Method and device for identifying tissue culture seedling grabbing positions based on topological features

Abstract

This application proposes a method and device for identifying the grasping position of a tissue culture seedling based on topological features, comprising the following steps: obtaining a depth image of the tissue culture seedling to be grasped, calculating the winding matrix of the depth image, and calculating the central eigenvalue of the depth image based on the winding matrix; obtaining the area of ​​the tissue culture seedling to be grasped based on the central eigenvalue, and then obtaining the thermal winding matrix and thermal density matrix of the area of ​​the tissue culture seedling to be grasped; performing a weighted summation of the thermal winding matrix, thermal density matrix, and the central area mask to obtain a grasping heat map, selecting an area in the grasping heat map with a thermal value below a preset threshold as the base of the tissue culture seedling stem, and using the base of the tissue culture seedling stem as the grasping position. This scheme utilizes the topological features of the tissue culture seedling to quantify the spatial winding relationship between the stem and leaves, deformation complexity, and geometric center position, and generates a heat map to accurately locate the grasping position of the stem base.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence, and in particular to a method and device for identifying the position of tissue culture seedlings based on topological features. Background Art

[0002] Transplanting tissue culture seedlings is one of the important technical links in tissue culture rapid propagation work. The success or failure of transplanting tissue culture seedlings will directly affect the survival rate and later growth quality. At present, the planting work in the transplanting operation still relies on skilled workers to complete, and the labor cost is high, highlighting the urgent need for automated transplanting.

[0003] The core challenge of automating the transplanting of tissue culture seedlings lies in their unique biological characteristics. Unlike cuttings, tissue culture seedlings have significant morphological heterogeneity. Plants in the same batch vary greatly in terms of leaf number, curvature, and expansion direction. The varying degrees of differentiation also lead to discrete distributions of key parameters such as plant height and leaves. This poses a huge challenge to extracting the grasping position and angle of the tissue culture seedlings. The ideal grasping position should be near the root at the base of the stem. This area not only has the highest mechanical strength, but also minimizes interference between the roots and leaves, thereby reducing the risk of tissue damage during transplantation.

[0004] Current robotic grasping and detection technology has obvious limitations in tissue culture seedling application scenarios. For example, deep learning solutions based on 2D vision rely on large-scale labeled data training, and have problems such as poor modeling and weak environmental adaptability. They are difficult to cope with the dynamic morphological changes of tissue culture seedlings. The point cloud solutions commonly used in industry are mainly aimed at standardized rigid industrial objects with fixed shapes. They are designed for standardized rigid parts and achieve grasping through geometric feature matching, but cannot handle the flexible deformation and unstructured features of plant organs, resulting in insufficient positioning accuracy.

[0005] In summary, how to accurately and automatically grasp and transplant tissue culture seedlings of different morphologies is a problem that urgently needs to be solved in the existing technology. Summary of the Invention

[0006] The embodiments of the present application provide a method and device for identifying the grasping position of tissue culture seedlings based on topological features, which utilize the topological features of tissue culture seedlings to quantify the spatial winding relationship between stems and leaves, deformation complexity and geometric center position, and generate a heat map to accurately locate the grasping position of the stem base.

[0007] In a first aspect, an embodiment of the present application provides a method for identifying a tissue culture seedling grabbing position based on topological features, the method comprising:

[0008] Acquire a depth image of the tissue culture seedling to be grasped, calculate a winding matrix of the depth image, and calculate a central eigenvalue of the depth image based on the winding matrix, wherein the central eigenvalue is used to represent the geometric center position of the tissue culture seedling to be grasped in the depth image;

[0009] A central region mask of the depth image is obtained based on the central eigenvalue, and the central region mask is used to extract the tissue culture seedling region to be grasped in the depth image. A sliding window process is performed on the tissue culture seedling region to be grasped using a preset window size, and a winding eigenvalue and a density eigenvalue of each sliding window are obtained to form a thermal winding matrix and a thermal density matrix, respectively. The winding eigenvalue is used to represent the overall winding relationship between the stem and leaves of the tissue culture seedling to be grasped in the corresponding sliding window, and the density eigenvalue is used to represent the deformation complexity of the tissue culture seedling to be grasped in the corresponding sliding window.

