Intelligent statistical method, system and electronic equipment for sugarcane germination rate and storage medium

Abstract

The application discloses a sugarcane germination rate intelligent statistical method and system, electronic equipment and storage medium, and relates to the technical field of sugarcane planting. The method comprises the following steps: constructing a target data set according to original images of a plurality of sugarcane seedlings; extracting pixel distribution features of the sugarcane seedlings from the target data set, classifying the extracted pixel distribution features, and obtaining a germination state classification result of the sugarcane seedlings; judging whether the germination state classification result meets a preset condition, and obtaining a judgment result; and when the judgment result is yes, obtaining a final germination rate statistical result of the sugarcane seedlings according to the target data set. Through image recognition and data analysis technology, the application can automatically and accurately identify and count the germination rate of the sugarcane seedlings, reduces errors caused by human factors, improves the reliability of the statistical result, greatly improves the work efficiency, and is suitable for large-scale sugarcane planting scenarios.

Description

Intelligent statistical methods, systems, electronic devices, and storage media for sugarcane germination rate Technical Field

[0001] This invention relates to the field of sugarcane planting technology, and in particular to an intelligent statistical method, system, electronic device and storage medium for sugarcane germination rate. Background Technology

[0002] Accurately calculating the germination rate of sugarcane seedlings is a crucial issue in sugarcane cultivation. Traditional statistical methods rely mainly on manual observation and recording. However, the judgment standards of different personnel may vary, leading to highly subjective statistical results that lack uniformity and accuracy. Moreover, manual statistics require a significant amount of time and manpower, especially in large-scale planting situations where checking the germination status of each seedling individually is a massive undertaking, making it difficult to obtain accurate data quickly. Summary of the Invention

[0003] The technical problem to be solved by this invention is to address the shortcomings of existing technologies. Specifically, it provides an intelligent statistical method, system, electronic device, and storage medium for sugarcane germination rate, as detailed below:

[0004] 1) In a first aspect, the present invention provides an intelligent statistical method for sugarcane germination rate, the specific technical solution of which is as follows:

[0005] S1. Construct the target dataset based on the original images of multiple sugarcane seedlings;

[0006] S2. Extract the pixel distribution features of sugarcane seedlings from the target dataset, classify the extracted pixel distribution features, and obtain the classification results of the germination status of sugarcane seedlings.

[0007] S3. Determine whether the germination status classification result meets the preset conditions and obtain the judgment result;

[0008] S4. When the judgment result is yes, the final germination rate statistics of sugarcane seedlings are obtained based on the target dataset.

[0009] The beneficial effects of the intelligent statistical method for sugarcane germination rate provided by this invention are as follows:

[0010] This invention utilizes image recognition and data analysis technologies to automatically and accurately identify and count the germination rate of sugarcane seedlings, reducing errors caused by human factors, improving the reliability of statistical results, and greatly increasing work efficiency. It is suitable for large-scale sugarcane planting scenarios.

[0011] Based on the above scheme, the intelligent statistical method for sugarcane germination rate of the present invention can be further improved as follows.

[0012] Furthermore, the extracted pixel distribution features are classified, including:

[0013] A support vector machine with radial basis functions is used to classify the extracted pixel distribution features.

[0014] Furthermore, based on the original images of multiple sugarcane seedlings, a target dataset was constructed, including:

[0015] Original images of multiple sugarcane seedlings are collected according to a preset cycle, and temperature and humidity sensors are used to collect temperature and humidity data synchronously. The original images, temperature data, and humidity data are matched according to the collection timestamp to generate a synchronous dataset.

[0016] Construct the target dataset based on the synchronized dataset.

[0017] Furthermore, it also includes:

[0018] If the judgment result is negative, adjust the acquisition frequency of the original images of sugarcane seedlings, reacquire multiple original images of sugarcane seedlings according to the adjusted acquisition frequency, and return to execute S1 until the judgment result is positive.

[0019] 2) Secondly, the present invention also provides an intelligent statistical system for sugarcane germination rate, the specific technical solution of which is as follows:

[0020] It includes a target dataset construction module, a germination status classification result acquisition module, a judgment module, and a statistics module;

[0021] The target dataset building module is used to: construct a target dataset based on the original images of multiple sugarcane seedlings;

[0022] The germination status classification result acquisition module is used to: extract the pixel distribution features of sugarcane seedlings from the target dataset, classify the extracted pixel distribution features, and obtain the germination status classification result of sugarcane seedlings;

[0023] The judgment module is used to: determine whether the germination status classification result meets the preset conditions and obtain the judgment result;

[0024] The statistics module is used to: when the judgment result is yes, obtain the final germination rate statistics of sugarcane seedlings based on the target dataset.

[0025] Based on the above scheme, the intelligent statistical system for sugarcane germination rate of the present invention can be further improved as follows.

[0026] Furthermore, the germination status classification result acquisition module is also specifically used for:

[0027] A support vector machine with radial basis functions is used to classify the extracted pixel distribution features.

[0028] Furthermore, the target dataset construction module is specifically used for:

[0029] Original images of multiple sugarcane seedlings are collected according to a preset cycle, and temperature and humidity sensors are used to collect temperature and humidity data synchronously. The original images, temperature data, and humidity data are matched according to the collection timestamp to generate a synchronous dataset.

[0030] Construct the target dataset based on the synchronized dataset.

[0031] Furthermore, it also includes calling the execution module, which is used for:

[0032] If the judgment result is negative, adjust the acquisition frequency of the original images of sugarcane seedlings, reacquire multiple original images of sugarcane seedlings according to the adjusted acquisition frequency, and re-call the target dataset construction module, the germination status classification result acquisition module, and the judgment module until the judgment result is positive.

