Rice disease identification method and system based on aerial image of planting area

Abstract

The invention relates to the field of rice disease identification, in particular to a rice disease identification method and system based on aerial images of a planting area. The method comprises the following steps: acquiring a plurality of preprocessed rice region images according to a preset path, and constructing a rice map sequence; segmenting any rice region image by using a watershed algorithm to obtain a plurality of sub-regions, and constructing multi-dimensional feature vectors of the sub-regions; the distance between any two sub-regions is calculated according to the feature vectors of the sub-regions, the sub-regions are clustered according to the distance to obtain clustering clusters, the disease degree of the rice region image is calculated according to the clustering clusters, and the clustering clusters comprise normal clustering clusters and disease clustering clusters; and traversing to obtain the disease degree of each rice region image in the rice map sequence so as to calculate the disease damage degree of the planting region and complete disease identification. Through the technical scheme of the invention, the accuracy of the rice disease condition identification result can be improved.

Description

Technical Field

[0001] The present invention relates to the field of rice disease identification, and more specifically, to a method and system for identifying rice disease based on aerial images of a planting area. Background Art

[0002] As one of the most important food crops in the world, the growth, development, yield and quality of rice are easily affected by pests and diseases. Common rice diseases such as rice blast, bacterial blight, brown spot and false smut can lead to reduced yield and quality, and even large-scale production reduction in severe cases. Traditional disease identification methods usually rely on manual inspections, but this method is not only time-consuming and labor-intensive, but also prone to missed or misdiagnosed cases. Therefore, how to use modern technology to improve the efficiency and accuracy of rice disease identification has become an important topic in current agricultural research and practice.

[0003] Remote sensing technology, especially aerial images based on drones or satellites, has been frequently used in the agricultural field, mainly for pest and disease monitoring, crop growth assessment and soil quality analysis. Aerial photography technology can quickly and comprehensively monitor large areas of farmland and obtain important information about crop growth, pest and disease distribution, soil conditions, etc. This technology provides efficient data support for agricultural management, helping farmers and agricultural experts better understand the real-time situation of farmland and make scientific decisions.

[0004] The existing Chinese patent application document with publication number CN108563979A discloses a method for distinguishing rice blast disease based on aerial farmland images. The application document uses the maximum inter-class variance method to segment the color component combined image based on color features to obtain rice disease spots. However, there may be a situation where the RGB value of the rice disease is close to the RGB value of rice. Segmenting the image only by using the color image component combination method will cause the problem of inaccurate segmentation results. Summary of the invention

[0005] In order to solve the problem of inaccurate segmentation results, the present invention proposes a rice disease identification method and system based on aerial images of planting areas.

[0006] In a first aspect, the present invention discloses a rice disease identification method based on aerial images of a planting area, comprising: obtaining a plurality of pre-processed rice area images according to a preset path, and constructing a rice image sequence; using a watershed algorithm to segment any rice area image to obtain a plurality of sub-areas, constructing a multi-dimensional feature vector of the sub-area, wherein one dimension in the feature vector corresponds to one water level line; calculating the distance between any two sub-areas according to the feature vector of the sub-area, clustering the sub-areas according to the distance to obtain cluster clusters, and calculating the disease degree of the rice area image according to the cluster clusters, wherein the cluster clusters include normal cluster clusters and disease cluster clusters; traversing to obtain the disease degree of each rice area image in the rice image sequence, so as to calculate the disease hazard degree of the planting area, and completing disease identification; wherein the disease degree satisfies the relationship: , Indicates the severity of the disease, Indicates disease clusters The area of represents the area of ​​a normal cluster, Indicates disease clusters The ratio of the number of neutron regions to the total number of sub-regions, represents the normalization function, Represents a logarithmic function.

[0007] The disease severity relationship is calculated by calculating the ratio of the area of ​​the disease cluster to the area of ​​the normal area, which can reflect the comparison between the overall range of rice disease and the healthy area, and intuitively reflect the spread of the disease. The entropy value is used to take into account the degree of aggregation and chaos of different disease areas. The larger the entropy value, the more types of diseases there are and the greater the severity of the disease.

[0008] Preferably, the preprocessing includes: obtaining a planting area image according to a preset path, and extracting a rice area image from the planting area image using an MTCNN model; enhancing the contrast and brightness of the rice area image using histogram equalization and contrast stretching, and denoising the rice area image using Gaussian filtering.