[0010] A weighted summation of the thermal winding matrix, the thermal density matrix and the central area mask is performed to obtain a grasping thermal map. In the grasping thermal map, an area with a thermal value lower than a preset threshold is selected as the base of the tissue culture seedling stem, and the base of the tissue culture seedling stem is used as the grasping position.

[0011] In a second aspect, an embodiment of the present application provides a topological feature-based tissue culture seedling grasping position identification device, comprising:

[0012] an acquisition module, configured to acquire a depth image of the tissue culture seedling to be grasped, calculate a winding matrix of the depth image, and calculate a central eigenvalue of the depth image based on the winding matrix, wherein the central eigenvalue is used to represent the geometric center position of the tissue culture seedling to be grasped in the depth image;

[0013] a topology calculation module, which obtains a central area mask of the depth image based on the central eigenvalue, uses the central area mask to extract the tissue culture seedling area to be grasped in the depth image, performs sliding window processing on the tissue culture seedling area to be grasped using a preset window size, and obtains the winding eigenvalue and density eigenvalue of each sliding window to respectively form a thermal winding matrix and a thermal density matrix, wherein the winding eigenvalue is used to represent the overall winding relationship between the stem and leaves of the tissue culture seedling to be grasped in the corresponding sliding window, and the density eigenvalue is used to represent the deformation complexity of the tissue culture seedling to be grasped in the corresponding sliding window;

[0014] The automatic grasping module is used to perform weighted summation on the thermal winding matrix, the thermal density matrix and the central area mask to obtain a grasping heat map, select the area with a thermal value lower than a preset threshold in the grasping heat map as the base of the tissue culture seedling stem, and automatically grasp the base of the tissue culture seedling stem.

[0015] In a third aspect, an embodiment of the present application provides an electronic device comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute a method for identifying a tissue culture seedling grasping position based on topological features.

[0016] In a fourth aspect, an embodiment of the present application provides a readable storage medium, in which a computer program is stored. The computer program includes a program code for controlling a process to execute a process. When the program code is executed by a processor, a method for identifying the position of a tissue culture seedling grasping based on topological features is implemented.

[0017] The main contributions and innovations of the present invention are as follows:

[0018] The embodiments of the present application address the problem of morphological heterogeneity in tissue culture seedlings. Without relying on large-scale labeled data, the winding eigenvalues, density eigenvalues, and center eigenvalues ​​are used to quantify the spatial winding relationship between stems and leaves, deformation complexity, and geometric center position, and combined with weighted heat maps to accurately locate the base of the stem, avoid interference between leaves and roots, and reduce the risk of transplant damage. Compared with traditional 2D visual deep learning solutions, the model has stronger generalization and environmental adaptability, does not require manual parameter adjustment, and is highly robust. Compared with industrial point cloud solutions, it can effectively handle the flexible deformation and unstructured features of plant organs, improve positioning accuracy, and provide an efficient and reliable grasping position recognition method for automated transplanting of tissue culture seedlings, helping to reduce labor costs and improve transplanting efficiency and survival rate.

