Crop variety identification method, device, electronic device and storage medium

Abstract

The present invention provides a crop variety identification method, device, electronic device and storage medium, which relates to the field of artificial intelligence technology. The method includes: obtaining plant phenotypic parameter values ​​of each target dimension of the crop to be identified after time-space alignment as each phenotypic characteristic data corresponding to the crop to be identified; performing quantum state characteristic representation on each phenotypic characteristic data corresponding to the crop to be identified, obtaining quantum superposition state representation of each phenotypic characteristic data corresponding to the crop to be identified, as quantum state phenotypic characteristic data corresponding to the crop to be identified; inputting the quantum state phenotypic characteristic data corresponding to the crop to be identified into a crop variety identification model, and obtaining the variety identification result of the crop to be identified output by the crop variety identification model. The crop variety identification method, device, electronic device and storage medium provided by the present invention can use the crop variety identification model to more accurately identify the differences between different varieties of the same type, thereby improving the accuracy of crop variety identification.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a crop variety identification method, device, electronic equipment and storage medium. Background Art

[0002] Plant phenotypic parameters refer to parameters that can reflect the phenotype of plants. Plant phenotypic parameters generally include morphological plant phenotypic parameters and physiological plant phenotypic parameters. Morphological plant phenotypic parameters mainly describe the appearance and structural characteristics of plants, including but not limited to crop height, stem diameter, leaf area or leaf area index, leaf angle, stem length, and plant spacing. Physiological plant phenotypic parameters mainly reflect the physiological functions and metabolic states of plants, including but not limited to chlorophyll content, photosynthetic rate, water stress, biomass, salt tolerance, and leaf water content.

[0003] In the related art, the traditional crop variety identification method can identify the crop variety based on the plant phenotypic parameters of the crop. However, since the phenotypic differences between different varieties of crops of the same type are very small, it is difficult to accurately identify the crop variety based on the above traditional crop variety identification method. Therefore, how to more accurately identify the crop variety is a technical problem to be solved in this field. Summary of the invention

[0004] The present invention provides a crop variety identification method, device, electronic device and storage medium, which are used to solve the defect that it is difficult to accurately identify crop varieties in the prior art, and to achieve more accurate identification of crop varieties.

[0005] The present invention provides a crop variety identification method, comprising the following steps.

[0006] Obtaining the plant phenotypic parameter value of each target dimension after the crop to be identified is aligned in time and space as each phenotypic characteristic data corresponding to the crop to be identified;

[0007] Performing quantum state characteristic representation on each phenotypic characteristic data corresponding to the crop to be identified, and obtaining a quantum superposition state representation of each phenotypic characteristic data corresponding to the crop to be identified as the quantum state phenotypic characteristic data corresponding to the crop to be identified;

[0008] Inputting the quantum state phenotype characteristic data corresponding to the crop to be identified into a crop variety identification model, and obtaining a variety identification result of the crop to be identified output by the crop variety identification model;

[0009] Among them, the crop variety identification model is constructed based on a quantum neural network and is trained based on the quantum state phenotypic characteristic data corresponding to the sample crop and the variety identification result of the sample crop; the sample crop and the crop to be identified belong to the same type of crop; the quantum state phenotypic characteristic data corresponding to the sample crop is obtained by performing quantum superposition state representation on the phenotypic characteristic data corresponding to the sample crop; each phenotypic characteristic data corresponding to the sample crop is the plant phenotypic parameter value of each target dimension after the sample crop is aligned in time and space.

[0010] According to a crop variety identification method provided by the present invention, when there are multiple crops to be identified, each of the crops to be identified belongs to the same type of crop;

[0011] In the case that there are multiple crops to be identified, after inputting the quantum state phenotype feature data corresponding to the crops to be identified into the crop variety identification model and obtaining the variety identification result of the crops to be identified output by the crop variety identification model, the method further includes: based on the variety identification result of each crop to be identified, grouping the crops to be identified of the same variety into one group to obtain a plurality of crop groups;

[0012] In the case where there are multiple crop groups, based on the quantum state phenotype characteristic data corresponding to each crop to be identified, obtaining a quantum state characteristic score of each crop to be identified;

[0013] Based on the quantum state characteristic scores of each crop to be identified in each crop group, obtaining a comprehensive evaluation score of the crop variety corresponding to each crop group;

[0014] Based on the comprehensive evaluation scores of the crop varieties corresponding to each of the crop groups, the optimal crop variety among each of the crops to be identified is determined.

[0015] According to a crop variety identification method provided by the present invention, the step of obtaining the quantum state characteristic score of each crop to be identified based on the quantum state phenotype characteristic data corresponding to each crop to be identified includes:

[0016] Based on the quantum state phenotypic characteristic data corresponding to each crop to be identified and each phenotypic characteristic data corresponding to each crop to be identified, a probability distribution value of each phenotypic characteristic data corresponding to each crop to be identified is calculated;

[0017] Based on the probability distribution value of each phenotypic characteristic data corresponding to each crop to be identified, the quantum state characteristic score of each crop to be identified is calculated.

[0018] According to a crop variety identification method provided by the present invention, the comprehensive evaluation score of the crop variety corresponding to each crop group is obtained based on the quantum state characteristic score of each crop to be identified in each crop group, including:

[0019] The weighted sum of the quantum state characteristic scores of each crop to be identified in each crop group is calculated as the comprehensive evaluation score of the crop variety corresponding to each crop group, and the weight corresponding to each crop to be identified in each crop group is determined based on the principal component analysis method.

[0020] According to a crop variety identification method provided by the present invention, the method of determining the best crop variety among the crops to be identified based on the comprehensive evaluation scores of the crop varieties corresponding to the crop groups comprises:

[0021] Based on the comprehensive evaluation scores of the crop varieties corresponding to each of the crop groups, the optimal crop variety among the crops to be identified is determined using a target algorithm, wherein the target algorithm includes a maximum likelihood estimation algorithm and / or a Bayesian optimization algorithm.

[0022] According to a crop variety identification method provided by the present invention, after determining the best crop variety among the crops to be identified based on the comprehensive evaluation scores of the crop varieties corresponding to the crop groups, the method further includes:

[0023] The crop to be evaluated with the highest quantum state characteristic score in the crop group corresponding to the optimal crop variety is determined as the optimal crop among the crops to be identified.

[0024] According to a crop variety identification method provided by the present invention, the step of obtaining the plant phenotypic parameter value of each target dimension of the crop to be identified after time-space alignment includes:

[0025] Acquiring multimodal plant phenotypic data of the crop to be identified and three-dimensional point cloud data of the crop to be identified;

[0026] Based on the internal and external parameter data of each crop phenotyping sensor when collecting the plant identification data of each mode of the crop to be identified and the internal and external parameter data of the three-dimensional point cloud data acquisition device when collecting the three-dimensional point cloud data of the crop to be identified, the multi-modal plant phenotyping data of the crop to be identified and the three-dimensional point cloud data of the crop to be identified are temporally and spatially aligned to obtain the multi-modal plant phenotyping data of the crop to be identified and the three-dimensional point cloud data of the crop to be identified after the temporal and spatial alignment;

[0027] The multimodal plant phenotypic data of the crop to be identified and the three-dimensional point cloud data of the crop to be identified after time-space alignment are subjected to data fusion to obtain target three-dimensional point cloud data of the crop to be identified, wherein the target three-dimensional point cloud data of the crop to be identified includes the plant phenotypic parameter value of each target dimension of the crop to be identified;

[0028] The plant phenotypic parameter value of each target dimension of the crop to be identified after the time-space alignment is obtained from the target three-dimensional point cloud data of the crop to be identified.

