning frame specifically refers to the length and width of the positioning frame and the pixel coordinates of the intersection of the frame of the positioning frame. It should be noted that if the positioning frame is a rectangle, the coarse position information is the length and width of the rectangle and the pixel coordinates of the intersection of the rectangular frame; if the positioning frame is a circle, the coarse position information is the length and width of the rectangular frame corresponding to the circle and the pixel coordinates of the intersection of the rectangular frame.

[0043] like Figure 6 As shown, in step S22, the step of "obtaining the pixel coordinates of the center point of the positioning frame by using an image recognition algorithm for the pre-processed image" specifically includes:

[0044] S221, first perform edge detection on the preprocessed image to identify the edge of the positioning frame;

[0045] S222, extracting the edge of the positioning frame to obtain a contour image of the positioning frame;

[0046] S223, obtaining the pixel coordinates of the center point of the positioning frame according to the contour image of the positioning frame.

[0047] In some embodiments, if the positioning frame is a rectangle, the center point of the positioning frame is the intersection of the diagonal lines; if the positioning frame is a circle, the center point of the positioning frame is the center of the circle.

[0048] In the above steps, a deep learning algorithm is first used to coarsely locate the sample image, and then an image recognition algorithm is used for fine positioning to obtain the position of each positioning frame. By combining the deep learning algorithm with the image recognition algorithm, compared with the existing method, the position of the positioning frame can be accurately located, the overall positioning efficiency can be improved, and the recognition time and recognition cost can be shortened.

[0049] In step S4, the multiple offset positions include multiple first offset positions, and the step of “determining multiple offset positions according to the first feature information and the second feature information” specifically includes:

[0050] S41, calculating the difference coordinates between the pixel coordinates of the corresponding positioning frame center point according to the first initial calibration position and the first target calibration position as the first offset position.

[0051] In some embodiments, for example, Figure 2 As shown, there are eight positioning frames, which are numbered 1-8, and the first offset positions of the eight positioning frames are calculated respectively. Taking the positioning frame numbered 1 as an example, the pixel coordinates of the center point of the positioning frame at the first initial calibration position are (X 1 , Y 1 ), the pixel coordinates of the center point of the positioning frame of the first target calibration position are (X 1 ', Y 1 '), the pixel coordinates of the first offset position are (X a1 , Y a1 ), where X a1 =X 1 -X 1 ', Y a1 =Y 1 -Y 1 ', and so on, the pixel coordinates (X a2 , Y a2 )、(X a3 , Y a3 )、(X a4 , Y a4 )、(X a5 , Y a5 )、(X a6 , Y a6 )、(X a7 , Y a7 )、(X a8 , Y a8 ).

[0052] Furthermore, if Figure 2 and Figure 3As shown, the midpoint of the connecting line of the center points of adjacent positioning frames is the calibration point. In step S2, the first feature information also includes multiple second initial calibration positions of the multiple calibration points; the step of "obtaining the first feature information according to the first sample image" further includes:

[0053] S210, determining the pixel coordinates corresponding to the calibration point according to the first initial calibration position of the adjacent positioning frame as the second initial calibration position;

[0054] In step S3, the second feature information further includes a plurality of second target calibration positions of the plurality of calibration points; and the step of “obtaining the second feature information according to the second sample image” further includes:

[0055] S310: Determine pixel coordinates corresponding to the calibration point according to the second initial calibration position of the adjacent positioning frame as the second target calibration position.

[0056] The “adjacent positioning frames” refer to the positioning frames adjacent to each other along the rotation direction of the straw tray. Figure 2 The adjacent positioning frames shown in the figure are number 1 and number 8, number 2 and number 7, number 3 and number 6, number 4 and number 5 respectively.

[0057] In step S210 and step S310, in some embodiments, as Figure 3 As shown in the figure, taking the positioning frames No. 1 and No. 8 as an example, the two positioning frames are adjacent positioning frames along the rotation direction of the straw plate, the midpoint of the connecting line between the two positioning frames is the calibration point, recorded as point M, and the pixel coordinates of the second initial calibration position are (X 11 , Y 11 ), where the pixel coordinates of the first initial calibration position of the positioning frame numbered 1 are (X 1 , Y 1 ), the pixel coordinates of the first initial calibration position of the positioning frame No. 8 are (X 8 , Y 8 ), then, X 11 =X 1 +X 8 / 2,Y 11 =Y 1 +Y 8 / 2, the pixel coordinates of the second target calibration position are (X 11 ', Y 11 '), the pixel coordinates of the first initial calibration position of the positioning frame numbered 1 are (X 1 ', Y 1 '), the pixel coordinates of the first initial calibration position of the positioning frame No. 8 are (X 8 ', Y 8 '), then, X 11 '=X 1'+X 8 ' / 2,Y 11 '=Y 1 '+Y 8 ' / 2, and so on, respectively calculate the second initial calibration position and the second target calibration position between the positioning frames No. 2 and No. 7, the positioning frames No. 3 and No. 6, and the positioning frames No. 4 and No. 5, and thus obtain the pixel coordinates of the second initial calibration position as (X 22 , Y 22 )、(X 33 , Y 33 )、(X 44 , Y 44 ), the pixel coordinates of the second target calibration position are (X 22 ', Y 22 '),(X 33 ', Y 33 '),(X 44 ', Y 44 ').