[0019] The details of one or more embodiments of the present application are set forth in the following drawings and description to make other features, objects, and advantages of the present application more readily apparent. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0021] Figure 1 is a flow chart of a method for identifying a tissue culture seedling grabbing position based on topological features according to an embodiment of the present application;

[0022] Figure 2 is a relationship diagram of a density eigenvalue, a winding eigenvalue, and a center eigenvalue according to an embodiment of the present application;

[0023] Figure 3 is a schematic diagram of generating a crawl heat map according to an embodiment of the present application;

[0024] Figure 4 This is an experimental result of grabbing the base of the stem of a tissue culture seedling according to an embodiment of the present application;

[0025] Figure 5 This is a structural block diagram of a topological feature-based tissue culture seedling grabbing position identification device according to an embodiment of the present application;

[0026] Figure 6 Schematic diagram of the hardware structure of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0027] Exemplary embodiments will be described in detail herein, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numerals in different figures represent identical or similar elements, unless otherwise indicated. The implementations described in the following exemplary embodiments are not intended to represent all implementations consistent with one or more embodiments of this specification. Rather, they are merely examples of apparatuses and methods consistent with certain aspects of one or more embodiments of this specification, as detailed in the appended claims.

[0028] It should be noted that in other embodiments, the steps of the corresponding method are not necessarily performed in the order shown and described in this specification. In some other embodiments, the method may include more or fewer steps than those described in this specification. In addition, a single step described in this specification may be broken down into multiple steps for description in other embodiments, and multiple steps described in this specification may be combined into a single step for description in other embodiments.

[0029] Example 1

[0030] The embodiment of the present application provides a method for identifying the grasping position of a tissue culture seedling based on topological features, which uses the topological features of the tissue culture seedling to quantify the spatial winding relationship between the stem and the leaves, the deformation complexity and the geometric center position, and generates a heat map to accurately locate the grasping position of the stem base. Specifically, referring to Figure 1 , the method comprising:

[0031] Acquire a depth image of the tissue culture seedling to be grasped, calculate a winding matrix of the depth image, and calculate a central eigenvalue of the depth image based on the winding matrix, wherein the central eigenvalue is used to represent the geometric center position of the tissue culture seedling to be grasped in the depth image;

[0032] A central region mask of the depth image is obtained based on the central eigenvalue, and the central region mask is used to extract the tissue culture seedling region to be grasped in the depth image. A sliding window process is performed on the tissue culture seedling region to be grasped using a preset window size, and a winding eigenvalue and a density eigenvalue of each sliding window are obtained to form a thermal winding matrix and a thermal density matrix, respectively. The winding eigenvalue is used to represent the overall winding relationship between the stem and leaves of the tissue culture seedling to be grasped in the corresponding sliding window, and the density eigenvalue is used to represent the deformation complexity of the tissue culture seedling to be grasped in the corresponding sliding window.

[0033] A weighted summation of the thermal winding matrix, the thermal density matrix and the central area mask is performed to obtain a grasping thermal map. In the grasping thermal map, an area with a thermal value lower than a preset threshold is selected as the base of the tissue culture seedling stem, and the base of the tissue culture seedling stem is used as the grasping position.

[0034] In some embodiments, the present solution uses a binocular vision camera to collect 3D depth information of the tissue culture seedlings to be grasped to obtain a depth image of the tissue culture seedlings to be grasped, performs edge information detection on the depth image to obtain an edge information matrix, and calculates the Gaussian winding integral of every two pairs of vectors in the edge information matrix to form a winding matrix.

[0035] Specifically, the edge information matrix is ​​expressed as , and the edge information matrix is ​​a three-dimensional matrix, and the winding matrix T is an n*n matrix. The formula for calculating the Gaussian winding integral of each two pairs of vectors in the edge information matrix is ​​expressed as follows:

[0036]

[0037] in, Represents a vector and Gaussian surround integral, × represents cross product, · represents dot product, is the differential operator.

[0038] Specifically, edge information detection can be used to detect the contour boundaries of organs such as the stems and leaves of the tissue culture seedlings to be captured in the depth image, and a three-dimensional edge information matrix L can be constructed in the form of curves. The Gaussian wrapping integral is used to quantify the degree of winding tightness and direction of each two vectors in the edge information matrix, that is, the curves, thereby reflecting the winding interference situation, and a winding matrix T is used to record the winding situation of the entire tissue culture seedling in the depth image.