[0033] 3) In a third aspect, the present invention also provides an electronic device, the electronic device including a processor coupled to a memory, the memory storing at least one computer program, the at least one computer program being loaded and executed by the processor, so that the electronic device implements any of the above-mentioned intelligent statistical methods for sugarcane germination rate.

[0034] 4) In a fourth aspect, the present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements any of the above-mentioned intelligent statistical methods for sugarcane germination rate.

[0035] It should be noted that the beneficial effects of the technical solutions of the second to fourth aspects of the present invention and their corresponding possible implementations can be found in the above description of the technical effects of the first aspect and its corresponding possible implementations, and will not be repeated here. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments of the present invention will be briefly introduced below:

[0037] Figure 1 is a flowchart illustrating an intelligent statistical method for sugarcane germination rate according to an embodiment of the present invention;

[0038] Figure 2 is a schematic diagram of the structure of an intelligent statistical system for sugarcane germination rate according to an embodiment of the present invention;

[0039] Figure 3 is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0040] The principles and features of the present invention are described below. The examples given are only for explaining the present invention and are not intended to limit the scope of the present invention.

[0041] The technical solution of the present invention and how the technical solution of the present invention solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of the present invention will now be described with reference to the accompanying drawings.

[0042] As shown in Figure 1, an intelligent statistical method for sugarcane germination rate according to an embodiment of the present invention includes the following steps:

[0043] S1. Construct the target dataset based on the original images of multiple sugarcane seedlings;

[0044] S2. Extract the pixel distribution features of sugarcane seedlings from the target dataset, classify the extracted pixel distribution features, and obtain the classification results of the germination status of sugarcane seedlings.

[0045] S3. Determine whether the germination status classification result meets the preset conditions, and obtain the judgment result;

[0046] S4. When the judgment result is yes, the final germination rate statistics of sugarcane seedlings are obtained based on the target dataset.

[0047] Optionally, the extracted pixel distribution features are classified, including:

[0048] A support vector machine with radial basis functions is used to classify the extracted pixel distribution features.

[0049] Optionally, a target dataset is constructed based on the original images of multiple sugarcane seedlings, including:

[0050] Original images of multiple sugarcane seedlings are collected according to a preset cycle, and temperature and humidity sensors are used to collect temperature and humidity data synchronously. The original images, temperature data, and humidity data are matched according to the collection timestamp to generate a synchronous dataset.

[0051] Construct the target dataset based on the synchronized dataset.

[0052] Optionally, it also includes:

[0053] If the judgment result is negative, adjust the acquisition frequency of the original images of sugarcane seedlings, reacquire multiple original images of sugarcane seedlings according to the adjusted acquisition frequency, and return to execute S1 until the judgment result is positive.

[0054] The following embodiments illustrate an intelligent statistical method for sugarcane germination rate according to the present invention, specifically including the following steps:

[0055] S101. Collect original images of multiple sugarcane seedlings at a preset frequency, and simultaneously collect temperature and humidity data using temperature and humidity sensors. Match the original images, temperature data, and humidity data according to the collection timestamp to generate a synchronized dataset.

[0056] S102. Extract pixel information of each original image of sugarcane seedlings from the synchronous dataset, and calculate the edge intensity of any original image using the boundary gradient method. Based on the edge intensity of the original image, use the Canny operator to generate the initial edge feature image corresponding to the original image. Perform Gaussian filtering on the initial edge feature image to eliminate noise interference. If there are blurred areas in the initial edge feature image, use the nonlocal mean denoising algorithm to sharpen them, and obtain the final edge feature image corresponding to the original image. Repeat this process until the final edge feature image corresponding to each original image is obtained.

[0057] The specific implementation process of calculating and extracting the edge intensity of any original image using the boundary gradient method is as follows:

[0058] The original image is converted to grayscale using a weighted average method. Gaussian filtering or median filtering is then used to smooth the grayscale image, reducing noise impact on edge detection. The Sobel or Prewitt operator is used to calculate the gradients in the x and y directions of the grayscale image to determine the edge direction and components. For each pixel in the image, the selected gradient operator is used to calculate its x and y gradient components, which are then expressed using the formula Gradient = sqrt(dx). 2 +dy2) calculates the gradient magnitude, where dx and dy are the gradient components in the x and y directions, respectively, and Gradient represents the edge intensity.

[0059] S103. Based on the spatial correlation matrix of any final edge feature image, obtain the dependency relationship between each pixel of the final edge feature image and its surrounding pixels, and generate the self-attention weights corresponding to the final edge feature image; multiply the self-attention weights by the pixel values ​​of each pixel in the edge feature image to obtain an attention mask; normalize the attention mask to generate a two-dimensional weight map; perform weighted fusion of the two-dimensional weight map and the edge feature image to obtain the initial weighted feature image corresponding to the final edge feature image. If the self-attention weight of any region in the initial weighted feature image is lower than a preset weight threshold, then use a non-local mean algorithm to perform local feature enhancement on the region to obtain the enhanced weighted feature image, which is recorded as the final weighted feature image, until the final weighted feature image corresponding to each final edge feature image is obtained.

[0060] The process of obtaining the spatial correlation matrix of any final edge feature image includes:

[0061] The grayscale values ​​of the edge feature image are quantized to the range of 0 to N_g-1, where N_g is the number of grayscale levels. The spatial relationship between pixel pairs is determined, typically represented by distance d and angle θ. For example, d can be 1, 2, etc., and θ can be 0°, 45°, 90°, 135°, etc. For each pair of pixels satisfying the spatial relationship, the frequency of grayscale value combinations is counted and normalized to obtain a probability matrix P(i,j,d,θ), where i and j are grayscale levels. A spatial correlation matrix is ​​calculated using the GLCM (Gray-Level Co-occurrence Matrix), where each element represents the correlation (dependency) of pixels at different grayscale levels in the image under a specific spatial relationship.