[0029] The present invention also provides a crop variety identification device, comprising the following modules:

[0030] A data acquisition module, used to acquire the plant phenotypic parameter value of each target dimension of the crop to be identified after the time-space alignment, as each phenotypic characteristic data corresponding to the crop to be identified;

[0031] A quantum computing module, used for performing quantum state characteristic representation on each phenotypic characteristic data corresponding to the crop to be identified, and obtaining a quantum superposition state representation of each phenotypic characteristic data corresponding to the crop to be identified as the quantum state phenotypic characteristic data corresponding to the crop to be identified;

[0032] A variety identification module, used for inputting the quantum state phenotype characteristic data corresponding to the crop to be identified into a crop variety identification model, and obtaining the variety identification result of the crop to be identified output by the crop variety identification model;

[0033] Among them, the crop variety identification model is constructed based on a quantum neural network and is trained based on the quantum state phenotypic characteristic data corresponding to the sample crop and the variety identification result of the sample crop; the sample crop and the crop to be identified belong to the same type of crop; the quantum state phenotypic characteristic data corresponding to the sample crop is obtained by performing quantum superposition state representation on the phenotypic characteristic data corresponding to the sample crop; each phenotypic characteristic data corresponding to the sample crop is the plant phenotypic parameter value of each target dimension after the sample crop is aligned in time and space.

[0034] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the crop variety identification method as described above is implemented.

[0035] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the crop variety identification method as described in any one of the above is implemented.

[0036] The present invention also provides a computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the crop variety identification method as described above is implemented.

[0037] The crop variety identification method, device, electronic device and storage medium provided by the present invention obtain the plant phenotypic parameter value of each target dimension of the crop to be identified after time-space alignment as each phenotypic characteristic data corresponding to the crop to be identified, then perform quantum state characteristic representation on each phenotypic characteristic data corresponding to the crop to be identified, obtain the quantum superposition state representation of each phenotypic characteristic data corresponding to the crop to be identified as the quantum state phenotypic characteristic data corresponding to the crop to be identified, and then input the quantum state phenotypic characteristic data corresponding to the crop to be identified into the crop variety identification model, obtain the variety identification result of the crop to be identified output by the crop variety identification model, and construct a crop variety identification model by combining the concept of quantum computing with the classical neural network theory. A new crop variety identification model is built, which can utilize the parallelism of quantum computing and the characteristics of quantum superposition states to enable the crop variety identification model to capture greater feature information and learn the subtle differences between the phenotypic characteristics of crops of different varieties of the same type, thereby enhancing the crop variety identification model's ability to identify the differences between different varieties of the same type, thereby improving the accuracy of crop variety identification. Using quantum state feature representation to process crop phenotypic data can enable the model to operate more effectively in high-dimensional data space, which can improve the efficiency of crop variety identification. Quantum computing has potential advantages in data encryption and security, and can provide more accurate guidance and services for agricultural production. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0039] Figure 1 This is one of the flow charts of the crop variety identification method provided by the present invention.

[0040] Figure 2 This is the second flow chart of the crop variety identification method provided by the present invention.

[0041] Figure 3 It is a structural schematic diagram of the crop variety identification device provided by the present invention.

[0042] Figure 4 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] In the description of the invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" 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 it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0045] In the description of the present application, the terms "first", "second", etc. are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are usually a class, and the number of objects is not limited. For example, the first object can be one or more. In addition, in the description of the present application, "and / or" represents at least one of the connected objects, and the character " / " generally represents that the front and back associated objects are in an "or" relationship.

[0046] It should be noted that plant phenotype refers to the physical, physiological and biochemical properties that can reflect the structural and functional characteristics of plant cells, tissues, organs, plants and populations. Its essence is actually the temporal three-dimensional expression of the plant gene map, its regional differentiation characteristics and the laws of intergenerational evolution. It is an important factor in determining traits such as crop yield, quality and stress resistance.

[0047] With the rise of precision agriculture technology in the field of modern agriculture, plant phenotypic parameters have become an important data basis for crop breeding and crop production management. In related technologies, traditional crop variety identification methods can identify crop varieties based on plant phenotypic parameters of crops.

[0048] However, since the phenotypic differences between different varieties of crops of the same type are very small, it is difficult to accurately identify the crop varieties based on the above traditional crop variety identification methods.

[0049] In this regard, the present invention provides a method for identifying crop varieties. The method for identifying crop varieties provided by the present invention utilizes quantum computing and neural network technology to construct a method that can accurately and efficiently identify varieties. Through the high parallelism and complex optimization capabilities of quantum computing, the accuracy and efficiency of crop variety identification can be greatly improved, and the best individuals can be accurately screened for further breeding.

[0050] Combine the following Figure 1-Figure 2 The crop variety identification method of the present invention is described.

[0051] Figure 1 FIG. 1 is one of the flow charts of the crop variety identification method provided by the present invention, such as Figure 1 As shown, the method includes the following: Step 101, obtaining the plant phenotypic parameter value of each target dimension after the crop to be identified is aligned in time and space, as each phenotypic characteristic data corresponding to the crop to be identified.

[0052] It should be noted that the execution subject in the embodiment of the present invention is a crop variety identification device. The crop variety identification device can be configured in electronic devices such as computers and servers.

[0053] Specifically, the crop to be identified is the identification object of the crop variety identification method provided by the present invention. The crop to be identified can be determined based on actual needs, and the crop to be identified is not specifically limited in the embodiment of the present invention.

[0054] Optionally, the types of crops to be identified may include, but are not limited to, food crops such as wheat, rice, corn, soybeans, and potatoes, and the types of crops to be identified may also include vegetable crops and cash crops.

[0055] It should be noted that crop varieties refer to a group of crop populations that have relatively stable genetic characteristics and consistent biological, morphological and economic traits, formed through artificial or natural selection under certain ecological and economic conditions and according to human needs. Crops of the same type can include multiple different varieties.

[0056] In the embodiment of the present invention, the plant phenotypic parameter values ​​of each target dimension after the crop to be identified is aligned in time and space can be obtained in a variety of ways. For example, in the embodiment of the present invention, the plant phenotypic parameter values ​​of each target dimension after the crop to be identified is aligned in time can be obtained through a variety of technologies such as data query, data processing, deep learning, image processing, and point cloud sorting.

[0057] It should be noted that the target dimension in the embodiment of the present invention may be determined according to actual requirements and / or prior knowledge. The target dimension is not specifically limited in the embodiment of the present invention.

[0058] Optionally, the target dimension may include morphological dimensions, such as plant height, leaf area, volume, canopy structure, color characteristics, and texture characteristics; the target dimension may also include spectral dimensions, such as vegetation and / or water indices such as the Normalized Difference Vegetation Index (NDVI), Enhanced Vegetation Index (EVI), and Normalized Difference Water Index (NDWI); the target dimension may also include temperature dimensions, such as the Normalized Relative Canopy Temperature (NRCT) and the Crop Water Stress Index (CWSI) that can reflect the moisture status and overall effect of the crop.

[0059] The plant phenotypic parameter values ​​in the morphological dimension can be obtained using radar sensors, depth cameras, RGB image sensors, etc. The plant phenotypic parameter values ​​in the spectral dimension can be calculated based on spectral images collected by multispectral sensors or remote sensing satellites.