[0039] In some embodiments, the formula for calculating the central eigenvalue of the depth image based on the winding matrix is ​​expressed as follows:

[0040]

[0041] in, are the coordinates of the central eigenvalue, is the element in the winding matrix T.

[0042] Specifically, since the winding matrix records the mutual winding relationship between the stems and leaves of the tissue culture seedlings to be grasped, the center of mass of the winding matrix tends to be in the area with the highest curve density and the most frequent interactions. Therefore, this scheme uses the center of mass of the winding matrix as the central eigenvalue of the depth image to accurately locate the position of the tissue culture seedlings to be grasped.

[0043] In some embodiments, the mask size is preset, the size of the central area mask is the same as the mask size, and the geometric center point of the central area mask is the central eigenvalue.

[0044] Specifically, although the shapes of tissue culture seedlings are different, their overall size and height are basically the same. Therefore, this scheme obtains the central area mask in the depth image based on the central eigenvalue in a preset mask size, thereby ensuring that the central area mask includes all tissue culture seedling areas, so as to better obtain the base of the tissue culture seedling stem.

[0045] In some embodiments, the central area mask is used to mask the depth image to obtain the area of ​​the tissue culture plantlet to be grasped, and the area of ​​the tissue culture plantlet to be grasped includes all tissue culture plantlet structures.

[0046] Specifically, the central region mask is used to retain only the tissue culture seedling region in the depth image, thereby shielding irrelevant regions far from the center to reduce the probability of misidentification in non-target regions.

[0047] In some specific embodiments, the sliding window processing used in this solution is a conventional sliding window method. For example, the size of the area of ​​tissue culture seedlings to be captured is h*h, the preset sliding window size is w*w, the step size is s, and the sliding window starts from (0,0) of the area of ​​tissue culture seedlings to be captured until the sliding window covers the entire area of ​​tissue culture seedlings to be captured.

[0048] Specifically, during the sliding window processing, the winding eigenvalue w and the density eigenvalue d within each sliding window are calculated and stored in the corresponding positions of the thermal winding matrix and the thermal density matrix, respectively. That is, the position of each winding eigenvalue on the thermal winding matrix corresponds to the position of the corresponding sliding window in the area of ​​the tissue culture seedlings to be grasped, and the position of each density eigenvalue on the thermal density matrix corresponds to the position of the corresponding sliding window in the area of ​​the tissue culture seedlings to be grasped.

[0049] For example, if the position of the first sliding window is (0, 0), the winding eigenvalue w of the first sliding window is stored at the (0, 0) position of the thermal winding matrix, and the density eigenvalue d of the first sliding window is stored at the (0, 0) position of the thermal density matrix.

[0050] In some embodiments, edge information detection is performed on each sliding window to obtain a window edge information matrix, and the Gaussian wrapping integral of each pair of vectors in the window edge information matrix is ​​calculated to form a window wrapping matrix. The calculation formula of the wrapping eigenvalue is as follows:

[0051]

[0052] in, is the winding eigenvalue, is an element in the window wrapping matrix, and i and j represent the position coordinates in the window wrapping matrix.

[0053] That is to say, the calculation method of the window winding matrix in this solution is the same as the calculation method of the winding matrix of the entire depth image.

[0054] In some embodiments, a density threshold is preset, and the ratio of the number of elements greater than the density threshold to the number of all non-zero elements in each window winding matrix is ​​calculated as the density eigenvalue of the corresponding sliding window, wherein the density threshold is the mean of the non-zero elements in the corresponding window winding matrix.

[0055] Specifically, the mean of the non-zero elements in the window winding matrix is ​​used as the density threshold to calculate the density eigenvalue, thereby adapting to local structural differences and accurately distinguishing stems and leaves without the need for manual parameter adjustment and with strong robustness.