[0062] The process of generating the self-attention weights corresponding to the final edge feature image includes:

[0063] The spatial correlation matrix is ​​used as the input feature matrix X, where each element represents the correlation (dependency) between pixels at different locations in the image. The input feature matrix X is transformed into query vector, key vector, and value vector through learnable weight matrices WQ, WK, and WV. Query vector: Q = X·WQ, key vector: K = X·WK, value vector: V = X·WV. The similarity between the query vector and the key vector is calculated by dot product to obtain the self-attention score matrix A. The self-attention score matrix A is normalized by applying the softmax function to obtain the self-attention weight matrix W (i.e., self-attention weights).

[0064] Dilation and erosion operations are performed on each final weighted feature image to extract the initial contours of multiple sugarcane seedling germination regions. The temperature and humidity data are fused with the initial contours of each germination region using a weighted average method to obtain the fusion results. Based on the fusion results, the Otsu threshold segmentation algorithm is used to distinguish between germination regions and non-germination regions to obtain the germination region identification results of each final weighted feature image.

[0065] Specifically, dilation and erosion operations are performed on each final weighted feature image to extract the initial contours of multiple sugarcane seedling germination regions. The specific implementation process is as follows:

[0066] The final weighted feature image is converted to a grayscale image. The grayscale image is then converted to a binary image for morphological operations. Thresholding can be used to segment pixels, setting values ​​greater than a threshold to 1 (foreground) and values ​​less than the threshold to 0 (background). In the dilation operation, a 3×3 square kernel is selected, and a convolutional kernel is used to dilate the binary image. Dilation expands the foreground region, connects adjacent regions, and fills in small holes. In the erosion operation, a 3×3 square kernel is selected, and a convolutional kernel is used to erode the dilated image. Erosion shrinks the foreground region and removes small noise points. Finally, OpenCV's `findContours` function is used to extract the initial contours of multiple sugarcane seedling germination regions from the eroded image.

[0067] In this process, a weighted average method is used to perform multimodal fusion of temperature and humidity data with the initial contour of each germination region to obtain the fusion result. The specific implementation process is as follows:

[0068] Temperature and humidity data are normalized to ensure they fall within the same numerical range. For example, temperature and humidity data can be normalized to the [0,1] interval, and the contour image of the germination region (the region image enclosed by the initial contour of the germination region) is converted to a format that matches the temperature and humidity data. If the temperature and humidity data are time-series data, the time series of the contour image needs to be aligned. Weights are assigned to each modality based on the importance of temperature and humidity data to sugarcane seedling germination. The weight assignment can be based on prior knowledge or determined through optimization algorithms, ensuring that the sum of all weights is 1 to maintain the rationality of the weighted average. For the contour of each germination region, the temperature and humidity data are multiplied by their corresponding weights, and the results are summed to obtain the fusion result, which is a matrix.

[0069] Based on the fusion results, the Otsu threshold segmentation algorithm is used to distinguish between germinating and non-germinating regions. The specific implementation process is as follows:

[0070] The fusion result is normalized to ensure the data range is between 0 and 255, facilitating subsequent thresholding. The normalized fusion result is then converted to an image format, forming a fused image. The fused image is a grayscale image, where pixel values ​​represent the intensity of the fused features. The number of pixels at each grayscale level in the fused image is counted to form a grayscale histogram. All possible thresholds (from 0 to 255) are iterated over. For each threshold, the inter-class variance of the foreground (pixels greater than the threshold) and background (pixels less than or equal to the threshold) in the grayscale histogram is calculated. The threshold that maximizes the inter-class variance is selected as the optimal segmentation threshold. This threshold divides the image into two classes: germinating regions and non-germinating regions.

[0071] Based on the identification results of each germination region, the corresponding final weighted feature image is labeled to obtain multiple semantically labeled images.

[0072] S104. Traverse any pixel in the semantically labeled image and check if the label is consistent with the label of the neighboring pixels. If the label of any pixel is a germination label and there is a germination label in the neighboring pixels, then mark the pixel as a germination connected region. Use a depth-first search algorithm to recursively detect all germination connected regions in the semantically labeled image and determine the boundary of each germination connected region. Generate an initial germination boundary image based on the boundary of the germination connected region.

[0073] Obtain a preset area threshold and calculate the area of ​​each germination connected region in the initial germination boundary image; mark germination connected regions with an area smaller than the area threshold as noise regions; remove the noise regions from the initial germination boundary image to obtain the noise-removed germination boundary image.

[0074] The contour information of the germination region in the noise-removed germination boundary image is extracted; it is determined whether there are breaks in the contour information; if there are breaks in the contour information, the contour information is repaired using a morphological closing operation method to obtain a repaired germination region contour image; the area of ​​the germination region in the repaired germination region contour image is calculated, and the calculated area is compared with a preset area difference threshold; if the difference between the calculated area and the preset area difference threshold is greater than a preset adjustment threshold, the area difference threshold is adjusted according to the difference, and the germination region is re-annotated and detected using the adjusted area difference threshold to generate the final germination boundary image corresponding to the semantically annotated image, until the final germination boundary image corresponding to each semantically annotated image is obtained.