[0009] The MTCNN model first accurately extracts the rice area in the planting area image to ensure that subsequent processing focuses on the key parts, thereby reducing interference from irrelevant areas. Then, histogram equalization and contrast stretching techniques are used to enhance the brightness and contrast of the rice area image, making the details clearer, especially in uneven lighting or low contrast conditions, which can better highlight the texture and disease characteristics of rice. At the same time, Gaussian filtering is used to remove noise in the image, reducing the impact of environmental interference on image analysis, making subsequent disease identification and analysis more accurate.

[0010] Preferably, the step of extracting the rice area image from the planting area image further comprises: performing threshold segmentation on the planting area image using G channel information to obtain the rice area image.

[0011] Rice leaves usually have a higher reflectivity in the green band. Compared with other channels, the G channel can better highlight the characteristics of rice vegetation. By extracting the G channel image, the distribution of rice areas can be more accurately reflected.

[0012] Preferably, the feature vector is composed of the sub-region area and the number of edge pixels in the sub-region or is composed of the number of edge pixels in the sub-region alone.

[0013] Preferably, the distance includes: calculating the distance between any two sub-regions using dynamic time warping based on feature vectors of the sub-regions.

[0014] Preferably, the distance also satisfies the relationship: , Indicates sub-area and subregions The distance , Indicates sub-area The eigenvector and subregion of The covariance of the eigenvectors of Indicates sub-area The standard deviation of the eigenvectors, Indicates sub-area The standard deviation of the eigenvectors, Represents an exponential function.

[0015] The covariance measures the linear relationship between the feature vectors of two sub-regions, while the standard deviation characterizes the dispersion of the features of each sub-region, further adjusting the scale of the similarity measurement. This distance measurement method can not only reveal the relationship between different diseased areas or healthy areas in the rice area, but also effectively distinguish the distribution pattern and evolution trend of the disease during image analysis.

[0016] Preferably, the calculation of the disease hazard degree of the planting area includes: setting the rice region images whose disease degree is not less than a preset disease threshold in the rice image sequence to 1, setting the rice region images whose disease degree is less than the preset disease threshold in the rice image sequence to 0, converting the rice image sequence into a binary sequence, and counting the number of consecutive occurrences of 1; the disease hazard degree satisfies the relationship: , Indicates the severity of the disease. represents the mean disease severity of the rice area images in the rice image sequence, represents the variance of the number of consecutive occurrences of 1, Represents the normalization function.

[0017] The binary sequence reflects the persistence and spread of the disease. The distribution and volatility of the disease are comprehensively evaluated by calculating the mean of the disease severity and the variance of the number of occurrences of consecutive severe areas.

[0018] In a second aspect, the present invention discloses a rice disease identification system based on aerial images of planting areas, comprising: a processor; and a memory, wherein the memory stores computer instructions, and when the computer instructions are executed by the processor, the system executes the above-mentioned rice disease identification method based on aerial images of planting areas.

[0019] Beneficial effects of the present invention: The present invention first optimizes the image quality through preprocessing technology, extracts the rice area and enhances its contrast and brightness, making the disease more obvious. Then, the watershed algorithm is used to finely segment the rice image, refine the rice area into multiple sub-areas, calculate its multi-dimensional feature vector, and identify the difference between the disease clusters and normal areas through dynamic time warping and distance calculation methods, so as to accurately evaluate the disease severity of each area. Finally, by statistically analyzing the disease severity and the changing trend in the image sequence, the disease hazard level of the entire planting area can be quantified.

[0020] The present invention can efficiently and accurately identify disease conditions in a large area of ​​rice planting areas, avoids the hysteresis of manual detection, and can effectively improve the efficiency of disease and insect pest management in rice planting. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein: Figure 1 It is a flow chart of a method for identifying rice disease based on aerial images of a planting area according to an embodiment of the present invention. DETAILED DESCRIPTION

[0022] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments 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 those skilled in the art without creative work are within the scope of protection of the present invention.

[0023] It should be understood that when the terms "first", "second", etc. are used in the claims, descriptions, and drawings of the present invention, they are only used to distinguish different objects, rather than to describe a specific order. The terms "include" and "comprise" used in the description and claims of the present invention indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their collections.