[0056] In some specific embodiments, the density eigenvalue, winding eigenvalue and center eigenvalue in this solution are all topological features. The relationship among the density eigenvalue, winding eigenvalue and center eigenvalue is as follows: Figure 2 As shown in the figure, in the topological structure, the winding eigenvalue is used to describe the topological invariant of the degree of winding of the spatial curve itself, so as to quantify the global twisting and crossing of the curve in three-dimensional space. Therefore, this scheme uses the winding eigenvalue to represent the overall winding relationship between the stem and leaves of the tissue culture seedling to be grasped in the corresponding sliding window in space; the density eigenvalue refers to the distribution of local curvature or torsion in the topological structure, describing the degree of geometric deformation per unit length of the curve. When the density is 0, both curves participate in the torsion process, and the deformation is evenly distributed on the two curves. When the density value is extremely large or extremely small, the system exhibits significant non-equilibrium characteristics, in which one curve maintains a quasi-linear shape, and the other spirally winds around it, forming a master-slave topological structure. When the density changes from negative to positive or from positive to negative, the system undergoes a topological role exchange. Therefore, this scheme uses the density eigenvalue to represent the deformation complexity of the tissue culture seedling to be grasped in the corresponding sliding window; the center eigenvalue describes the reference point or axis of the overall symmetry of the curve in the topological structure. Therefore, this scheme uses the center eigenvalue to represent the geometric center position of the tissue culture seedling to be grasped in the depth image.

[0057] In some embodiments, a schematic diagram of generating a crawl heat map is shown as follows: Figure 3 As shown, the formula for obtaining the crawling heat map by weighted summing the thermal winding matrix, the thermal density matrix and the central area mask is as follows:

[0058]

[0059] Among them, E is the crawling heat map, is the thermal winding matrix, is the thermal density matrix, is the center area mask, is the weight of the thermal winding matrix, is the weight of the thermal density matrix, is the weight of the center region mask.

[0060] Specifically, 、 、 The added value is 1. Further, when the thermal density matrix The matrix average of More than the entire map When the value is To increase the weight of the thermal density matrix, at the same time, to ensure 、 、 The added value is 1, set , where the d value of the entire image is a preset parameter.

[0061] In some embodiments, the grasping heat map is subjected to sliding window processing to obtain the area in the grasping heat map where the thermal value is lower than a preset threshold as the base of the tissue culture seedling stem. In this solution, the grasping measurement index is used to calculate the grasping posture, and the interference areas with high thermal values, such as leaves, roots, etc., are excluded, and then the base of the tissue culture seedling stem is automatically grasped.

[0062] Furthermore, after identifying the base of the tissue culture seedling stem, any grasping posture generation algorithm can be used in combination with the graspability evaluation algorithm to grasp the base of the tissue culture seedling stem. This scheme does not limit this. In this scheme, a fast graspability evaluation algorithm is used to grasp the base of the tissue culture seedling stem. The experimental results of grasping the base of the tissue culture seedling stem are as follows: Figure 4 shown.

[0063] Specifically, this solution accurately identifies the base of the stem of the tissue culture seedling and thus predicts the grasping angle to achieve accurate grasping of the tissue culture seedling.

[0064] Example 2

[0065] Based on the same concept, refer to Figure 5, this application also proposes a topological feature-based tissue culture seedling grabbing position recognition device, including:

[0066] an acquisition module, configured to acquire a depth image of the tissue culture seedling to be grasped, calculate a winding matrix of the depth image, and calculate a central eigenvalue of the depth image based on the winding matrix, wherein the central eigenvalue is used to represent the geometric center position of the tissue culture seedling to be grasped in the depth image;

[0067] a topology calculation module, which obtains a central area mask of the depth image based on the central eigenvalue, uses the central area mask to extract the tissue culture seedling area to be grasped in the depth image, performs sliding window processing on the tissue culture seedling area to be grasped using a preset window size, and obtains the winding eigenvalue and density eigenvalue of each sliding window to respectively form a thermal winding matrix and a thermal density matrix, wherein the winding eigenvalue is used to represent the overall winding relationship between the stem and leaves of the tissue culture seedling to be grasped in the corresponding sliding window, and the density eigenvalue is used to represent the deformation complexity of the tissue culture seedling to be grasped in the corresponding sliding window;

[0068] The automatic grasping module is used to perform weighted summation on the thermal winding matrix, the thermal density matrix and the central area mask to obtain a grasping thermal map, select the area with a thermal value lower than a preset threshold in the grasping thermal map as the base of the tissue culture seedling stem, and use the base of the tissue culture seedling stem as the grasping position.

