CN110659659A - Method and system for intelligently identifying and early warning pests - Google Patents
Method and system for intelligently identifying and early warning pests Download PDFInfo
- Publication number
- CN110659659A CN110659659A CN201810794841.2A CN201810794841A CN110659659A CN 110659659 A CN110659659 A CN 110659659A CN 201810794841 A CN201810794841 A CN 201810794841A CN 110659659 A CN110659659 A CN 110659659A
- Authority
- CN
- China
- Prior art keywords
- pest
- pests
- neural network
- early warning
- image data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24133—Distances to prototypes
- G06F18/24137—Distances to cluster centroïds
- G06F18/2414—Smoothing the distance, e.g. radial basis function networks [RBFN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2431—Multiple classes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to the field of biological information, in particular to a method and a system for intelligently identifying and early warning pests. The pest image data are collected and enhanced; denoising and labeling the pest image data; training and transfer learning are carried out on pest image data through a convolutional neural network; leading the target pest image into a trained neural network to predict the pest category; and counting and generating the existing data information of the pest image and the pest early warning data information. The method can be expanded to be used for detecting field plants, insects and pests, is not influenced by the outside in many aspects, and realizes real-time classification; can automatic identification pest kind and statistics pest quantity, provide higher identification accuracy degree and liberate more manpowers, save time and manpower to can carry out the early warning to the pest.
Description
Technical Field
The invention relates to the field of biological information, in particular to a method and a system for intelligently identifying and early warning pests.
Background
In the field of biological information, pest discrimination and early warning are mainly relied on at the present stage:
1. manual identification: the traditional manual identification is mainly to judge the pest species directly with naked eyes or by means of tools such as a microscope and the like and count the number of the pest species. The method has the advantages of great workload, low recognition efficiency, inevitable human errors caused by long-time work, high requirements on detection personnel and rich pest-related prior knowledge.
2. By means of voice recognition: and (3) building a pest sound collection related environment, and identifying pests through sounds emitted by behaviors such as pest chirping, moving, ingestion and the like. The method is difficult to set up a sound collection environment, the identification precision is easily influenced by various noises, and the identified pest species are less, so the method is difficult to apply in an actual scene.
3. The traditional image algorithm: the method of pest trapping and manual analysis/pattern recognition is long in time and labor-consuming. The traditional image algorithm mainly detects pests by analyzing various forms, sizes and colors of the pests. The identification process is simple, the algorithm adaptability is poor, and the method is easily influenced by various factors of environment change.
Disclosure of Invention
Aiming at the defects of large calculated amount, high difficulty in feature extraction, easiness in losing other features and the like, the invention provides the method and the system for intelligently identifying and early warning the pests, so that the time and the labor are saved for intelligently identifying and early warning the pests, and the accuracy is high.
The invention is realized by the following technical scheme:
a method for intelligently identifying and early warning pests specifically comprises the following steps:
step S10, pest image data are collected and enhanced;
step S20, denoising and labeling the pest image data;
step S30, training and transfer learning are carried out on pest image data through a neural network;
step S40, leading the target pest image into the trained neural network to predict the pest category;
and step S50, counting the existing data information of the pest image and the pest early warning data information.
Further, before step S10, the method further includes the steps of:
step S100, data transformation is carried out on the pest image.
Further, the pest image data are marked to mark the specific position and the category of each pest in the picture;
further, in step S30, the neural network is a convolutional neural network,
the convolutional neural network is specifically as follows: an end-to-end deep neural network;
the end-to-end deep neural network comprises a feature extraction network and a target prediction network;
the characteristic extraction network is a deep convolutional neural network with 43 layers and a residual error structure;
the target prediction network is a multi-scale prediction structure convolutional neural network.