[0060] After obtaining the plant phenotypic parameter value of each target dimension after the crop to be identified is aligned in time and space, the prepared phenotypic parameter value of each target dimension after the crop to be identified is aligned in time can be determined as each phenotypic characteristic data corresponding to the crop to be identified.

[0061] It can be understood that the plant phenotypic parameter value of each target dimension after the crop to be identified is aligned in time and space, and the preparation phenotypic parameter value of each target dimension after the crop to be identified is aligned in time.

[0062] As an optional embodiment, obtaining the plant phenotypic parameter value of each target dimension after the crop to be identified is aligned in time and space, including: obtaining multimodal plant phenotypic data of the crop to be identified and three-dimensional point cloud data of the crop to be identified.

[0063] Figure 2 FIG. 2 is a flow chart of the crop variety identification method provided by the present invention. Figure 2 As shown, in an embodiment of the present invention, multimodal plant phenotyping data of a crop to be identified can be obtained based on a crop phenotyping system including a plurality of crop phenotyping sensors.

[0064] The crop phenotyping system in the embodiment of the present invention may include but is not limited to an RGB visible light camera, which is used to obtain visible light images of crops and provide color, texture and morphological data; a multispectral camera, which is used to collect spectral data in multiple specific spectral bands to help analyze the health status, photosynthesis efficiency, etc. of crops; a long-wave infrared camera (thermal infrared), which is used to capture the temperature distribution of crops and reflect the moisture status and transpiration of crops; a laser radar (LiDAR) sensor, which is used to collect accurate three-dimensional structural information, including the height, volume and shape of crops; a depth camera, which is used to further enhance the morphological and three-dimensional structural information of crops through active or passive depth perception.

[0065] It should be noted that the above-mentioned crop phenotyping system includes a mobile body (such as an unmanned vehicle or a robotic phenotyping platform), and the above-mentioned crop phenotyping sensors can be uniformly installed on the above-mentioned mobile body in a fixed manner, and the relative positions of the crop phenotyping sensors are completely fixed.

[0066] It should be noted that in the crop field test, the above-mentioned mobile body can drive each crop phenotyping sensor to collect multimodal plant phenotyping data of the crop to be identified. In order to better realize the data fusion between each crop phenotyping sensor at the same time and higher-level real-time fusion. In the embodiment of the present invention, the ROS distributed operating system can be used. During the collection process, the ROS node of each crop phenotyping sensor is subscribed, and then the plant phenotyping data subscribed at the same time is saved.

[0067] It should be noted that, in the embodiment of the present invention, the laser radar sensor in the above-mentioned crop phenotyping system can be used to obtain the three-dimensional point cloud data of the crop to be identified.

[0068] After obtaining the multimodal plant phenotypic data of the crop to be identified and the three-dimensional point cloud data of the crop to be identified, data preprocessing may be performed on the image data in the plant phenotypic data of the crop to be identified and the three-dimensional point cloud data of the crop to be identified.

[0069] The preprocessing of the image data in the plant phenotyping data of the crop to be identified mainly includes image segmentation to obtain the crop ROI mask (two-dimensional ROI) and color restoration.

[0070] Specifically, in order to increase the separation of the crop to be identified and the background so as to efficiently obtain a better crop ROI, it is necessary to quickly perform relevant single plant segmentation on the crop to be identified. However, the single plant segmentation of the three-dimensional point cloud data of the crop to be identified is not as accurate as the image segmentation, and the existing image segmentation efficiency is significantly higher. In addition, due to the low resolution, the segmentation accuracy of multispectral and thermal images is not as good as that of RGB images. Therefore, the present invention uses image segmentation networks such as FastSAM to segment the crop area of ​​the crop to be identified from the original RGB image (i.e., the maximum outline surrounding the crop), draws the outline on a new canvas (or inverts all other outline pixels to black) to obtain a complete crop mask of the crop to be identified, and then restores the color of the original crop outline area to the segmented binary image to achieve color restoration.

[0071] In the preprocessing of the three-dimensional point cloud data of the crops to be identified, the main task is to remove noise and optimize the data volume to improve the quality and processing efficiency of the point cloud, which mainly involves key steps such as denoising and downsampling.

[0072] Among them, due to the uncertainty of the sensor, the original point cloud may contain noise points, which will affect the subsequent phenotyping analysis. Denoising is aimed at removing these abnormal points to make the point cloud more accurate. Common denoising methods include statistical filtering and radius filtering. Statistical filtering removes points with large abnormal deviations based on the statistical characteristics within the point neighborhood; while radius filtering removes isolated points with too low density by analyzing the local density of each point. These methods balance the denoising effect and information retention by setting appropriate parameters. Point cloud data is usually very large, and direct processing will increase the computational burden. Therefore, it is necessary to downsample the point cloud to reduce the amount of data. The commonly used downsampling method is voxel grid downsampling, which divides the point cloud into a three-dimensional grid of fixed size and selects a representative point in each grid. This method can significantly reduce the number of points while ensuring the overall morphology of the point cloud, and optimize storage and computing performance. Point cloud data that has been denoised and downsampled can more efficiently participate in the subsequent multimodal phenotypic feature extraction and provide reliable data support for crop analysis.

[0073] Based on the internal and external parameter data of each crop phenotyping sensor when collecting plant identification data of each mode of the crop to be identified and the internal and external parameter data of the three-dimensional point cloud data acquisition device when collecting the three-dimensional point cloud data of the crop to be identified, the multimodal plant phenotypic data of the crop to be identified and the three-dimensional point cloud data of the crop to be identified are temporally and spatially aligned to obtain the multimodal plant phenotypic data of the crop to be identified and the three-dimensional point cloud data of the crop to be identified after temporal and spatial alignment.

[0074] Specifically, after completing the data preprocessing of the image data in the plant phenotypic data of the crop to be identified and the three-dimensional point cloud data of the crop to be identified, the image data in the plant phenotypic data of the crop to be identified and the three-dimensional point cloud data of the crop to be identified can be fused with other indicator phenotypic data of the crop to be identified under a common field of view, and the color of the part outside the mask of the crop is uniformly set to black (corresponding to the RGB component of [0,0,0]). Then, the corresponding black color component is removed in the fusion stage, and finally the ROI multimodal point cloud model (3D ROI) of the crop is obtained, in which the fused point cloud contains information such as position, color, temperature, spectrum, etc., to integrate the phenotypic information of different crop phenotypic sensors.

[0075] Data fusion is divided into the following two steps: data space synchronization and data time synchronization. Spatial synchronization is based on the location information of each crop phenotyping sensor (such as the external parameter matrix of LiDAR and RGB camera), and the prepared phenotyping data collected by different crop phenotyping sensors are spatially aligned and registered to a unified three-dimensional coordinate system. Time synchronization is based on the timestamp information at the time of collection and the posture information of the phenotyping platform, and the multimodal plant phenotyping data collected in the crop scene are time-synchronized, thereby obtaining the multimodal plant phenotyping data of the crop to be identified after time-space alignment and the three-dimensional point cloud data of the crop to be identified.