[0069] Example 3

[0070] This embodiment also provides an electronic device, referring to Figure 6 , includes a memory 404 and a processor 402, wherein the memory 404 stores a computer program, and the processor 402 is configured to run the computer program to perform the steps in any of the above method embodiments.

[0071] Specifically, the processor 402 may include a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0072] Memory 404 may include a large-capacity memory 404 for data or instructions. By way of example, and not limitation, memory 404 may include a hard disk drive (HDD), a floppy disk drive, a solid-state drive (SSD), flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 404 may include removable or non-removable (or fixed) media. Where appropriate, memory 404 may be internal or external to the data processing device. In certain embodiments, memory 404 is non-volatile memory. In certain embodiments, memory 404 includes read-only memory (ROM) and random access memory (RAM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically alterable ROM (EAROM) or a flash memory (FLASH), or a combination of two or more of these. In appropriate circumstances, the RAM may be a static random access memory (SRAM) or a dynamic random access memory (DRAM), wherein the DRAM may be a fast page mode dynamic random access memory 404 (FPMDRAM), an extended data output dynamic random access memory (EDODRAM), a synchronous dynamic random access memory (SDRAM), etc.

[0073] The memory 404 may be used to store or cache various data files required for processing and / or communication, as well as possible computer program instructions executed by the processor 402 .

[0074] The processor 402 reads and executes computer program instructions stored in the memory 404 to implement any one of the topological feature-based tissue culture seedling grabbing position identification methods in the above embodiments.

[0075] Optionally, the electronic device may further include a transmission device 406 and an input / output device 408 , wherein the transmission device 406 is connected to the processor 402 , and the input / output device 408 is connected to the processor 402 .

[0076] Transmission device 406 can be used to receive or transmit data via a network. Specific examples of such networks may include wired or wireless networks provided by the electronic device's communications provider. In one embodiment, the transmission device includes a network interface controller (NIC), which can be connected to other network devices via a base station to enable communication with the Internet. In another embodiment, transmission device 406 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0077] The input and output device 408 is used to input or output information. In this embodiment, the input information may be a depth image of the tissue culture seedling to be grasped, and the output information may be the base of the stem of the tissue culture seedling.

[0078] Optionally, in this embodiment, the processor 402 may be configured to execute the following steps through a computer program:

[0079] Acquire a depth image of the tissue culture seedling to be grasped, calculate a winding matrix of the depth image, and calculate a central eigenvalue of the depth image based on the winding matrix, wherein the central eigenvalue is used to represent the geometric center position of the tissue culture seedling to be grasped in the depth image;

[0080] A central region mask of the depth image is obtained based on the central eigenvalue, and the central region mask is used to extract the tissue culture seedling region to be grasped in the depth image. A sliding window process is performed on the tissue culture seedling region to be grasped using a preset window size, and a winding eigenvalue and a density eigenvalue of each sliding window are obtained to form a thermal winding matrix and a thermal density matrix, respectively. The winding eigenvalue is used to represent the overall winding relationship between the stem and leaves of the tissue culture seedling to be grasped in the corresponding sliding window, and the density eigenvalue is used to represent the deformation complexity of the tissue culture seedling to be grasped in the corresponding sliding window.