[0075] S105. Acquire any final germination boundary image and corresponding temperature and humidity sensor data. Based on a pre-established comprehensive temperature and humidity influence factor model, normalize the final germination boundary image to obtain a normalized final germination boundary image. Segment the normalized final germination boundary image, extract the pixel coordinates of the segmentation edges, and use these pixel coordinates as the position information of the segmentation edges. Use the least squares method to perform polynomial fitting on the position information of the segmentation edges to obtain a preliminary smoothed edge curve. Determine if there are curvature abrupt change points in the preliminary smoothed edge curve. If so, use Gaussian filtering to smooth the curvature abrupt change points to obtain an optimized smoothed edge curve. Based on the optimized smoothed edge curve, redraw the boundary of the germination region in the final germination boundary image to generate an initial smoothed boundary image. Extract the contours from the initial smoothed boundary image and calculate the length and area of ​​the contours to obtain the length-to-area ratio. If the length-to-area ratio is within a preset range, then the Otsu adaptive threshold is used to perform secondary optimization on the boundary of the germination region in the final germination boundary image to adjust the distribution of the segmented edges and regenerate the initial smooth boundary image until the length-to-area ratio is within a preset range, thus obtaining the final smooth boundary image corresponding to the final germination boundary image, and so on, until the final smooth boundary image corresponding to each final germination boundary image is obtained.

[0076] S106. Based on the final smooth boundary image corresponding to each final germination boundary image, each final germination boundary image is segmented to obtain multiple germination region images and multiple non-germination region images. Each germination region image is randomly rotated and scaled to obtain multiple data-enhanced germination region images. A bilinear interpolation algorithm is used to fill in regions with missing edges or discontinuous pixels in the data-enhanced germination region images to obtain multiple complete germination region images. The multiple complete germination region images and multiple non-germination region images are combined to form the target dataset.

[0077] S107. Extract the pixel distribution features of sugarcane seedlings from the target dataset; use a radial basis function support vector machine to classify the extracted pixel distribution features to obtain the classification results of the germination status of sugarcane seedlings; determine whether the classification results of the germination status meet the preset conditions and obtain the judgment result; when the judgment result is yes, obtain the final germination rate statistics of sugarcane seedlings based on multiple complete germination area images and multiple non-germination area images.

[0078] The specific implementation process for extracting the pixel distribution features of sugarcane seedlings from the target dataset is as follows:

[0079] The Otsu thresholding algorithm is used to segment the image, dividing the complete sprouting region image and multiple non-sprouting region images into foreground (green) and background, respectively. The pixel value distribution of the sprouting and non-sprouting regions is statistically analyzed, including the mean, variance, and histogram of pixel values. Texture features of the image can be calculated using methods such as the Gray-Level Co-occurrence Matrix (GLCM), i.e., pixel distribution features include both pixel value distribution and texture features.

[0080] The classification results of the germination status of sugarcane seedlings include: the distribution and proportion of green pixels. The preset condition is that the proportion of green pixels is greater than 18% or 20%. When the proportion of green pixels is greater than 18% or 20%, it is determined that the preset condition is met and the judgment result is yes.

[0081] Based on multiple complete images of germination regions and multiple images of non-germination regions, the final germination rate statistics of sugarcane seedlings were obtained. The specific calculation process is as follows:

[0082] Germination rate = the ratio between the number of images of complete germination regions and the sum of multiple images of complete germination regions and multiple images of non-germination regions × 100%.

[0083] Optionally, the final germination rate statistics can be corrected, specifically:

[0084] The absolute values ​​of the deviations between temperature data and corresponding preset temperature and humidity baselines are used as comprehensive temperature and humidity influence factors. Multiple comprehensive temperature and humidity influence factors are obtained from these factors. A weighted average method is used to calculate the normalized value of each comprehensive temperature and humidity influence factor. The difference between this normalized value and 1 is used as a correction factor. The final germination rate statistics are corrected based on this correction factor. For example, if the calculated germination rate is 80% and the corresponding correction factor is 0.9, then the corrected germination rate is 80% × 0.9 = 72%. By taking the comprehensive influence of temperature and humidity into account, the classification results of sugarcane seedling germination status are reasonably corrected to better reflect actual growth conditions.

[0085] The calculated germination rate is analyzed to assess the overall germination status of sugarcane seedlings and determine whether the germination rate has reached the expected target. The statistical results of the germination rate are then applied to sugarcane planting decisions, such as adjusting sugarcane planting density and optimizing irrigation and fertilization programs, to improve the planting efficiency of sugarcane.

[0086] If the judgment result is negative, adjust the time interval for acquiring the original images of sugarcane seedlings, reacquire multiple original images of sugarcane seedlings according to the adjusted time interval, and return to execute S101 until the judgment result is positive.

[0087] The method employs a PID control algorithm to dynamically adjust the time interval for acquiring original images of sugarcane seedlings based on environmental change information, thereby obtaining an adjusted time interval. Multiple original images of sugarcane seedlings are then acquired based on this adjusted time interval.

[0088] The environmental change information includes temperature and humidity changes. For example, when the temperature rises from 25°C to 35°C and the humidity drops from 70% to 50%, the germination of sugarcane seedlings may accelerate. The PID algorithm, through proportional, integral, and derivative parameters, adjusts the data collection interval from once per hour to once every 30 minutes to ensure timely capture of changes in germination characteristics. This dynamic adjustment effectively adapts to environmental fluctuations and improves the targeting of data collection.

[0089] The PID control algorithm is used to dynamically adjust the time interval for collecting original images of sugarcane seedlings based on environmental changes. The specific implementation process is as follows:

[0090] Based on temperature and humidity changes and the set target values, the error value is calculated, and the output value is calculated using a PID control algorithm. The time interval for acquiring raw images of sugarcane seedlings is dynamically adjusted according to the output value. For example, if the difference between the output value and the target value exceeds a preset deviation threshold, the acquisition time interval is shortened; conversely, the acquisition time interval is extended. The adjusted image acquisition time interval is applied to the actual acquisition process, and temperature and humidity changes are continuously monitored to form a closed-loop control, continuously optimizing the image acquisition time interval.

[0091] The target value can be the desired germination rate or the desired growth rate, and the corresponding output value is the output germination rate or the output growth rate.

[0092] In the above embodiments, although the steps are numbered S1, S2, etc., they are only specific embodiments given by the present invention. Those skilled in the art can adjust the execution order of S1, S2, etc. according to the actual situation, which is also within the protection scope of the present invention. It can be understood that in some embodiments, some or all of the above embodiments may be included.