[0024] The present invention provides a method for identifying rice disease based on aerial images of planting areas. Figure 1 As shown, the rice disease identification method based on aerial images of the planting area includes steps S1 to S4, which are described in detail below.

[0025] S1, obtaining a number of pre-processed rice area images according to a preset path, and constructing a rice image sequence.

[0026] In one embodiment, a high-resolution image of the planting area is acquired by a drone according to a preset flight path, and the image is processed using an MTCNN (Multi-task Cascaded Convolutional Networks) model to automatically identify and extract multiple rice area images in the image.

[0027] For each rice area image, the histogram equalization technique is first applied to improve the brightness distribution of the image, thereby enhancing the image details; then, the contrast stretching method is used to further enhance the contrast of the image, making the features of the rice area more prominent. Next, in order to eliminate the noise in the rice area image, the Gaussian filter algorithm is used to smooth the image, reduce the noise interference in the details, and improve the image quality. After these preprocessing steps, high-quality and clear rice area images are obtained. Finally, according to the preset flight path of the drone, these processed rice area images are organized into a rice image sequence in chronological or spatial order.

[0028] In one embodiment, extracting the rice area image from the planting area image further includes: performing threshold segmentation on the planting area image using G channel information to obtain the rice area image.

[0029] It should be noted that, first, the G channel (i.e., green channel) of the planting area image is extracted, because rice leaves usually have a higher reflectivity in the green band, and compared with other channels (such as the red channel and the blue channel), the G channel can better highlight the characteristics of rice vegetation. By extracting the G channel image, the distribution of rice areas can be more accurately reflected.

[0030] Next, select a suitable threshold, which can usually be performed by manually setting the threshold or using an automated thresholding method. The purpose of threshold segmentation is to distinguish the rice area from the background area in the image. Specifically, in the G channel image, the rice area usually has a higher pixel value, while the background area (such as soil, open space, etc.) has a lower pixel value. By setting a threshold, the area with a pixel value higher than the threshold is marked as the rice area, and the area below the threshold is marked as the background area.

[0031] After completing the threshold segmentation, a binary rice region image can be obtained, in which the rice region is usually represented by white and the background region is represented by black. In order to further refine the extraction of the rice region, morphological operations can be used to remove noise, smooth the boundaries, and improve the accuracy of the segmentation results.

[0032] S2, using the watershed algorithm to segment any rice area image to obtain several sub-regions, and construct a multi-dimensional feature vector of the sub-region.

[0033] It should be noted that when analyzing the rice area, the rice area is first accurately segmented through image processing technology to extract key parts such as leaves, because the disease of rice is usually manifested as abnormality of the leaves. Rice diseases often cause visible changes on the leaves, such as insect bite holes, spots, and browning on the leaves. These diseased areas often have different colors, textures, or morphological characteristics from healthy leaves. Therefore, image segmentation can not only help identify healthy areas of rice, but also accurately distinguish damaged leaf areas. After obtaining different areas through segmentation, the size, distribution, and type of lesions of these diseased areas can be further analyzed to assess the severity of the rice disease.

[0034] The watershed algorithm divides the image into multiple different sub-regions by simulating the process of water erosion in the terrain, each of which represents an area with similar visual features. On this basis, in order to more comprehensively analyze the characteristics of the rice area, a multi-dimensional feature vector can be constructed for each sub-region, where each dimension corresponds to the "water level line" in the image - that is, in the watershed algorithm, different stages of water level rise will gradually merge adjacent areas and eventually form independent segmented areas. Each dimension reflects different levels of image features, such as color, texture, shape, edge information, etc. Specifically, these "water levels" can help capture different segmentation levels in the image, thereby expressing the various features of the region as numerical vectors.

[0035] In one embodiment, the feature vector is composed of the sub-region area and the number of edge pixels in the sub-region or is composed of the number of edge pixels in the sub-region alone.

[0036] When the feature vector is composed of the number of edge pixels in the sub-region, it is 1 times The matrix of Represents the number of dimensions. The value of each dimension in the feature vector represents the water level. When , the number of edge pixels in the sub-region.