[0076] Among them, the intrinsic parameters are mainly used to describe the internal characteristics of the camera, such as focal length, principal point position and lens distortion coefficient. The steps of camera intrinsic parameter calibration are as follows: first, use a calibration plate with known geometric characteristics (such as a chessboard) to shoot, and ensure that multiple images are obtained from different angles to cover different areas in the field of view; then extract image feature points, identify the corner points of the chessboard in the calibration image, and record the positions of these feature points in the image coordinate system; then solve the intrinsic parameter matrix, match the image coordinates of the feature points with their geometric positions in the real world, and use the camera calibration algorithm (such as Zhang Zhengyou calibration method) to solve the camera's intrinsic parameters, including focal length, principal point position, distortion coefficient, etc. Finally, by minimizing the reprojection error, optimize the camera's intrinsic parameter results to make the calibration more accurate.

[0077] The extrinsic parameters describe the relationship between the camera and the world coordinate system, mainly the position (rotation and translation). The steps of extrinsic parameter calibration are as follows: First, use multiple sensors to shoot relevant landmarks in the same scene, extract their common feature points, and usually identify the features of the same object (such as corners or planes) in the image and point cloud; then solve the extrinsic parameter matrix, and use the PnP (Perspective-n-Point) algorithm to solve the rotation matrix and translation vector of the camera, that is, the extrinsic parameters, by matching the three-dimensional coordinates of the common viewpoint with the coordinates in the image; finally, verify the calibration accuracy, apply the calibration results to the new data, and check whether the three-dimensional points in the image are accurately mapped to the corresponding positions in the actual scene. If the error is large, recalibration or optimization is required. Usually, LiDAR or other devices are used as references to the global coordinate system, and the data of each camera is aligned with the global coordinate system through extrinsic calibration.

[0078] It should be noted that in crop scenarios, crop phenotyping sensors often collect dynamic data, such as the movement of unmanned vehicles or the movement of animals by wind. Therefore, the time synchronization of each crop phenotyping sensor is crucial for data fusion. Each crop phenotyping sensor generates a corresponding timestamp when collecting data. To avoid clock drift problems between different devices, all crop phenotyping sensors need to use the same time base (such as the current unified system clock of ROS). Timestamp synchronization is usually divided into software synchronization and hardware synchronization. Software synchronization usually uses linear interpolation or more complex time synchronization algorithms to adjust the data collection time of different crop phenotyping sensors to align them with the reference time. Hardware synchronization unifies their timestamps based on the external trigger mechanism inside the crop phenotyping sensor.

[0079] In addition, PTP time synchronization is also commonly used for timestamp synchronization of crop phenotyping sensors in network devices (such as lidar, multispectral cameras, etc.), which is mainly divided into two ways: software and hardware. Software time synchronization is implemented through the operating system and network protocol stack, while hardware time synchronization requires network devices (such as switches) to support the PTP protocol.

[0080] The sampling frequencies of various crop phenotyping sensors may be different, and further fine matching of the data frames with synchronized timestamps is required. For example, the LiDAR point cloud data may receive 10 frames of point cloud messages per second, while the RGB camera may generate dozens of frames per second. For fusion, the image frame closest to the LiDAR point cloud generation time can be selected according to the frame rate of the RGB camera, or the RGB frame can be downsampled. The frame with the greatest impact on the fusion result is selected as the key frame to ensure effective alignment of multimodal data.

[0081] The multimodal plant phenotypic data of the crop to be identified and the three-dimensional point cloud data of the crop to be identified after time and space alignment are fused to obtain the target three-dimensional point cloud data of the crop to be identified. The target three-dimensional point cloud data of the crop to be identified contains the plant phenotypic parameter value of each target dimension of the crop to be identified.

[0082] Specifically, based on the spatial alignment and time synchronization of the multimodal plant phenotypic data of the crop to be identified and the three-dimensional point cloud data of the crop to be identified, the multimodal plant phenotypic data of the crop to be identified can be fused into the three-dimensional point cloud data of the crop to be identified to obtain the target three-dimensional point cloud data of the crop to be identified. Each point in the target three-dimensional point cloud data of the crop to be identified contains multi-dimensional plant phenotypic data.

[0083] The three-dimensional point cloud data of the crop to be identified obtained after data fusion contains rich plant phenotypic data such as the global three-dimensional morphology, color texture, spectral reflectance, temperature distribution, etc. of the crop to be identified, which can provide a reliable basis for the extraction of phenotypic parameter values ​​in multi-dimensional preparation.

[0084] The plant phenotypic parameter value of each target dimension of the crop to be identified is obtained from the target three-dimensional point cloud data of the crop to be identified after time-space alignment.

[0085] Specifically, after obtaining the target three-dimensional point cloud data of the crop to be identified, the plant phenotypic parameter value of each target dimension of the crop to be identified after spatiotemporal alignment can be obtained based on the target three-dimensional point cloud data of the crop to be identified through data extraction, mathematical statistics, and numerical calculation.

[0086] The embodiment of the present invention can ensure the consistency of these data in time and space by performing spatiotemporal alignment of multimodal plant phenotypic data and three-dimensional point cloud data based on the internal and external reference data of each crop phenotypic sensor and three-dimensional point cloud data acquisition device when collecting data, which helps to reduce errors caused by slight differences (such as position, timestamp, etc.) when different sensors or devices collect data, thereby improving the accuracy of subsequent analysis. By fusing the multimodal plant phenotypic data and three-dimensional point cloud data after spatiotemporal alignment, not only the details of the original data can be retained, but also the target three-dimensional point cloud data can be more comprehensively reflected by integrating information from different modalities. Through automated and intelligent data processing processes, the efficiency and accuracy of crop variety identification can be further improved, providing strong technical support for agricultural scientific research, crop breeding, production management and other fields.

[0087] Step 102: Perform quantum state feature representation on each phenotypic characteristic data corresponding to the crop to be identified, and obtain quantum superposition state representation of each phenotypic characteristic data corresponding to the crop to be identified as the quantum state phenotypic characteristic data corresponding to the crop to be identified.

[0088] Specifically, in the embodiments of the present invention, Identify the crops to be identified, is a positive integer greater than 0.

[0089] For In the embodiment of the present invention, crops to be identified can be Indicates The phenotypic characteristic data corresponding to the crops to be identified are Indicates The crop to be identified corresponds to the Phenotypic characteristic data, namely The crops to be identified are aligned in time and space. The plant phenotypic parameter value of the target dimension, , represents the total number of each target dimension. Accordingly, .

[0090] It should be noted that, in the embodiment of the present invention, Phenotypic characteristic data corresponding to the crops to be identified It exists in the form of a vector. Phenotypic characteristic data corresponding to the crops to be identified The dimension is .

[0091] The quantum state characteristics are represented by Phenotypic characteristic data corresponding to the crops to be identified Mapped into the quantum state. Phenotypic characteristic data corresponding to the crops to be identified To represent the quantum state characteristics, we can Phenotypic characteristic data corresponding to the crops to be identified Expressed as The superposition state on the quantum bit, Phenotypic characteristic data corresponding to the crops to be identified The corresponding quantum state It can be expressed as:

[0092]

[0093] in, represents different quantum ground states (qubits), Represents the probability amplitude corresponding to each quantum ground state.

[0094] For Phenotypic characteristic data corresponding to the crops to be identified Tensor decomposition can reduce high-dimensional data into multiple compact feature subspaces, and use quantum state vectorization to represent these feature vectors. By mapping classical data with quantum states, we can obtain the first Phenotypic characteristic data corresponding to the crops to be identified The quantum superposition representation of , the mapping formula is as follows:

[0095]

[0096] in, Indicates The crop to be identified corresponds to the The weight of the phenotypic characteristic data, The crop to be identified corresponds to the The weight of phenotypic characteristics It can be predefined based on prior knowledge and / or actual conditions.