In order to achieve the above object, the present invention further provides a system for intelligently identifying and early warning pests, the system comprising:
the data acquisition unit is used for collecting pest image data and enhancing the pest image data;
the data labeling unit is used for denoising and labeling the pest image data;
the training and transferring unit is used for training and transferring and learning the pest image data through a neural network;
the prediction unit is used for leading the target pest image into the trained neural network to predict the pest category;
and the statistic and early warning unit is used for counting the existing data information and the early warning data information of the pests for generating the pest images.
Further, the system further comprises:
and the image transformation module is used for carrying out data transformation on the pest image.
In order to achieve the above object, the present invention further provides a platform for intelligently identifying and early warning pests, which comprises a processor, a memory and a platform control program for intelligently identifying and early warning pests;
wherein the processor executes the platform control program, the platform control program for intelligently identifying and early warning pests is stored in the memory, and the platform control program for intelligently identifying and early warning pests implements the method steps for intelligently identifying and early warning pests.
In order to achieve the above object, the present invention further provides a computer readable storage medium, where the computer readable storage medium stores a platform control program for intelligently identifying and early warning pests, and the platform control program for intelligently identifying and early warning pests implements the method steps for intelligently identifying and early warning pests.
Compared with the prior art, the invention has the following beneficial effects:
through the method for intelligently identifying and early warning the pests,
step S10, pest image data are collected and enhanced;
step S20, denoising and labeling the pest image data;
step S30, training and transfer learning are carried out on pest image data through a neural network;
step S40, leading the target pest image into the trained neural network to predict the pest category;
and step S50, counting the existing data information of the pest image and the pest early warning data information.
And correspondingly the system unit:
the data acquisition unit is used for collecting pest image data and enhancing the pest image data;
the data labeling unit is used for denoising and labeling the pest image data;
the training and transferring unit is used for training and transferring and learning the pest image data through a neural network;
the prediction unit is used for leading the target pest image into the trained neural network to predict the pest category;
and the statistic and early warning unit is used for counting the existing data information and the early warning data information of the pests for generating the pest images.
The method can be expanded to be used for detecting field plants, insects and pests, is not influenced by the outside in many aspects, and realizes real-time classification; can automatic identification pest kind and statistics pest quantity, provide higher identification accuracy degree and liberate more manpowers, save time and manpower to can carry out the early warning to the pest.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a schematic diagram of an architecture of a method for intelligently identifying and early warning pests according to the present invention;
FIG. 2 is a schematic diagram of the overall network structure of a method for intelligently identifying and early warning pests according to the present invention;
FIG. 3 is a schematic diagram of a method for field lesion leaf detection; FIG. 4 is a schematic diagram of a system for intelligently identifying and early warning pests according to the present invention;
FIG. 5 is a schematic diagram of a system module architecture for intelligently identifying and early warning pests according to the present invention;
the objects, features and advantages of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
For better understanding of the objects, aspects and advantages of the present invention, reference will now be made to the following detailed description taken in conjunction with the accompanying drawings, and other advantages and capabilities of the present invention will become apparent to those skilled in the art from the description.
The invention is capable of other and different embodiments and its several details are capable of modification in various other respects, all without departing from the spirit and scope of the present invention.
It should be noted that, if directional indications (such as up, down, left, right, front, and back … …) are involved in the embodiment of the present invention, the directional indications are only used to explain the relative positional relationship between the components, the movement situation, and the like in a specific posture (as shown in the drawing), and if the specific posture is changed, the directional indications are changed accordingly.
In addition, if there is a description of "first", "second", etc. in an embodiment of the present invention, the description of "first", "second", etc. is for descriptive purposes only and is not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. Secondly, the technical solutions in the embodiments can be combined with each other, but it must be based on the realization of those skilled in the art, and when the technical solutions are contradictory or cannot be realized, the combination of the technical solutions should be considered to be absent and not be within the protection scope of the present invention.
Preferably, the method for intelligently identifying and early warning pests is applied to one or more terminals or servers. The terminal is a device capable of automatically performing numerical calculation and/or information processing according to a preset or stored instruction, and the hardware includes, but is not limited to, a microprocessor, an Application Specific Integrated Circuit (ASIC), a Programmable Gate Array (FPGA), a Digital Signal Processor (DSP), an embedded device, and the like.