[0037] When the feature vector is composed of the sub-region area and the number of edge pixels of the sub-region, it is 2 times The matrix of Represents the number of dimensions. The value of each dimension in the feature vector represents the water level. When , the number of edge pixels in the sub-region and the area of ​​the sub-region.

[0038] S3, calculating the distance between any two sub-regions according to the feature vectors of the sub-regions, clustering the sub-regions according to the distances to obtain clusters, and calculating the disease severity of the rice region image according to the clusters.

[0039] In one embodiment, dynamic time warping is used to calculate the distance between any two sub-regions based on the feature vectors of the sub-regions.

[0040] The sub-regions are clustered according to the distance to obtain clusters, which include normal clusters and disease clusters. To distinguish normal clusters, a cluster of all disease-free clusters can be added as a template cluster, and the pixel points between each cluster and the template cluster are calculated to distinguish normal clusters.

[0041] The disease severity of the rice area image is calculated based on the clustering clusters, and the disease severity satisfies the relationship: , Indicates the severity of the disease, Indicates disease clusters The area of represents the area of ​​a normal cluster, Indicates disease clusters The ratio of the number of neutron regions to the total number of sub-regions, represents the normalization function, Represents a logarithmic function.

[0042] By calculating the ratio of the area of ​​the disease clusters to the area of ​​the normal area (leaf), the comparison between the overall disease range of rice and the healthy area can be reflected, and the degree of disease spread can be intuitively reflected. The overall value is the entropy value. Taking into account the degree of aggregation and chaos of different disease areas, the larger the entropy value, the more types of diseases there are and the greater the severity of the disease.

[0043] In one embodiment, the distance also satisfies the relationship: , Indicates sub-area and subregions The distance , Indicates sub-area The eigenvector and subregion of The covariance of the eigenvectors of Indicates sub-area The standard deviation of the eigenvectors, Indicates sub-area The standard deviation of the eigenvectors, Represents an exponential function.

[0044] The covariance measures the linear relationship between the feature vectors of two sub-regions, while the standard deviation characterizes the dispersion of the features of each sub-region, further adjusting the scale of the similarity measurement. By using an exponential function to transform these quantities, the distance between sub-regions with higher similarity can be shorter, while the distance between sub-regions with lower similarity can be farther, thus providing a smooth and flexible similarity measurement method. This distance measurement method can not only reveal the relationship between different diseased areas or healthy areas in the rice area, but also effectively distinguish the distribution pattern and evolution trend of the disease during image analysis.

[0045] S4, traverse and obtain the disease severity of each rice area image in the rice image sequence to calculate the disease hazard degree of the planting area and complete disease identification.

[0046] It should be noted that the degree of disease damage in the planting area is not only related to the single disease degree in each image, but also closely related to the spatial distribution of the disease degree. When the areas with more serious disease degrees are concentrated in some adjacent rice area images, it usually indicates that the disease is limited to certain local areas of the planting area, and the spread and diffusion of the disease have not yet formed a large-scale impact. At this time, the prevention and control measures may be concentrated in these specific areas to effectively control the development of the disease. However, if the areas with serious disease appear in the rice area images of different positions, it means that the disease not only occurs in certain local areas, but has also spread to multiple places in the planting area, indicating that the disease is spreading and the difficulty of control has greatly increased. At this time, the disease in the entire planting area has begun to show a diffusive trend, which means that the disease is spreading faster and may cross multiple areas. Prevention and control measures need to be more comprehensive and timely to avoid further spread of the disease and cause greater losses.

[0047] In one embodiment, the rice area images in the rice image sequence whose disease severity is not less than a preset disease threshold are set to 1, and the rice area images in the rice image sequence whose disease severity is less than the preset disease threshold are set to 0. The rice image sequence is converted into a binary sequence, and the number of consecutive occurrences of 1 is counted.

[0048] The degree of disease damage satisfies the relationship: , Indicates the severity of the disease. represents the mean disease severity of the rice area images in the rice image sequence, represents the variance of the number of consecutive occurrences of 1, Represents the normalization function.

[0049] For example, the binary sequence of statistics is , then the number of times 1 appears consecutively is 2, 1, and 3. The variance in the formula is the variance of 2, 1, and 3.

[0050] The degree of harm of the disease is compared with the danger threshold, and in response to the degree of harm of the disease being not less than the danger threshold, an alarm signal is generated and sent to complete the disease identification.