[0097] Get the Phenotypic characteristic data corresponding to the crops to be identified The quantum superposition representation of Afterwards, you can Phenotypic characteristic data corresponding to the crops to be identified The quantum superposition representation of Determined as the Quantum state phenotypic characteristic data corresponding to the crops to be identified .

[0098] In the embodiment of the present invention, by performing a superposition state representation of quantum state feature representations on each phenotypic characteristic data corresponding to the crop to be identified, a quantum superposition state representation of each phenotypic characteristic data corresponding to the crop to be identified is obtained, and each phenotypic characteristic data corresponding to the crop to be identified can be constructed as a quantum state, thereby being able to establish a more complex variety characteristic representation in the superposition and entanglement of quantum states, and can enhance the characteristic representation of each phenotypic characteristic data corresponding to the crop to be identified, and can provide a more accurate data basis for crop variety identification.

[0099] Step 103: input the quantum state phenotype characteristic data corresponding to the crop to be identified into the crop variety identification model, and obtain the variety identification result of the crop to be identified output by the crop variety identification model;

[0100] Among them, the crop variety recognition model is constructed based on a quantum neural network and is trained based on the quantum state phenotypic characteristic data corresponding to the sample crops and the variety recognition results of the sample crops; the sample crops and the crops to be identified belong to the same type of crops; the quantum state phenotypic characteristic data corresponding to the sample crops are obtained by quantum superposition state representation of each phenotypic characteristic data corresponding to the sample crops; each phenotypic characteristic data corresponding to the sample crop is the plant phenotypic parameter value of each target dimension after the sample crop is aligned in time and space.

[0101] Specifically, in an embodiment of the present invention, a crop variety recognition model can be constructed based on quantum neural networks (QNNs), and the quantum state phenotypic characteristic data corresponding to the sample crops is used as samples, and the variety recognition results of the sample crops are used as sample labels. The above-mentioned crop variety recognition model is trained to obtain a trained crop variety recognition model.

[0102] It should be noted that the sample crops and the crops to be identified in the embodiment of the present invention are crops of the same type. There are multiple sample crops, and the sample crops may include multiple different varieties of crops.

[0103] It should be noted that the quantum state phenotype characteristic data corresponding to the sample crop can be obtained in the same manner as that of obtaining the quantum state phenotype characteristic data corresponding to the crop to be identified, which will not be described in detail in the embodiments of the present invention.

[0104] It should be noted that the variety identification results of the sample crops can be marked by technicians.

[0105] The training process of the crop variety recognition model can utilize the characteristics of quantum parallel computing to explore multiple complex feature combinations in a very short time and adjust the weight of each feature through quantum gate operations. The training target of the crop variety recognition model is the loss function Minimizing, the loss function of the crop variety recognition model is as follows:

[0106]

[0107] in, It is represented as the parameter matrix of the crop variety identification model; represents the bias term, It represents the predicted variety recognition result of the sample crop output by the crop variety recognition model under training. represents the predicted variety identification result (sample label) of the sample crop, Represents the loss sub-function (cross entropy loss or squared difference loss can be used).

[0108] Since quantum computing can accelerate the traditional gradient descent process, the crop variety identification model built based on quantum neural networks can efficiently optimize model parameters.

[0109] Get the Quantum state phenotypic characteristic data corresponding to the crops to be identified Afterwards, you can Quantum state phenotypic characteristic data corresponding to the crops to be identified Input the trained crop variety recognition model, and then obtain the output of the above trained crop variety recognition model. The variety identification results of the crops to be identified.

[0110] The embodiment of the present invention obtains the plant phenotypic parameter value of each target dimension of the crop to be identified after time-space alignment as each phenotypic characteristic data corresponding to the crop to be identified, then performs quantum state characteristic representation on each phenotypic characteristic data corresponding to the crop to be identified, obtains the quantum superposition state representation of each phenotypic characteristic data corresponding to the crop to be identified as the quantum state phenotypic characteristic data corresponding to the crop to be identified, and then inputs the quantum state phenotypic characteristic data corresponding to the crop to be identified into the crop variety identification model, obtains the variety identification result of the crop to be identified output by the crop variety identification model, and constructs a new crop variety identification model by combining the concept of quantum computing with the classical neural network theory. The type can utilize the parallelism of quantum computing and the characteristics of quantum superposition states to enable the crop variety recognition model to capture greater feature information and learn the subtle differences between the phenotypic characteristics of crops of different varieties of the same type, thereby enhancing the crop variety recognition model's ability to identify the differences between different varieties of the same type, thereby improving the accuracy of crop variety recognition. Using quantum state feature representation to process crop phenotypic data can enable the model to operate more effectively in high-dimensional data space, which can improve the efficiency of crop variety recognition. Quantum computing has potential advantages in data encryption and security, and can provide more accurate guidance and services for agricultural production.

[0111] As an optional embodiment, when there are multiple crops to be identified, each crop to be identified belongs to the same type of crop.

[0112] When there are multiple crops to be identified, the quantum state phenotypic characteristic data corresponding to the crops to be identified are input into the crop variety identification model, and the variety identification results of the crops to be identified output by the crop variety identification model are obtained. The method also includes: based on the variety identification results of each crop to be identified, the crops to be identified of the same variety are grouped together to obtain several crop groups.

[0113] When there are multiple crop groups, the quantum state characteristic score of each crop to be identified is obtained based on the quantum state phenotypic characteristic data corresponding to each crop to be identified.

[0114] It can be understood that, when there are multiple crop groups, it can be explained that the crops to be identified include crops of multiple varieties.

[0115] Specifically, in the embodiments of the present invention, Identify the crop group, is a positive integer greater than 0. The crops to be identified in the crop groups are all varieties , that is, The crop varieties corresponding to the crop groups are varieties .

[0116] For In the crop group In the embodiment of the present invention, the crop to be identified can be In the crop group Quantum state phenotypic characteristic data corresponding to the crops to be identified , through numerical calculation, we can get In the crop group The quantum state feature score of the crops to be identified .

[0117] As an optional embodiment, based on the quantum state phenotypic characteristic data corresponding to each crop to be identified, the quantum state characteristic score of each crop to be identified is obtained, including: based on the quantum state phenotypic characteristic data corresponding to each crop to be identified and each phenotypic characteristic data corresponding to each crop to be identified, the probability distribution value of each phenotypic characteristic data corresponding to each crop to be identified is calculated.

[0118] Specifically, for the In the crop group The crops to be identified can be In the crop group Quantum state phenotypic characteristic data corresponding to the crops to be identified and In the crop group The crop to be identified corresponds to the Phenotypic characteristics data , calculated by the following formula In the crop group The probability distribution value of the crops to be identified :

[0119]

[0120] in, Indicates In the crop group Quantum state phenotypic characteristic data corresponding to the crops to be identified and In the crop group The crop to be identified corresponds to the Phenotypic characteristics data The inner product of In the crop group The crop to be identified corresponds to the Phenotypic characteristics data For In the crop group Quantum state phenotypic characteristic data corresponding to the crops to be identified The size of the contribution.

[0121] Based on the probability distribution value of each phenotypic characteristic data corresponding to each crop to be identified, the quantum state characteristic score of each crop to be identified is calculated.