[0058] Further, based on step S210 and step S310, in step S4, the multiple offset positions also include multiple second offset positions, and the step of “determining multiple offset positions according to the first feature information and the second feature information” specifically includes:

[0059] S410, calculating the difference coordinates between the corresponding pixel coordinates according to the second initial calibration position and the second target calibration position as the second offset position.

[0060] In some embodiments, illustratively, continue as Figure 2 As shown, continuing to take the positioning frames No. 1 and No. 8 as an example, the pixel coordinates of the second initial calibration position are (X 11 , Y 11 ), the pixel coordinates of the second target calibration position are (X 11 ', Y 11 '), then the pixel coordinates of the second offset position are (X b1 , Y b1 ), where X b1 =X 11 -X 11 ', Y b1 =Y 11 -Y 11 ', and so on, the pixel coordinates of the second offset position are obtained in sequence (X b2 , Y b2 )、(X b3 , Y b3 )、(X b4 , Y b4 ).

[0061] The deviation correction control method of the present invention also includes:

[0062] Before the straw plate rotates, the straw to be grasped and the corresponding initial position are obtained;

[0063] Control the robot arm to move to the initial position.

[0064] In some embodiments, the initial position of the straw is determined by the robot arm using its own three-point positioning function according to the camera calibration image, that is, before the straw tray rotates, that is, the straw tray is at the top of the tank body and is at room temperature, the robot arm moves in advance to the initial position corresponding to the straw to be grasped. In addition, if the straw needs to be grasped before the straw tray rotates, the straw can be grasped directly at the determined initial position.

[0065] like Figure 7 As shown, in step S5, the step of "controlling the robot arm to move according to a plurality of offset positions to grab the straw" specifically includes:

[0066] S51, matching a corresponding offset position according to the straw to be grasped;

[0067] S52, controlling the robot arm to move according to the corresponding offset position to grab the straw.

[0068] In some embodiments, those skilled in the art store in advance the mapping relationship between the straws and the offset positions in the database. For example, the first offset position of the positioning frame numbered 1 corresponds to the straw as follows: Figure 2 The area indicated by A in the figure, the first offset position of the positioning frame numbered 8 corresponds to the straw as shown in FIG. Figure 2 In the area indicated by B, the straw corresponding to the second offset position of the center point M of the line between number 1 and number 8 is as follows: Figure 2 The area indicated by C in the figure is deduced by analogy, and the present invention will not be repeated here. By simultaneously determining the first offset position and the second offset position, the corresponding relationship between the straw and the offset position can be more finely divided, so that the offset position of each straw can be determined more accurately, thereby improving the grasping accuracy of the robot arm.

[0069] Specifically, in actual operation, as the straw tray is extended into the interior of the tank and is in a deep low temperature environment, the straw tray rotates to offset its position. The cloud server or the local machine matches the corresponding offset position according to the straw to be grasped. When the straw needs to be grasped, the robotic arm is controlled to move according to the corresponding offset position to grasp the straw.

[0070] The deviation correction control method of the present invention also includes:

[0071] The pixel coordinates of the corresponding offset position are converted into corresponding world coordinates and sent to the robotic arm.

[0072] In step S52, the step of "controlling the robot arm to move according to the corresponding offset position to grab the straw" specifically includes: controlling the robot arm to move according to the world coordinates of the corresponding offset position to grab the straw.

[0073] In some embodiments, the conversion between pixel coordinates and world coordinates adopts the conventional conversion method, which will not be described in detail in the present invention.

[0074] In summary, the deviation correction control method of the present invention obtains a first sample image of the straw disk that has not rotated and a second sample image of the straw disk that has rotated when the straw disk rotates; obtains first feature information based on the first sample image; wherein the first feature information includes multiple first initial calibration positions of the multiple positioning frames; obtains second feature information based on the second sample image; wherein the second feature information includes multiple first target calibration positions of the multiple positioning frames; determines multiple offset positions based on the first feature information and the second feature information; and controls the robotic arm to grab the straws according to the offset positions. The present invention obtains the positions of the straw disk before and after rotation respectively, and then determines the offset position based on the position before and after rotation, so that the robotic arm grabs the straws according to the offset position, thereby ensuring accurate grabbing of the robotic arm.