[0051] An embodiment of the present invention also discloses a rice disease identification system based on aerial images of planting areas, including a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a rice disease identification method based on aerial images of planting areas according to the present invention is implemented.

[0052] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface, and their configuration and functions are known in the art, so they will not be described in detail here.

[0053] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, apparatus or device. For example, a computer-readable storage medium may be any appropriate magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory RRAM (Resistive Random Access Memory), a dynamic random access memory DRAM (Dynamic Random Access Memory), a static random access memory SRAM (Static Random-Access Memory), an enhanced dynamic random access memory EDRAM (Enhanced Dynamic Random Access Memory), a high-bandwidth memory HBM (High-Bandwidth Memory), a hybrid memory cube HMC (Hybrid Memory Cube), etc., or any other medium that can be used to store the required information and can be accessed by an application, a module or both. Any such computer storage medium may be part of a device or accessible or connectable to a device.

[0054] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

[0055] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A rice disease identification method based on aerial images of planting areas, characterized in that: include: Acquire a number of pre-processed rice area images according to a preset path, and construct a rice image sequence; The watershed algorithm is used to segment any rice area image to obtain several sub-areas, and a multi-dimensional feature vector of the sub-area is constructed, where one dimension in the feature vector corresponds to one water level line. The distance between any two sub-regions is calculated according to the feature vectors of the sub-regions, the sub-regions are clustered according to the distance to obtain clusters, and the disease severity of the rice region image is calculated according to the clusters, and the clusters include normal clusters and disease clusters; Traverse and obtain the disease severity of each rice region image in the rice image sequence to calculate the disease severity of the planting area and complete disease identification; Among them, the severity of the disease satisfies the relationship: , Indicates the severity of the disease, Indicates disease clusters The area of represents the area of ​​a normal cluster, Indicates disease clusters The ratio of the number of neutron regions to the total number of sub-regions, represents the normalization function, Represents a logarithmic function.

2. The rice disease identification method based on aerial images of planting areas according to claim 1 is characterized in that: The pre-processing comprises: Obtain the planting area image according to the preset path, and use the MTCNN model to extract the rice area image in the planting area image; Histogram equalization and contrast stretching are used to enhance the contrast and brightness of rice area images, and Gaussian filtering is used to denoise the rice area images.

3. The rice disease identification method based on aerial images of planting areas according to claim 2 is characterized in that: The step of extracting the rice area image from the planting area image further comprises: The G channel information is used to perform threshold segmentation on the planting area image to obtain the rice area image.

4. The rice disease identification method based on aerial images of planting areas according to claim 1 is characterized in that: The feature vector is composed of the sub-region area and the number of edge pixels in the sub-region or is composed of the number of edge pixels in the sub-region alone.

5. The rice disease identification method based on aerial images of planting areas according to claim 1 is characterized in that: The distances include: The distance between any two sub-regions is calculated using dynamic time warping based on the feature vectors of the sub-regions.

6. The rice disease identification method based on aerial images of planting areas according to claim 1 is characterized in that: The distance also satisfies the relationship: , Indicates sub-area and subregions The distance , Indicates sub-area The eigenvector and subregion of The covariance of the eigenvectors of Indicates sub-area The standard deviation of the eigenvectors, Indicates sub-area The standard deviation of the eigenvectors, Represents an exponential function.

7. The rice disease identification method based on aerial images of planting areas according to claim 1 is characterized in that: The calculation of the degree of disease damage in the planting area includes: The rice region images whose disease severity is not less than the preset disease severity threshold in the rice image sequence are set to 1, and the rice region images whose disease severity is less than the preset disease severity threshold in the rice image sequence are set to 0, and the rice image sequence is converted into a binary sequence, and the number of consecutive occurrences of 1 is counted; The degree of disease damage satisfies the relationship: , Indicates the severity of the disease. represents the mean disease severity of the rice area images in the rice image sequence, represents the variance of the number of consecutive occurrences of 1, Represents the normalization function.

8. A rice disease identification system based on aerial images of planting areas, characterized in that: include: Processor; and A memory storing computer instructions, wherein when the computer instructions are executed by a processor, the system executes the rice disease identification method based on aerial images of a planting area according to any one of claims 1 to 7.

Images