[0122] Specifically, obtain the In the crop group The probability distribution value of the crops to be identified Afterwards, the following formula can be used to calculate the In the crop group The quantum state feature score of the crops to be identified :

[0123]

[0124] in, Indicates The weight of the plant phenotyping parameters in the target dimension. Weights of plant phenotypic parameters in the target dimension It can be determined by prior knowledge and / or actual conditions (e.g., through statistical analysis or model training).

[0125] Based on the quantum state characteristic scores of each crop to be identified in each crop group, a comprehensive evaluation score of the crop variety corresponding to each crop group is obtained.

[0126] Specifically, obtain the In the crop group The quantum state feature score of the crops to be identified Afterwards, based on In the crop group The quantum state feature score of the crops to be identified , through numerical calculation, we can get Varieties corresponding to each crop group Comprehensive evaluation score .

[0127] As an optional embodiment, based on the quantum state characteristic scores of each crop to be identified in each crop group, a comprehensive evaluation score of the crop variety corresponding to each crop group is obtained, including: calculating the weighted sum of the quantum state characteristic scores of each crop to be identified in each crop group as the comprehensive evaluation score of the crop variety corresponding to each crop group, and the weight corresponding to each crop to be identified in each crop group is determined based on the principal component analysis method.

[0128] Specifically, obtain the In the crop group The quantum state feature score of the crops to be identified Afterwards, the following formula can be used to calculate Varieties corresponding to each crop group Comprehensive evaluation score :

[0129]

[0130] in, Indicates The total number of crops to be identified in the crop group; Indicates In the crop group The weight corresponding to the crop to be identified.

[0131] No. In the crop group The weights corresponding to the crops to be identified It can be determined based on the principal component analysis (PCA) method. The specific determination method includes: In the crop group The quantum state feature score of the crops to be identified After dimensionality reduction by principal component analysis, according to In the crop group The quantum state feature score of the crops to be identified The corresponding variance contributes to the total variance, and determines the In the crop group The weights corresponding to the crops to be identified The higher the variance contribution rate of the principal component, the greater the weight corresponding to the individual.

[0132] Based on the comprehensive evaluation scores of the crop varieties corresponding to each crop group, the optimal crop variety among each crop to be identified is determined.

[0133] Specifically, after obtaining the comprehensive scoring scores of the crop varieties corresponding to each crop group, the optimal varieties of each crop to be identified can be determined based on the comprehensive scoring scores of the crop varieties corresponding to each crop group through numerical calculation, mathematical statistics, and deep learning.

[0134] As an optional embodiment, based on the comprehensive evaluation scores of the crop varieties corresponding to each crop group, the optimal crop variety among each crop to be identified is determined, including: based on the comprehensive evaluation scores of the crop varieties corresponding to each crop group, using a target algorithm to determine the optimal crop variety among each crop to be identified, the target algorithm including a maximum likelihood estimation algorithm and / or a Bayesian optimization algorithm.

[0135] It should be noted that due to the Varieties corresponding to each crop group Comprehensive evaluation score There are uncertain factors (i.e., there are probabilities) in the calculation process. Therefore, in the embodiment of the present invention, after obtaining the comprehensive scoring scores of the crop varieties corresponding to each crop group, the maximum likelihood estimation and Bayesian optimization algorithm can be used to determine the optimal crop variety among the crops to be identified.

[0136] Optionally, the Bayesian optimization algorithm is a optimization algorithm suitable for an objective function with uncertainty. For the comprehensive scoring scores of the crop varieties corresponding to each crop group, the Bayesian optimization algorithm can explore and utilize information by constructing a probability model, for example, simulating the comprehensive index function through a Gaussian process (GP) model, and finally finding the variety with the best comprehensive index by updating the posterior distribution of the model according to each calculation. The Bayesian optimization algorithm in the embodiment of the present invention can be expressed as:

[0137]

[0138] in, represents the mean function, represents the covariance function, represents the crop variety and crop varieties The relationship between Indicates the same category as the crop variety A different crop variety.

[0139] After Bayesian optimization, the Gaussian process (GP) probability model constructed by maximizing the comprehensive index can be used to screen out the optimal crop varieties for each crop to be estimated. , the formula is:

[0140]

[0141] The embodiment of the present invention divides crops to be identified of the same variety into a group based on the variety identification result of each crop to be identified. After obtaining a plurality of crop groups, if the number of crop groups is multiple, the quantum state characteristic score of each crop to be identified is obtained based on the quantum state phenotypic characteristic data corresponding to each crop to be identified, and then the comprehensive evaluation score of the crop variety corresponding to each crop group is obtained based on the quantum state characteristic score of each crop to be identified in each crop group, and the optimal crop variety of each crop to be identified is determined based on the comprehensive evaluation score of the crop variety corresponding to each crop group, and the quantum state phenotypic characteristic data can be used to accurately identify and comprehensively evaluate crop varieties, thereby realizing the optimal selection of crop varieties. Through automated variety evaluation, the decision-making efficiency of crop variety optimization can be improved. The invention is particularly suitable for large-scale crop management and variety selection processes, and provides strong technical support for agricultural production.

[0142] As an optional embodiment, after determining the best crop variety among the crops to be identified based on the comprehensive evaluation scores of the crop varieties corresponding to each crop group, the method also includes: determining the crop to be evaluated with the highest quantum state characteristic score in the crop group corresponding to the best crop variety as the best crop among the crops to be identified.

[0143] Specifically, determine the optimal crop variety for each crop to be evaluated Afterwards, the best crop varieties can be The crop to be evaluated with the highest quantum state characteristic score in the corresponding crop group is determined as the best crop among all crops to be identified. , the formula is as follows:

[0144]

[0145] in, The best crop variety The corresponding crop group The quantum state feature score of the crop to be identified.

[0146] Selected best crops It will be used for further breeding or preserved as germplasm resources.

[0147] The embodiment of the present invention combines the analysis results of phenotypic parameter values ​​of different dimensions to establish a comprehensive evaluation model, and finally selects the individuals with the most outstanding performance in multi-dimensional traits, and finally realizes the breeding and screening of high-quality crops. This link not only improves the overall quality of crop varieties through accurate phenotypic identification, but also provides a basis for breeders to accurately select and improve. The final result is that high-quality crop varieties with higher yields, stronger disease resistance, and wider adaptability can be screened, achieving a major breakthrough in breeding work. This phenotypic analysis technology based on multi-source sensor data helps to improve breeding efficiency, shorten the breeding cycle, and significantly increase the market competitiveness of new varieties.

[0148] The crop variety identification method provided by the present invention can simultaneously obtain multi-dimensional phenotypic parameter values ​​of crops from RGB, spectrum, thermal infrared, LiDAR and depth camera through data fusion of multi-source crop phenotyping sensors. Compared with the traditional single sensor system, the present invention effectively enhances the ability to capture different phenotypic characteristics of crops.

[0149] The crop variety identification method provided by the present invention can effectively improve the accuracy of crop phenotypic identification by representing multidimensional phenotypic parameters in quantum states, especially significantly enhancing the sensitivity to different varieties and individual differences. Through the construction and training of quantum neural network models, the crop variety identification method provided by the present invention can achieve more accurate variety identification.

[0150] The crop variety identification method provided by the present invention uses the probability of quantum state characteristic scores of different individuals of the same variety of crops to perform weighted combination to construct a comprehensive index of the variety, which can systematically and comprehensively describe various characteristics of the crop. The crop variety identification method provided by the present invention uses Bayesian optimization to solve the maximum value of the comprehensive index of the variety to obtain the optimal variety and the optimal individual.