[0093] As shown in Figure 2, an intelligent statistical system 200 for sugarcane germination rate according to an embodiment of the present invention includes a target dataset construction module 201, a germination status classification result acquisition module 202, a judgment module 203, and a statistical module 204.

[0094] The target dataset construction module 201 is used to: construct a target dataset based on the original images of multiple sugarcane seedlings;

[0095] The germination status classification result acquisition module 202 is used to: extract the pixel distribution features of sugarcane seedlings from the target dataset, classify the extracted pixel distribution features, and obtain the germination status classification result of sugarcane seedlings;

[0096] The judgment module 203 is used to: determine whether the germination status classification result meets the preset conditions, and obtain the judgment result;

[0097] The statistics module 204 is used to: when the judgment result is yes, obtain the final germination rate statistics of sugarcane seedlings based on the target dataset.

[0098] Optionally, in the above technical solution, the germination status classification result acquisition module 202 is further specifically used for:

[0099] A support vector machine with radial basis functions is used to classify the extracted pixel distribution features.

[0100] Optionally, in the above technical solution, the target dataset construction module 201 is specifically used for:

[0101] Original images of multiple sugarcane seedlings are collected according to a preset cycle, and temperature and humidity sensors are used to collect temperature and humidity data synchronously. The original images, temperature data, and humidity data are matched according to the collection timestamp to generate a synchronous dataset.

[0102] Construct the target dataset based on the synchronized dataset.

[0103] Optionally, the above technical solution also includes a call execution module, which is used for:

[0104] If the judgment result is negative, adjust the acquisition frequency of the original images of sugarcane seedlings, reacquire multiple original images of sugarcane seedlings according to the adjusted acquisition frequency, and re-call the target dataset construction module 201, the germination status classification result acquisition module 202, and the judgment module 203 until the judgment result is positive.

[0105] It should be noted that the beneficial effects of the intelligent sugarcane germination rate statistical system 200 provided in the above embodiments are the same as those of the intelligent sugarcane germination rate statistical method described above, and will not be repeated here. Furthermore, the system provided in the above embodiments is only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the system can be divided into different functional modules according to the actual situation to complete all or part of the functions described above. In addition, the system and method embodiments provided in the above embodiments belong to the same concept, and their specific implementation process is detailed in the method embodiments, and will not be repeated here.

[0106] The intelligent statistical system for sugarcane germination rate of the present invention can be a computer program (including program code) running on a computer device. For example, the intelligent statistical system for sugarcane germination rate of the present invention is an application software that can be used to execute the corresponding steps in the intelligent statistical method for sugarcane germination rate of the present invention.

[0107] In some embodiments, the intelligent sugarcane germination rate statistical system of the present invention can be implemented in a combination of hardware and software. As an example, the intelligent sugarcane germination rate statistical system of the present invention can be a processor in the form of a hardware decoding processor, which is programmed to execute the intelligent sugarcane germination rate statistical method of the present invention. For example, the processor in the form of a hardware decoding processor can be one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), or other electronic components.

[0108] The modules described in the embodiments of this invention can be implemented in software or hardware. The names of the modules are not, in some cases, limiting the scope of the module itself.

[0109] An electronic device according to an embodiment of the present invention includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements any of the above-mentioned intelligent statistical methods for sugarcane germination rate. That is, an electronic device according to an embodiment of the present invention may include, but is not limited to: a processor and a memory; the memory is used to store the computer program; the processor is used to execute the intelligent statistical method for sugarcane germination rate shown in any embodiment of the present invention by calling the computer program.

[0110] In one optional embodiment, an electronic device is provided, as shown in FIG3. The electronic device 4000 shown in FIG3 includes a processor 4001 and a memory 4003. The processor 4001 and the memory 4003 are connected, for example, via a bus 4002. Optionally, the electronic device 4000 may further include a transceiver 4004, which can be used for data interaction between the electronic device and other electronic devices, such as sending and / or receiving data. It should be noted that in practical applications, the transceiver 4004 is not limited to one type, and the structure of the electronic device 4000 does not constitute a limitation on the embodiments of the present invention.

[0111] Processor 4001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this invention. Processor 4001 may also be a combination that implements computational functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0112] Bus 4002 may include a pathway for transmitting information between the aforementioned components. Bus 4002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. Bus 4002 can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent bus 4002 in Figure 3, but this does not mean that there is only one bus or one type of bus.

[0113] The memory 4003 may be ROM (Read Only Memory) or other types of static storage devices capable of storing static information and instructions, RAM (Random Access Memory) or other types of dynamic storage devices capable of storing information and instructions, or EEPROM (Electrically Erasable Programmable Read Only Memory), CD-ROM (Compact Disc Read Only Memory) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto.

[0114] The memory 4003 stores the application code (computer program) for executing the present invention, and its execution is controlled by the processor 4001. The processor 4001 executes the application code stored in the memory 4003 to implement the content shown in the foregoing method embodiments.

[0115] Among them, electronic devices can also be terminal devices, which can be any device that can install applications, including at least one of smartphones, tablets, laptops, desktop computers, smart speakers, smartwatches, smart TVs, and smart in-vehicle devices.

[0116] It should be noted that the electronic device shown in Figure 3 is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0117] An embodiment of the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements any of the above-mentioned intelligent statistical methods for sugarcane germination rate.

[0118] Alternatively, the computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a compact disc read-only memory (CD-ROM), magnetic tape, a floppy disk, and an optical data storage device, etc.

[0119] In an exemplary embodiment, a computer program product or computer program is also provided, which includes computer instructions stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the electronic device to perform any of the above-described intelligent statistical methods for sugarcane germination rate.

