Method for identifying plant diseases and insect pests on crops

Abstract

The invention belongs to the technical field of crop disease and insect pest recognition, and particularly relates to a crop disease and insect pest recognition method which comprises the steps of 1, collecting disease and insect pest pictures; 2, preprocessing the pest and disease damage picture; step 3, training a multi-channel network model composed of a convolutional neural network and a capsule network by using the disease and insect pest picture to obtain a disease and insect pest recognition model; and 4, collecting crop pictures, and carrying out disease and pest identification by using a disease and pest identification model. The method comprises the following steps: collecting and preprocessing disease and insect pest pictures, training a multi-channel network model composed of a convolutional neural network and a capsule network to obtain a disease and insect pest recognition model, and further performing disease and insect pest recognition on crop pictures to obtain whether diseases and insect pests exist on crops or not. The plant diseases and insect pests on crops can be timely found and recognized in a large range, the plant diseases and insect pests are treated, and serious influences on crop growth are avoided.

Description

Technical Field

[0001] The present application belongs to the technical field of crop disease and insect pest identification, and specifically relates to a crop disease and insect pest identification method. Background Art

[0002] Crops often suffer from diseases and pests, such as powdery mildew, ergot, leaf rust, etc., which seriously affect the growth of crops. Therefore, when diseases and pests occur on crops, they need to be controlled in a timely manner.

[0003] To control the diseases and pests on crops, we first need to identify the diseases and pests on the crops. Only after we have identified the diseases and pests on the crops can we carry out targeted control.

[0004] At present, the identification of crop diseases and insect pests mostly relies on manual inspections. This method consumes a lot of manpower and is prone to omissions. It is difficult to ensure that crop diseases and insect pests can be discovered and identified in a timely manner over a large area, which may lead to missed opportunities for control and seriously affect crop growth.

[0005] This application is proposed in view of the above-mentioned technical defects.

[0006] It should be noted that the disclosure of the above background technology content is only used to assist in understanding the inventive concept and technical solution of the present invention, and it does not necessarily belong to the prior art of this patent application. In the absence of clear evidence that the above content has been disclosed on the filing date of this application, the above background technology should not be used to evaluate the novelty and creativity of the present application. Summary of the invention

[0007] The purpose of the present application is to provide a method for identifying pests and diseases on crops to overcome or alleviate at least one of the known technical deficiencies.

[0008] The technical solution of this application is:

[0009] A method for identifying plant diseases and insect pests on crops, comprising:

[0010] Step 1: Collect pictures of pests and diseases;

[0011] Step 2: Preprocess the pest and disease images;

[0012] Step 3: Use pest and disease images to train a multi-channel network model composed of a convolutional neural network and a capsule network to obtain a pest and disease recognition model;

[0013] Step 4: Collect crop pictures and identify pests and diseases using the pest and disease identification model.

[0014] Optionally, in the above-mentioned crop pest and disease identification method, in step 2, the pest and disease image is preprocessed, including smoothing filtering and sharpening processing.

[0015] Optionally, in the above-mentioned crop disease and pest identification method, the multi-channel network model is a deep learning model composed of multiple convolutional layers, pooling layers, primary capsule layers and softmax layers, and the fully connected layer is discarded.

[0016] Optionally, in the above-mentioned crop pest and disease identification method, in the multi-channel network model, after the convolution layer extracts features, a pooling layer is applied to reduce the size of the pest and disease image to reduce the calculation parameters in the network, and accurate identification of the pest and disease is achieved in the softmax layer.

[0017] Optionally, in the above-mentioned crop pest and disease identification method, in the multi-channel network model, the Grabcut algorithm in OpenCV is used to segment the pest and disease image, texture information and boundary information in the pest and disease image are collected, each pixel of the pest and disease image is regarded as a node, and the dissimilarity measure between two pixels is represented by weighting the edges between the nodes, and the maximum flow or minimum flow algorithm is used to cut the graph, and the obtained minimum cut corresponds to the target boundary to be extracted.

[0018] This application has at least the following beneficial technical effects:

[0019] A method for identifying pests and diseases on crops is provided. The method collects pest and disease pictures, pre-processes the pest and disease pictures, and then trains a multi-channel network model composed of a convolutional neural network and a capsule network to obtain a pest and disease identification model. The pest and disease identification model can then be used to identify pests and diseases on crop pictures to determine whether pests and diseases are present on crops. The method can timely detect and identify pests and diseases on crops over a large area, control the pests and diseases, and avoid serious impact on crop growth. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 is a schematic diagram of a method for identifying plant diseases and insect pests on crops provided in an embodiment of the present application;

[0021] Figure 2 It is a schematic diagram of the pest and disease pictures provided in the embodiments of the present application;

[0022] Figure 3 This is a schematic diagram of the cut picture of the pests and diseases provided in the embodiment of the present application.

[0023] In order to better illustrate this embodiment, some contents of the drawings may be omitted, which is only used for exemplary description and should not be construed as a limitation of this patent. DETAILED DESCRIPTION

[0024] In order to make the technical solution and advantages of the present application clearer, the technical solution of the present application will be further described in detail in detail and in detail in conjunction with the accompanying drawings. It can be understood that the specific embodiments described here are only partial embodiments of the present application, which are only used to explain the present application, not to limit the present application. It should be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, and other related parts can refer to the general design. In the absence of conflict, the embodiments in the present application and the technical features in the embodiments can be combined with each other to obtain new embodiments.