[0075] Based on the above-mentioned embodiments, the present invention further provides a biological storage library, comprising a controller, wherein the controller is configured to execute the above-mentioned deviation correction control method.

[0076] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. A deviation correction control method for a biological storage library, characterized in that: The biological storage library comprises a tank body, a straw tray and a mechanical arm, wherein the straw tray is rotatably arranged in the tank body, a plurality of straw frames and a plurality of positioning frames are arranged on the straw tray, the plurality of positioning frames are arranged on both sides of the straw tray, the plurality of straw frames store straws, and the mechanical arm is arranged to be able to grab the straws; The deviation correction control method comprises: When the straw tray rotates, obtaining a first sample image of the straw tray that has not rotated and a second sample image of the straw tray that has rotated; Acquire first feature information according to the first sample image; wherein the first feature information includes a plurality of first initial calibration positions of the plurality of positioning frames; Acquire second feature information according to the second sample image; wherein the second feature information includes a plurality of first target calibration positions of the plurality of positioning frames; Determining a plurality of offset positions according to the first feature information and the second feature information; The robot arm is controlled to move according to the plurality of offset positions to grasp the straw.

2. The deviation correction control method of the biological storage bank according to claim 1, characterized in that: The first initial calibration position and the first target calibration position refer to the pixel coordinates of the center point of the positioning frame; The steps of "obtaining first feature information according to the first sample image" and "obtaining second feature information according to the second sample image" specifically include: First, the sample image is input into the trained image model for comparison to obtain a pre-processed image with the rough position information of the positioning frame; The pixel coordinates of the center point of the positioning frame are obtained by using an image recognition algorithm on the preprocessed image.

3. The deviation correction control method of the biological storage bank according to claim 2, characterized in that: The step of "obtaining the pixel coordinates of the center point of the positioning frame by using an image recognition algorithm on the preprocessed image" specifically includes: First, edge detection is performed on the preprocessed image to identify the edge of the positioning frame; Extracting the edge of the positioning frame to obtain a contour image of the positioning frame; The pixel coordinates of the center point of the positioning frame are obtained according to the contour image of the positioning frame.

4. The deviation correction control method of the biological storage bank according to claim 3, characterized in that: The midpoint of the connecting line of the adjacent center points of the positioning frame is a calibration point, and the first characteristic information further includes a plurality of second initial calibration positions of the plurality of calibration points; The step of "obtaining first feature information according to the first sample image" further includes: Determining the pixel coordinates corresponding to the calibration point according to the first initial calibration position adjacent to the positioning frame as the second initial calibration position; Among them, "adjacent positioning frames" refer to the positioning frames adjacent to each other along the rotation direction of the straw tray.

5. The deviation correction control method of the biological storage bank according to claim 4, characterized in that: The second characteristic information also includes a plurality of second target calibration positions of the plurality of calibration points; The step of "obtaining second feature information according to the second sample image" further includes: The pixel coordinates corresponding to the calibration point are determined according to the second initial calibration position adjacent to the positioning frame to serve as the second target calibration position.

6. The deviation correction control method of the biological storage bank according to claim 1, characterized in that: The multiple offset positions include multiple first offset positions, and the step of "determining multiple offset positions according to the first feature information and the second feature information" specifically includes: The difference coordinates between the pixel coordinates of the corresponding center point of the positioning frame are calculated according to the first initial calibration position and the first target calibration position to serve as the first offset position.

7. The deviation correction control method of the biological storage bank according to claim 6, characterized in that: The multiple offset positions also include multiple second offset positions, and the step of "determining multiple offset positions according to the first feature information and the second feature information" specifically includes: The difference coordinates between the corresponding pixel coordinates are calculated according to the second initial calibration position and the second target calibration position to serve as the second offset position.

8. The deviation correction control method of the biological storage bank according to claim 1, characterized in that: The deviation correction control method further comprises: Before the straw tray rotates, obtaining the straws to be grasped and the corresponding initial positions; Controlling the robotic arm to move to the initial position; The step of "controlling the robot arm to move according to the multiple offset positions to grab the straw" specifically includes: Matching a corresponding offset position according to the straw to be grasped; The mechanical arm is controlled to move according to the corresponding offset position to grab the straw.

9. The deviation correction control method of the biological storage bank according to claim 8, characterized in that: The deviation correction control method further comprises: Convert the pixel coordinates of the corresponding offset position into corresponding world coordinates, and send them to the robotic arm; The step of "controlling the mechanical arm to move according to the corresponding offset position to grab the straw" specifically includes: The robot arm is controlled to move according to the world coordinates of the corresponding offset position to grab the straw.

10. A biostorage storage device, characterized in that: It comprises a controller, which is configured to execute the correction control method as claimed in any one of claims 1 to 9.

Images