[0120] Computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof. These programming languages ​​include object-oriented programming languages—such as Java, Smalltalk, and C++—and conventional procedural programming languages—such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0121] It should be understood that the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of methods and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0122] The computer-readable storage medium provided in this invention can be, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EEPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0123] The aforementioned computer-readable storage medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to perform the method shown in the above embodiments.

[0124] The above description is merely a preferred embodiment of the present invention and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of disclosure in this invention is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this invention.

[0125] It should be noted that the terms "first," "second," etc., used in the specification and claims of this application are used to distinguish similar objects and represent a limitation on a specific order or sequence. Where appropriate, the order of use for similar objects can be interchanged so that the embodiments of this application described herein can be implemented in an order other than that shown or described.

[0126] Those skilled in the art will recognize that this invention can be implemented as a system, method, or computer program product. Therefore, this invention can be specifically implemented in the following forms: it can be entirely hardware, entirely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software, generally referred to herein as a "circuit," "module," or "system." Furthermore, in some embodiments, this invention can also be implemented as a computer program product contained in one or more computer-readable media, which includes computer-readable program code.

[0127] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A smart statistical method for sugarcane germination rate, characterized in that, include: S1. Construct the target dataset based on the original images of multiple sugarcane seedlings; The process of constructing the target dataset includes: S101, acquiring multiple original images of sugarcane seedlings at a preset frequency, and simultaneously acquiring temperature and humidity data using temperature and humidity sensors, matching the original images, temperature data, and humidity data according to the acquisition timestamps to generate a synchronized dataset; S102, extracting pixel information from each original image of the sugarcane seedlings from the synchronized dataset, calculating the edge intensity of any original image using the boundary gradient method, and generating an initial edge feature image corresponding to the original image using the Canny operator based on the edge intensity of the original image, performing Gaussian filtering on the initial edge feature image to eliminate noise interference, and if there are blurred areas in the initial edge feature image... A nonlocal mean denoising algorithm is used for sharpening to obtain the final edge feature image corresponding to the original image, until the final edge feature image corresponding to each original image is obtained; S103, the spatial correlation matrix is ​​calculated based on the gray-level co-occurrence matrix. Based on the spatial correlation matrix of any final edge feature image, the dependency relationship between each pixel of the final edge feature image and its surrounding pixels is obtained, and the self-attention weight corresponding to the final edge feature image is generated; the self-attention weight is multiplied by the pixel value of each pixel in the edge feature image to obtain an attention mask; the attention mask is normalized to generate a two-dimensional weight map; the two-dimensional weight map is weighted and fused with the edge feature image to obtain the final edge feature image. The initial weighted feature image corresponding to the image is used. If the self-attention weight of any region in the initial weighted feature image is lower than the preset weight threshold, the nonlocal mean algorithm is used to enhance the local features of the region to obtain the enhanced weighted feature image, which is recorded as the final weighted feature image. This process is repeated until the final weighted feature image corresponding to each final edge feature image is obtained. S104: Iterate through the pixels in any semantically labeled image and check whether the labels of the pixels are consistent with those of the neighboring pixels. If the label of any pixel is a germination label and there is a germination label in the neighboring pixels, then the pixel is marked as a germination connected region. A depth-first search algorithm is used to recursively detect all germination connected regions in the semantically labeled image and determine the boundaries of each germination connected region. Based on the boundaries of the germination connected regions, an initial germination boundary image is generated; a preset area threshold is obtained, and the area of ​​each germination connected region in the initial germination boundary image is calculated; germination connected regions with areas smaller than the area threshold are marked as noise regions; noise regions in the initial germination boundary image are removed to obtain a noise-removed germination boundary image; the contour information of the germination regions in the noise-removed germination boundary image is extracted; it is determined whether there are breaks in the contour information; if there are breaks in the contour information, the contour information is repaired using a morphological closing operation method to obtain a repaired germination region contour image; the area of ​​the germination region in the repaired germination region contour image is calculated, and the calculated area is compared with a preset area difference threshold.If the difference between the calculated area and the preset area difference threshold is greater than the preset adjustment threshold, the area difference threshold is adjusted according to the difference, and the germination region is re-annotated and detected using the adjusted area difference threshold to generate the final germination boundary image corresponding to the semantically annotated image, until the final germination boundary image corresponding to each semantically annotated image is obtained; S105, acquire any final germination boundary image and the corresponding temperature and humidity sensor data, and normalize the final germination boundary image according to the pre-established temperature and humidity comprehensive influence factor to obtain the normalized final germination boundary image. The normalized final germination boundary image is then segmented, and the pixel coordinates of the segmentation edges are extracted and the image of the segmentation edges is... Using primitive coordinates as the positional information of the segmentation edges, a least-squares method is used to perform polynomial fitting on the positional information of the segmentation edges to obtain a preliminary smooth edge curve. The presence of curvature abrupt changes in the preliminary smooth edge curve is then determined. If such abrupt changes exist, Gaussian filtering is used to smooth them, resulting in an optimized smooth edge curve. Based on the optimized smooth edge curve, the boundaries of the germination region in the final germination boundary image are redrawn, generating an initial smooth boundary image. The contours in the initial smooth boundary image are extracted, and their length and area are calculated to obtain the length-to-area ratio. It is then determined whether the length-to-area ratio is within a preset range. If not, an Otsu adaptive threshold is applied to the edges of the germination region in the final germination boundary image. The boundary is then optimized a second time to adjust the distribution of the segmented edges and regenerate the initial smooth boundary image until the length-to-area ratio is within a preset range, thus obtaining the final smooth boundary image corresponding to the final germination boundary image. This process is repeated until the final smooth boundary image corresponding to each final germination boundary image is obtained. The absolute values ​​of the deviations between the temperature data and the corresponding preset temperature reference value, and the absolute values ​​of the deviations between the humidity data and the corresponding preset humidity reference value are used as comprehensive temperature and humidity influence factors. S106: Based on the final smooth boundary image corresponding to each final germination boundary image, each final germination boundary image is segmented to obtain multiple germination region images and multiple non-germination region images. Each germination region image is then randomly rotated and scaled. The process involves: S1) Data augmentation to obtain multiple data-enhanced germination region images; S2) Using bilinear interpolation to fill in areas with missing edges or discontinuous pixels in the data-enhanced germination region images to obtain multiple complete germination region images; S3) Combining these complete germination region images with multiple non-germination region images to form a target dataset; S4) Extracting pixel distribution features from the target dataset and classifying these features to obtain a classification result of the germination status of the sugarcane seedlings; S5) Determining whether the classification result of the germination status meets preset conditions to obtain a judgment result; S6) When the judgment result is yes, obtaining the final germination rate statistics of the sugarcane seedlings based on the target dataset.The pixel distribution features include pixel value distribution and texture features. The classification results of sugarcane seedling germination status include the distribution and proportion of green pixels. The preset condition is that the proportion of green pixels is greater than 18% or 20%. When the proportion of green pixels is greater than 18% or 20%, it is determined that the preset condition is met, and the judgment result is yes.