[0151] In summary, the crop variety identification method provided by the present invention utilizes quantum computing and neural network technology to construct a method that can accurately and efficiently identify varieties. Through the high parallelism and complex optimization capabilities of quantum computing, the crop variety identification method provided by the present invention can greatly improve the accuracy and efficiency of crop variety identification, and can accurately screen out the best individuals for further breeding.

[0152] Figure 3 Schematic diagram of the structure of the crop variety identification device provided by the present invention. Figure 3 The crop variety identification device provided by the present invention is described. The crop variety identification device described below and the crop variety identification method provided by the present invention described above can be referred to each other. Figure 3 As shown, the device includes: a data acquisition module 301, a quantum computing module 302 and a variety identification module 303.

[0153] The data acquisition module 301 is used to acquire the plant phenotypic parameter value of each target dimension after the crop to be identified is aligned in time and space, as each phenotypic characteristic data corresponding to the crop to be identified;

[0154] The quantum computing module 302 is used to perform quantum state characteristic representation on each phenotypic characteristic data corresponding to the crop to be identified, and obtain the quantum superposition state representation of each phenotypic characteristic data corresponding to the crop to be identified as the quantum state phenotypic characteristic data corresponding to the crop to be identified;

[0155] The variety identification module 303 is used to input the quantum state phenotype characteristic data corresponding to the crop to be identified into the crop variety identification model, and obtain the variety identification result of the crop to be identified output by the crop variety identification model;

[0156] Among them, the crop variety recognition model is constructed based on a quantum neural network and is trained based on the quantum state phenotypic characteristic data corresponding to the sample crops and the variety recognition results of the sample crops; the sample crops and the crops to be identified belong to the same type of crops; the quantum state phenotypic characteristic data corresponding to the sample crops are obtained by quantum superposition state representation of each phenotypic characteristic data corresponding to the sample crops; each phenotypic characteristic data corresponding to the sample crop is the plant phenotypic parameter value of each target dimension after the sample crop is aligned in time and space.

[0157] Specifically, the data acquisition module 301, the quantum computing module 302 and the variety identification module 303 are electrically connected.

[0158] The crop variety identification device in the embodiment of the present invention obtains the plant phenotypic parameter value of each target dimension of the crop to be identified after time-space alignment as each phenotypic characteristic data corresponding to the crop to be identified, then performs quantum state characteristic representation on each phenotypic characteristic data corresponding to the crop to be identified, obtains the quantum superposition state representation of each phenotypic characteristic data corresponding to the crop to be identified as the quantum state phenotypic characteristic data corresponding to the crop to be identified, and then inputs the quantum state phenotypic characteristic data corresponding to the crop to be identified into the crop variety identification model, obtains the variety identification result of the crop to be identified output by the crop variety identification model, and constructs a new crop variety identification model by combining the concept of quantum computing with the classical neural network theory. The variety identification model can utilize the parallelism of quantum computing and the characteristics of quantum superposition states to enable the crop variety identification model to capture greater feature information and learn the subtle differences between the phenotypic characteristics of crops of different varieties of the same type, thereby enhancing the crop variety identification model's ability to identify the differences between different varieties of the same type, thereby improving the accuracy of crop variety identification. Using quantum state feature representation to process crop phenotypic data can enable the model to operate more effectively in high-dimensional data space, thereby improving the efficiency of crop variety identification. Quantum computing has potential advantages in data encryption and security, and can provide more accurate guidance and services for agricultural production.

[0159] Figure 4 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 4As shown, the electronic device may include: a processor (processor) 410 , a communication interface (Communications Interface) 420 , a memory (memory) 430 and a communication bus 440 , wherein the processor 410 , the communication interface 420 , and the memory 430 communicate with each other through the communication bus 440 . The processor 410 can call the logic instructions in the memory 430 to execute the crop variety identification method, which includes: obtaining the plant phenotypic parameter value of each target dimension after the crop to be identified is aligned in time and space, as each phenotypic characteristic data corresponding to the crop to be identified; performing quantum state characteristic representation on each phenotypic characteristic data corresponding to the crop to be identified, and obtaining the quantum superposition state representation of each phenotypic characteristic data corresponding to the crop to be identified, as the quantum state phenotypic characteristic data corresponding to the crop to be identified; inputting the quantum state phenotypic characteristic data corresponding to the crop to be identified into the crop variety identification model, and obtaining the variety identification result of the crop to be identified output by the crop variety identification model; wherein the crop variety identification model is constructed based on a quantum neural network, and is trained based on the quantum state phenotypic characteristic data corresponding to the sample crop and the variety identification result of the sample crop; the sample crop and the crop to be identified belong to the same type of crop; the quantum state phenotypic characteristic data corresponding to the sample crop is obtained by performing quantum superposition state representation on each phenotypic characteristic data corresponding to the sample crop; each phenotypic characteristic data corresponding to the sample crop is the plant phenotypic parameter value of each target dimension after the sample crop is aligned in time and space.

[0160] In addition, the logic instructions in the above-mentioned memory 430 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0161] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the crop variety identification method provided by the above-mentioned methods, and the method includes: obtaining the plant phenotypic parameter value of each target dimension of the crop to be identified after spatiotemporal alignment, as each phenotypic characteristic data corresponding to the crop to be identified; performing quantum state characteristic representation on each phenotypic characteristic data corresponding to the crop to be identified, obtaining the quantum superposition state representation of each phenotypic characteristic data corresponding to the crop to be identified, as the quantum state phenotypic characteristic data corresponding to the crop to be identified data; inputting the quantum state phenotypic characteristic data corresponding to the crop to be identified into the crop variety identification model, and obtaining the variety identification result of the crop to be identified output by the crop variety identification model; wherein the crop variety identification model is constructed based on a quantum neural network, and is trained based on the quantum state phenotypic characteristic data corresponding to the sample crop and the variety identification result of the sample crop; the sample crop and the crop to be identified belong to the same type of crop; the quantum state phenotypic characteristic data corresponding to the sample crop is obtained by performing a quantum superposition state representation on the phenotypic characteristic data corresponding to the sample crop; each phenotypic characteristic data corresponding to the sample crop is the plant phenotypic parameter value of each target dimension after the sample crop is aligned in time and space.

[0162] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which is implemented by a processor to execute the crop variety identification method provided by the above methods when the computer program is executed, the method comprising: obtaining the plant phenotypic parameter value of each target dimension of the crop to be identified after spatiotemporal alignment as each phenotypic characteristic data corresponding to the crop to be identified; performing quantum state characteristic representation on each phenotypic characteristic data corresponding to the crop to be identified, obtaining the quantum superposition state representation of each phenotypic characteristic data corresponding to the crop to be identified, as the quantum state phenotypic characteristic data corresponding to the crop to be identified; inputting the quantum state phenotypic characteristic data corresponding to the crop to be identified into a crop variety identification model, and obtaining the variety identification result of the crop to be identified output by the crop variety identification model; wherein the crop variety identification model is constructed based on a quantum neural network, and is trained based on the quantum state phenotypic characteristic data corresponding to the sample crop and the variety identification result of the sample crop; the sample crop and the crop to be identified belong to the same type of crop; the quantum state phenotypic characteristic data corresponding to the sample crop are obtained by performing quantum superposition state representation on each phenotypic characteristic data corresponding to the sample crop; each phenotypic characteristic data corresponding to the sample crop is the plant phenotypic parameter value of each target dimension after spatiotemporal alignment of the sample crop.