[0081] A weighted summation of the thermal winding matrix, the thermal density matrix and the central area mask is performed to obtain a grasping thermal map. In the grasping thermal map, an area with a thermal value lower than a preset threshold is selected as the base of the tissue culture seedling stem, and the base of the tissue culture seedling stem is used as the grasping position.

[0082] It should be noted that the specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementation modes, and this embodiment will not be repeated here.

[0083] In general, various embodiments may be implemented in hardware or dedicated circuitry, software, logic, or any combination thereof. Some aspects of the invention may be implemented in hardware, while other aspects may be implemented in firmware or software executed by a controller, microprocessor, or other computing device, but the invention is not limited thereto. Although various aspects of the invention may be shown and described as block diagrams, flow charts, or using some other graphical representation, it should be understood that, as non-limiting examples, the blocks, devices, systems, techniques, or methods described herein may be implemented in hardware, software, firmware, dedicated circuitry or logic, general-purpose hardware or a controller or other computing device, or some combination thereof.

[0084] The embodiments of the present invention may be implemented by computer software that is executable by a data processor of a mobile device, such as in a processor entity, or by hardware, or by a combination of software and hardware. Computer software or programs (also referred to as program products) including software routines, applets and / or macros may be stored in any device-readable data storage medium, and they include program instructions for performing specific tasks. A computer program product may include one or more computer executable components that are configured to perform an embodiment when the program is run. One or more computer executable components may be at least one software code or a portion thereof. In addition, it should be noted at this point that, for example, Figure 6 Any block of the logic flow in the program may represent program steps, or interconnected logic circuits, blocks and functions, or a combination of program steps and logic circuits, blocks and functions. The software may be stored on physical media such as memory chips or memory blocks implemented within the processor, magnetic media such as hard disks or floppy disks, and optical media such as, for example, DVDs and their data variants, CDs, etc. Physical media are non-transitory media.

[0085] Those skilled in the art should understand that the technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0086] The above embodiments merely illustrate several embodiments of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A method for identifying the position of tissue culture seedlings based on topological features, characterized in that: The following steps are involved: Acquire a depth image of the tissue culture seedling to be grasped, calculate the winding matrix of the depth image, and calculate the central eigenvalue of the depth image based on the winding matrix, wherein the central eigenvalue is used to represent the geometric center position of the tissue culture seedling to be grasped in the depth image. Perform edge information detection on the depth image to obtain an edge information matrix, calculate the Gaussian winding integral of every two pairs of vectors in the edge information matrix to form a winding matrix, and the formula for calculating the Gaussian winding integral is expressed as: in, Represents a vector and Gaussian surround integral, × represents cross product, · represents dot product, is the differential operator; A central area mask of the depth image is obtained based on the central eigenvalue, and the area of ​​the tissue culture seedling to be grasped in the depth image is extracted using the central area mask. The area of ​​the tissue culture seedling to be grasped is subjected to sliding window processing using a preset window size, and the winding eigenvalue and density eigenvalue of each sliding window are obtained to respectively form a thermal winding matrix and a thermal density matrix, wherein the winding eigenvalue is used to represent the overall winding relationship between the stem and leaves of the tissue culture seedling to be grasped in the corresponding sliding window in space, and the density eigenvalue is used to represent the deformation complexity of the tissue culture seedling to be grasped in the corresponding sliding window, wherein edge information detection is performed on each sliding window to obtain a window edge information matrix, and the Gaussian wrapping integral of each two pairs of vectors in the window edge information matrix is ​​calculated to form a window winding matrix, a density threshold is preset, and the ratio of the number of elements in each window winding matrix that is greater than the density threshold to the number of all non-zero elements is calculated as the density eigenvalue of the corresponding sliding window, and the density threshold is the mean of the non-zero elements in the winding matrix of the corresponding window; A weighted summation of the thermal winding matrix, the thermal density matrix and the central area mask is performed to obtain a grasping thermal map. In the grasping thermal map, an area with a thermal value lower than a preset threshold is selected as the base of the tissue culture seedling stem, and the base of the tissue culture seedling stem is used as the grasping position.