2. The intelligent statistical method for sugarcane germination rate according to claim 1, characterized in that, The extracted pixel distribution features are classified, including: using a support vector machine with radial basis functions to classify the extracted pixel distribution features.

3. The intelligent statistical method for sugarcane germination rate according to claim 1 or 2, characterized in that, Based on multiple original images of sugarcane seedlings, a target dataset is constructed, including: collecting original images of multiple sugarcane seedlings at a preset period, and simultaneously collecting temperature and humidity data using temperature and humidity sensors, and matching the original images, temperature data, and humidity data according to the collection timestamps to generate a synchronized dataset; and constructing the target dataset based on the synchronized dataset.

4. The intelligent statistical method for sugarcane germination rate according to claim 1 or 2, characterized in that, Also includes: When the judgment result is negative, the acquisition frequency of the original images of sugarcane seedlings is adjusted, and multiple original images of sugarcane seedlings are reacquired according to the adjusted acquisition frequency. The process returns to S1 until the judgment result is positive. Specifically, a PID control algorithm is used to dynamically adjust the time interval for acquiring the original images of sugarcane seedlings based on environmental change information, resulting in an adjusted time interval. Multiple original images of sugarcane seedlings are then acquired according to this adjusted time interval. The specific implementation process of using a PID control algorithm to dynamically adjust the time interval for acquiring the original images of sugarcane seedlings based on environmental change information is as follows: based on temperature and humidity changes... The system calculates the error value based on the set target value and uses a PID control algorithm to calculate the output value. Based on the output value, it dynamically adjusts the time interval for acquiring original images of sugarcane seedlings. Specifically, if the difference between the output value and the target value exceeds a preset deviation threshold, the acquisition time interval is shortened; otherwise, the acquisition time interval is extended. The adjusted image acquisition time interval is applied to the actual acquisition process, and temperature and humidity changes are continuously monitored to form a closed-loop control, continuously optimizing the image acquisition time interval. The target value is specifically the desired germination rate or desired growth rate, and the output value is the output germination rate or output growth rate.