[0163] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0164] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0165] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A crop variety identification method, characterized in that: include: Obtaining the plant phenotypic parameter value of each target dimension after the crop to be identified is aligned in time and space as each phenotypic characteristic data corresponding to the crop to be identified; Performing quantum state characteristic representation on each phenotypic characteristic data corresponding to the crop to be identified, and obtaining a quantum superposition state representation of each phenotypic characteristic data corresponding to the crop to be identified as the quantum state phenotypic characteristic data corresponding to the crop to be identified; Inputting the quantum state phenotype characteristic data corresponding to the crop to be identified into a crop variety identification model, and obtaining a variety identification result of the crop to be identified output by the crop variety identification model; The crop variety recognition model is constructed based on a quantum neural network and is trained based on quantum state phenotypic characteristic data corresponding to sample crops and variety recognition results of the sample crops; the sample crops and the crop to be identified belong to the same type of crop; the quantum state phenotypic characteristic data corresponding to the sample crops are obtained by performing quantum superposition state representation on each phenotypic characteristic data corresponding to the sample crops; each phenotypic characteristic data corresponding to the sample crops is the plant phenotypic parameter value of each target dimension after the sample crops are aligned in time and space; In the case where there are multiple crops to be identified, each of the crops to be identified belongs to the same type of crops; In the case that there are multiple crops to be identified, after inputting the quantum state phenotype feature data corresponding to the crops to be identified into the crop variety identification model and obtaining the variety identification result of the crops to be identified output by the crop variety identification model, the method further includes: Based on the variety identification result of each of the crops to be identified, the crops to be identified of the same variety are grouped together to obtain a plurality of crop groups; In the case where there are multiple crop groups, based on the quantum state phenotype characteristic data corresponding to each crop to be identified, obtaining a quantum state characteristic score of each crop to be identified; Based on the quantum state characteristic scores of each crop to be identified in each crop group, obtaining a comprehensive evaluation score of the crop variety corresponding to each crop group; Based on the comprehensive evaluation scores of the crop varieties corresponding to each of the crop groups, the optimal crop variety among each of the crops to be identified is determined.

2. The crop variety identification method according to claim 1, characterized in that: The step of obtaining the quantum state characteristic score of each crop to be identified based on the quantum state phenotype characteristic data corresponding to each crop to be identified includes: Based on the quantum state phenotypic characteristic data corresponding to each crop to be identified and each phenotypic characteristic data corresponding to each crop to be identified, a probability distribution value of each phenotypic characteristic data corresponding to each crop to be identified is calculated; Based on the probability distribution value of each phenotypic characteristic data corresponding to each crop to be identified, the quantum state characteristic score of each crop to be identified is calculated.

3. The crop variety identification method according to claim 2, characterized in that: The step of obtaining a comprehensive evaluation score of the crop varieties corresponding to each crop group based on the quantum state characteristic scores of each crop to be identified in each crop group includes: The weighted sum of the quantum state characteristic scores of each crop to be identified in each crop group is calculated as the comprehensive evaluation score of the crop variety corresponding to each crop group, and the weight corresponding to each crop to be identified in each crop group is determined based on the principal component analysis method.

4. The crop variety identification method according to claim 3, characterized in that: The step of determining the best crop variety among the crops to be identified based on the comprehensive evaluation scores of the crop varieties corresponding to the crop groups comprises: Based on the comprehensive evaluation scores of the crop varieties corresponding to each of the crop groups, the optimal crop variety among the crops to be identified is determined using a target algorithm, wherein the target algorithm includes a maximum likelihood estimation algorithm and / or a Bayesian optimization algorithm.

5. The crop variety identification method according to claim 4, characterized in that: After determining the best crop variety among the crops to be identified based on the comprehensive evaluation scores of the crop varieties corresponding to the crop groups, the method further includes: The crop to be evaluated with the highest quantum state characteristic score in the crop group corresponding to the optimal crop variety is determined as the optimal crop among the crops to be identified.

6. The crop variety identification method according to any one of claims 1 to 5, characterized in that: The step of obtaining the plant phenotypic parameter value of each target dimension after the crop to be identified is aligned in time and space includes: Acquiring multimodal plant phenotypic data of the crop to be identified and three-dimensional point cloud data of the crop to be identified; Based on the internal and external parameter data of each crop phenotyping sensor when collecting the plant identification data of each mode of the crop to be identified and the internal and external parameter data of the three-dimensional point cloud data acquisition device when collecting the three-dimensional point cloud data of the crop to be identified, the multi-modal plant phenotyping data of the crop to be identified and the three-dimensional point cloud data of the crop to be identified are temporally and spatially aligned to obtain the multi-modal plant phenotyping data of the crop to be identified and the three-dimensional point cloud data of the crop to be identified after the temporal and spatial alignment; The multimodal plant phenotypic data of the crop to be identified and the three-dimensional point cloud data of the crop to be identified after time-space alignment are subjected to data fusion to obtain target three-dimensional point cloud data of the crop to be identified, wherein the target three-dimensional point cloud data of the crop to be identified includes the plant phenotypic parameter value of each target dimension of the crop to be identified; The plant phenotypic parameter value of each target dimension of the crop to be identified after the time-space alignment is obtained from the target three-dimensional point cloud data of the crop to be identified.

7. A crop variety identification device, characterized in that: include: A data acquisition module, used to acquire the plant phenotypic parameter value of each target dimension of the crop to be identified after the time-space alignment, as each phenotypic characteristic data corresponding to the crop to be identified; A quantum computing module, used for performing quantum state characteristic representation on each phenotypic characteristic data corresponding to the crop to be identified, and obtaining a quantum superposition state representation of each phenotypic characteristic data corresponding to the crop to be identified as the quantum state phenotypic characteristic data corresponding to the crop to be identified; A variety identification module, used for inputting the quantum state phenotype characteristic data corresponding to the crop to be identified into a crop variety identification model, and obtaining the variety identification result of the crop to be identified output by the crop variety identification model; The crop variety recognition model is constructed based on a quantum neural network and is trained based on quantum state phenotypic characteristic data corresponding to sample crops and variety recognition results of the sample crops; the sample crops and the crop to be identified belong to the same type of crop; the quantum state phenotypic characteristic data corresponding to the sample crops are obtained by performing quantum superposition state representation on each phenotypic characteristic data corresponding to the sample crops; each phenotypic characteristic data corresponding to the sample crops is the plant phenotypic parameter value of each target dimension after the sample crops are aligned in time and space; In the case where there are multiple crops to be identified, each of the crops to be identified belongs to the same type of crops; In the case where there are multiple crops to be identified, the variety identification module is further used to: Based on the variety identification result of each of the crops to be identified, the crops to be identified of the same variety are grouped together to obtain a plurality of crop groups; In the case where there are multiple crop groups, based on the quantum state phenotype characteristic data corresponding to each crop to be identified, obtaining a quantum state characteristic score of each crop to be identified; Based on the quantum state characteristic scores of each crop to be identified in each crop group, obtaining a comprehensive evaluation score of the crop variety corresponding to each crop group; Based on the comprehensive evaluation scores of the crop varieties corresponding to each of the crop groups, the optimal crop variety among each of the crops to be identified is determined.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the crop variety identification method according to any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the crop variety identification method according to any one of claims 1 to 6 is implemented.

Images