2. A method for identifying a tissue culture seedling grabbing position based on topological features according to claim 1, characterized in that: The formula for calculating the central eigenvalue of the depth image based on the winding matrix is ​​as follows: in, are the coordinates of the central eigenvalue, is the element in the winding matrix T.

3. A method for identifying a tissue culture seedling grabbing position based on topological features according to claim 1, characterized in that: The mask size is preset, the size of the central area mask is the same as the mask size, and the geometric center point of the central area mask is the central eigenvalue.

4. A method for identifying the position of a tissue culture seedling based on topological features according to claim 1, characterized in that: The calculation formula of the winding eigenvalue is as follows: in, is the winding eigenvalue, is an element in the window wrapping matrix, and i and j represent the position coordinates in the window wrapping matrix.

5. The method for identifying the position of tissue culture seedlings based on topological features according to claim 1, wherein: The formula for obtaining the crawl heat map by weighted summing the thermal winding matrix, the thermal density matrix, and the center area mask is as follows: Among them, E is the crawling heat map, is the thermal winding matrix, is the thermal density matrix, is the center area mask, is the weight of the thermal winding matrix, is the weight of the thermal density matrix, is the weight of the center region mask.

6. A topological feature-based tissue culture seedling grabbing position recognition device, characterized in that: include: An acquisition module is used to acquire a depth image of the tissue culture seedling to be grasped, calculate a winding matrix of the depth image, and calculate a central eigenvalue of the depth image based on the winding matrix, wherein the central eigenvalue is used to represent the geometric center position of the tissue culture seedling to be grasped in the depth image, perform edge information detection on the depth image to obtain an edge information matrix, calculate the Gaussian winding integral of every two pairs of vectors in the edge information matrix to form a winding matrix, and the formula for calculating the Gaussian winding integral is expressed as: in, Represents a vector and Gaussian surround integral, × represents cross product, · represents dot product, is the differential operator; A topological calculation module obtains a central area mask of the depth image based on the central eigenvalue, uses the central area mask to extract the area of ​​the tissue culture seedling to be grasped in the depth image, performs sliding window processing on the area of ​​the tissue culture seedling to be grasped using a preset window size, and obtains the winding eigenvalue and density eigenvalue of each sliding window to respectively form a thermal winding matrix and a thermal density matrix, wherein the winding eigenvalue is used to represent the overall winding relationship between the stems and leaves of the tissue culture seedling to be grasped in the corresponding sliding window in space, and the density eigenvalue is used to represent the deformation complexity of the tissue culture seedling to be grasped in the corresponding sliding window, wherein edge information detection is performed on each sliding window to obtain a window edge information matrix, and the Gaussian wrapping integral of each two pairs of vectors in the window edge information matrix is ​​calculated to form a window winding matrix, a density threshold is preset, and the ratio of the number of elements in each window winding matrix that are greater than the density threshold to the number of all non-zero elements is calculated as the density eigenvalue of the corresponding sliding window, and the density threshold is the mean of the non-zero elements in the corresponding window winding matrix; The automatic grasping module is used to perform weighted summation on the thermal winding matrix, the thermal density matrix and the central area mask to obtain a grasping heat map, select the area with a thermal value lower than a preset threshold in the grasping heat map as the base of the tissue culture seedling stem, and automatically grasp the base of the tissue culture seedling stem.

7. An electronic device comprising a memory and a processor, characterized in that: The memory stores a computer program, and the processor is configured to run the computer program to execute the method for identifying the position of a tissue culture seedling based on topological features according to any one of claims 1 to 5.

8. A readable storage medium, characterized in that: The readable storage medium stores a computer program, which includes a program code for controlling a process to execute a process. When the program code is executed by a processor, the method for identifying the position of a tissue culture seedling based on topological features according to any one of claims 1 to 5 is implemented.

Images