5. A smart statistical system for sugarcane germination rate, characterized in that, It includes a target dataset construction module, a germination status classification result acquisition module, a judgment module, and a statistics module; The target dataset construction module is used to: construct a target dataset based on the original images of multiple sugarcane seedlings; The process of constructing the target dataset includes: acquiring multiple original images of sugarcane seedlings at a preset frequency, and simultaneously acquiring temperature and humidity data using temperature and humidity sensors. The original images, temperature data, and humidity data are then matched according to the acquisition timestamps to generate a synchronized dataset. Pixel information of each original image of the sugarcane seedlings is extracted from the synchronized dataset. The edge intensity of any given original image is calculated using the boundary gradient method. Based on the edge intensity, the Canny operator is used to generate an initial edge feature image corresponding to that original image. This initial edge feature image is then subjected to Gaussian filtering to eliminate noise interference. If blurred regions exist in the initial edge feature image, a nonlocal filtering method is used. The mean denoising algorithm is used to sharpen the image, resulting in the final edge feature image corresponding to the original image. This process is repeated for each original image to obtain its corresponding final edge feature image. A spatial correlation matrix is ​​calculated based on the gray-level co-occurrence matrix. Using this spatial correlation matrix, the dependency relationship between each pixel of the final edge feature image and its surrounding pixels is obtained, generating self-attention weights for that final edge feature image. These self-attention weights are multiplied by the pixel value of each pixel in the edge feature image to obtain an attention mask. This attention mask is then normalized to generate a two-dimensional weight map. Finally, this two-dimensional weight map is weighted and fused with the edge feature image to obtain the final edge feature image. An initial weighted feature image is generated. If the self-attention weight of any region in the initial weighted feature image is lower than a preset weight threshold, a nonlocal mean algorithm is used to enhance the local features of that region, resulting in an enhanced weighted feature image, which is recorded as the final weighted feature image. This process is repeated until the final weighted feature image corresponding to each final edge feature image is obtained. The labels of pixels in any semantically labeled image are checked against those of their neighboring pixels. If a pixel's label is a germination label and there are germination labels among its neighboring pixels, then that pixel is marked as a germination connected region. A depth-first search algorithm is used to recursively detect all germination connected regions in the semantically labeled image to determine the boundaries of each germination connected region. Based on the germination connectivity... The boundary of the region is defined to generate an initial germination boundary image; a preset area threshold is obtained, and the area of ​​each germination connected region in the initial germination boundary image is calculated; germination connected regions with an area smaller than the area threshold are marked as noise regions; noise regions in the initial germination boundary image are removed to obtain a noise-removed germination boundary image; the contour information of the germination region in the noise-removed germination boundary image is extracted; whether there are breaks in the contour information is determined; if there are breaks in the contour information, the contour information is repaired using a morphological closing operation method to obtain a repaired germination region contour image; the area of ​​the germination region in the repaired germination region contour image is calculated, and the calculated area is compared with a preset area difference threshold;If the difference between the calculated area and the preset area difference threshold is greater than the preset adjustment threshold, the area difference threshold is adjusted according to the difference. The germination region is then re-annotated and detected using the adjusted area difference threshold, generating the final germination boundary image corresponding to the semantically annotated image. This process is repeated until the final germination boundary image for each semantically annotated image is obtained. Any final germination boundary image and its corresponding temperature and humidity sensor data are acquired. Based on a pre-established comprehensive temperature and humidity influence factor, the final germination boundary image is normalized to obtain a normalized final germination boundary image. This normalized final germination boundary image is then segmented, and the pixel coordinates of the segmentation edges are extracted. These pixel coordinates are then used as the segmentation edges. The location information of the segmentation edges is obtained by performing polynomial fitting using the least squares method to obtain a preliminary smooth edge curve. The presence of curvature abrupt changes in the preliminary smooth edge curve is then assessed. If such abrupt changes exist, Gaussian filtering is used to smooth them, resulting in an optimized smooth edge curve. Based on this optimized smooth edge curve, the boundary of the germination region in the final germination boundary image is redrawn, generating an initial smooth boundary image. The contours in the initial smooth boundary image are extracted, and their length and area are calculated to obtain the length-to-area ratio. It is then determined whether the length-to-area ratio is within a preset range. If not, Otsu adaptive thresholding is used to perform secondary optimization on the boundary of the germination region in the final germination boundary image to adjust the segmentation edges. The distribution of the edge is used to regenerate the initial smooth boundary image until the length-to-area ratio is within a preset range, thus obtaining the final smooth boundary image corresponding to the final germination boundary image. This process is repeated until the final smooth boundary image corresponding to each final germination boundary image is obtained. The absolute values ​​of the deviations between temperature data and corresponding preset temperature and humidity reference values ​​are used as the comprehensive temperature and humidity influence factors. Based on the final smooth boundary image corresponding to each final germination boundary image, each final germination boundary image is segmented to obtain multiple germination region images and multiple non-germination region images. Each germination region image is then randomly rotated and scaled to obtain multiple data-enhanced germination region images. The bilinear interpolation algorithm is used to fill in areas with missing edges or discontinuous pixels in the data-enhanced germination region images, resulting in multiple complete germination region images. These multiple complete germination region images and multiple non-germination region images are combined to form the target dataset. The germination state classification result acquisition module is used to: extract the pixel distribution features of sugarcane seedlings from the target dataset, classify the extracted pixel distribution features, and obtain the germination state classification result of the sugarcane seedlings. The judgment module is used to: determine whether the germination state classification result meets preset conditions and obtain a judgment result. The statistics module is used to: when the judgment result is yes, obtain the final germination rate statistics result of the sugarcane seedlings based on the target dataset.The pixel distribution features include pixel value distribution and texture features. The classification results of sugarcane seedling germination status include the distribution and proportion of green pixels. The preset condition is that the proportion of green pixels is greater than 18% or 20%. When the proportion of green pixels is greater than 18% or 20%, it is determined that the preset condition is met, and the judgment result is yes.

6. The intelligent statistical system for sugarcane germination rate according to claim 5, characterized in that, The germination state classification result acquisition module is also specifically used to classify the extracted pixel distribution features using a support vector machine with radial basis functions.

7. A sugarcane germination rate intelligent statistical system according to claim 5 or 6, characterized in that, The target dataset construction module is specifically used to: collect original images of multiple sugarcane seedlings according to a preset cycle, and enable temperature and humidity sensors to collect temperature and humidity data synchronously, and match the original images, temperature data and humidity data according to the collection timestamp to generate a synchronous dataset; Based on the synchronized dataset, construct the target dataset.

8. A smart statistical system for sugarcane germination rate according to claim 5 or 6, characterized in that, The system also includes a call execution module, which is used to: when the judgment result is negative, adjust the acquisition frequency of the original images of sugarcane seedlings, reacquire multiple original images of sugarcane seedlings according to the adjusted acquisition frequency, and re-call the target dataset construction module, the germination status classification result acquisition module, and the judgment module until the judgment result is positive; wherein, a PID control algorithm is used to dynamically adjust the time interval for acquiring the original images of sugarcane seedlings according to environmental change information to obtain an adjusted time interval, and acquire multiple original images of sugarcane seedlings according to the adjusted time interval; the PID control algorithm is used to dynamically adjust the acquisition frequency of the original images of sugarcane seedlings according to environmental change information. The specific implementation process is as follows: Based on the temperature change information, humidity change information, and the set target value, the error value is calculated, and the output value is calculated using the PID control algorithm. Based on the output value, the time interval for acquiring the original images of sugarcane seedlings is dynamically adjusted. Specifically, if the difference between the output value and the target value exceeds the preset deviation threshold, the acquisition time interval is shortened; otherwise, the acquisition time interval is extended. The adjusted image acquisition time interval is applied to the actual acquisition process, and temperature and humidity changes are continuously monitored to form a closed-loop control, continuously optimizing the image acquisition time interval. The target value is specifically the desired germination rate or desired growth rate, and the output value is the output germination rate or output growth rate.

9. An electronic device, characterized in that, The method includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the intelligent statistical method for sugarcane germination rate as described in any one of claims 1 to 4.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the intelligent statistical method for sugarcane germination rate as described in any one of claims 1 to 4.

Images