[0025] In addition, unless otherwise defined, the technical terms or scientific terms used in the description of this application shall be the common meanings understood by the general technicians in the field to which this application belongs. The words "upper", "lower", "left", "right", "center", "vertical", "horizontal", "inside", "outside" and other words indicating orientation used in the description of this application are only used to indicate the relative direction or positional relationship, and do not imply that the device or element must have a specific orientation, be constructed and operated in a specific orientation. When the absolute position of the described object changes, its relative positional relationship may also change accordingly, so it cannot be understood as a limitation on this application. The "first", "second", "third" and similar terms used in the description of this application are only used for descriptive purposes to distinguish different components, and cannot be understood as indicating or implying relative importance. The similar words "one", "one" or "the" used in the description of this application should not be understood as an absolute limitation on quantity, but should be understood as the existence of at least one. The similar words "including" or "comprising" used in the description of this application mean that the elements or objects appearing in front of the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects.

[0026] In addition, it should be noted that, unless otherwise clearly specified and limited, the words "installed", "connected", "connected" and similar terms used in the description of this application should be understood in a broad sense. For example, the connection can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, an indirect connection through an intermediate medium, or the internal connection of two components. Technical personnel in the field can understand their specific meanings in this application according to the specific circumstances.

[0027] A method for identifying pests and diseases on crops, such as Figure 1 shown.

[0028] Step 1: Collect pictures of pests and diseases.

[0029] Collect pictures of powdery mildew, ergot, and leaf rust on crops. You can collect them from the Internet or by taking photos.

[0030] The collected pest and disease images can be saved in a folder, read with a Python program, renamed and marked with pest and disease categories, and the pest and disease image names and their pest and disease categories can be recorded in a CSV file as the overall data set.

[0031] Step 2: Preprocess the pest and disease images, including smoothing filtering and sharpening, so as to improve the accuracy of feature extraction, segmentation and classification of pest and disease images.

[0032] Step 3: Use pest and disease images to train the multi-channel network model composed of convolutional neural network and capsule network to obtain the pest and disease recognition model.

[0033] The multi-channel network model is a deep learning model composed of multiple convolutional layers, pooling layers, primary capsule layers and softmax layers.

[0034] In the multi-channel network model, the Grabcut algorithm in OpenCV is used to segment the pest image, collect the texture information and boundary information in the pest image, regard each pixel of the pest image as a node, and represent the dissimilarity measure between two pixels by weighting the edges between nodes. The maximum flow or minimum flow algorithm is used to cut the graph, and the minimum cut obtained corresponds to the target boundary to be extracted. In a specific example, the pest image is as follows: Figure 2 As shown, after cutting Figure 3 shown.

[0035] In the multi-channel network model, after the convolution layer extracts features, the pooling layer is used to reduce the size of the pest and disease images to reduce the calculation parameters in the network. Finally, the fully connected layer is discarded in combination with the Inception model idea, and accurate identification of pests and diseases is achieved in the softmax layer.

[0036] Step 4: Collect crop pictures and identify pests and diseases using the pest and disease identification model.

[0037] A rotating camera or a drone with a camera can be used to collect crop images at a certain frequency.

[0038] The pest and disease recognition model can be used to identify the presence of pests and diseases in crop images, and then determine the type and quantity of pests and diseases.

[0039] The crop pest and disease identification method disclosed in the above embodiment collects pest and disease images, pre-processes the pest and disease images, and then uses them to train a multi-channel network model composed of a convolutional neural network and a capsule network to obtain a pest and disease identification model, which can then be used to identify pests and diseases in crop images to determine whether there are pests and diseases on the crops. It can timely detect and identify pests and diseases on crops over a large range, control the pests and diseases, and avoid serious impact on crop growth.

[0040] So far, the technical solution of the present application has been described in conjunction with the preferred embodiments shown in the accompanying drawings. Those skilled in the art should understand that the scope of protection of the present application is obviously not limited to these specific embodiments. Without departing from the principles of the present application, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the scope of protection of the present application.

Claims

1. A method for identifying plant diseases and insect pests on crops, characterized in that: include: Step 1: Collect pictures of pests and diseases; Step 2: Preprocess the pest and disease images; Step 3: Use pest and disease images to train a multi-channel network model composed of a convolutional neural network and a capsule network to obtain a pest and disease recognition model; Step 4: Collect crop pictures and identify pests and diseases using the pest and disease identification model.

2. The method for identifying crop pests and diseases according to claim 1, characterized in that: In step 2, the pest and disease images are preprocessed, including smoothing filtering and sharpening.

3. The method for identifying crop pests and diseases according to claim 2, characterized in that: The multi-channel network model is a deep learning model composed of multiple convolutional layers, pooling layers, primary capsule layers and softmax layers, and the fully connected layer is discarded.

4. The method for identifying crop pests and diseases according to claim 3, characterized in that: In the multi-channel network model, after the convolution layer extracts features, the pooling layer is applied to reduce the size of the pest and disease images to reduce the calculation parameters in the network and achieve accurate identification of pests and diseases in the softmax layer.

5. The method for identifying crop pests and diseases according to claim 4, characterized in that: In the multi-channel network model, the Grabcut algorithm in OpenCV is used to segment the pest and disease images, collect texture information and boundary information in the pest and disease images, regard each pixel of the pest and disease image as a node, and represent the dissimilarity measure between two pixels by weighting the edges between the nodes. The maximum flow or minimum flow algorithm is used to cut the graph, and the obtained minimum cut corresponds to the target boundary to be extracted.

Images