The terminal can be a desktop computer, a notebook, a palm computer, a cloud server and other computing equipment. The terminal can be in man-machine interaction with a client in a keyboard mode, a mouse mode, a remote controller mode, a touch panel mode or a voice control device mode.
The invention provides a method and a system for intelligently identifying and early warning pests for intelligently identifying and early warning pest data.
Fig. 1 is a flow chart of a method for intelligently identifying and early warning pests according to an embodiment of the present invention.
In this embodiment, the method for intelligently identifying and early warning pests may be applied to a terminal or a fixed terminal having a display function, where the terminal is not limited to a personal computer, a smart phone, a tablet computer, a desktop or all-in-one machine with a camera, and the like.
The intelligent pest identification and early warning method can also be applied to a hardware environment formed by a terminal and a server connected with the terminal through a network. Networks include, but are not limited to: a wide area network, a metropolitan area network, or a local area network. The method for intelligently identifying and early warning pests can be executed by a server, can also be executed by a terminal, and can also be executed by the server and the terminal together.
For example, for a terminal which needs to perform intelligent pest identification and early warning, the functions of intelligent pest identification and early warning provided by the method of the present invention can be directly integrated on the terminal, or a client for implementing the method of the present invention is installed. For another example, the method provided by the present invention can also be operated on a server or other devices in the form of Software Development Kit (SDK), an interface for intelligently identifying and early warning pests is provided in the form of SDK, and the terminal or other devices can implement the functions of intelligently identifying and early warning pests through the provided interface.
As shown in FIG. 1, the present invention provides a method for intelligently identifying and warning pests, which specifically comprises the following steps, the sequence of the steps in the flow chart can be changed according to different requirements, and some steps can be omitted.
Step S10, pest image data are collected and enhanced;
in the embodiment of the present invention, before step S10, the method further includes the steps of:
step S100, data transformation is carried out on the pest image.
Specifically, the image data are various types of pictures, the trained image data are collected through a pest trapping platform and an image automatic acquisition platform, data preprocessing and enhancement are performed on original image data samples through image transformation algorithms (such as turning, mirroring, rotation, appropriate noise enhancement, appropriate blurring, affine transformation and transmission transformation), the original data are expanded by more than ten times, and the purpose of data enhancement is to expand the image data volume, so that the number of the image data used for training is more, and scenes are more diversified.
Step S20, denoising and labeling the pest image data;
the pest image data are marked to mark the specific position and the category of each pest in the picture;
specifically, the labeling is to label the type and specific position of each pest in the image, that is, manually label the pest image data, and manually label the specific position and the belonging category of each pest in the image. And after the marking is finished, generating a marking file with an xml format, wherein the file records the category, position coordinates and width and height of each pest in the picture. One picture corresponds to one label file, and finally the original picture and the label file are used for training the neural network.
The denoising treatment specifically comprises the following steps: and removing the particle and salt and pepper noises through median filtering.
Pest image data before being sent to training and transfer learning needs to be subjected to denoising processing;
specifically, the denoising processing of the pest image data before training and transfer learning is specifically to remove particles and salt and pepper noises through median filtering.
That is, filtering and denoising are firstly performed on a picture to be predicted, median filtering is used for removing particles and salt and pepper noises, and Gaussian filtering is used for denoising so that the picture is smoother. The filtered and denoised pest picture is sent into a trained deep convolutional neural network, the network can automatically predict the pest type and position, and finally the type of each pest and the specific position in the picture are given
Step S30, training and transfer learning are carried out on pest image data through a convolutional neural network;
in step S30, the neural network is a convolutional neural network:
the convolutional neural network is specifically as follows: an end-to-end deep neural network;
the end-to-end deep neural network comprises a feature extraction network and a target prediction network; the characteristic extraction network is a deep convolutional neural network with 43 layers and a residual error structure; the feature extraction network sets the size of the extracted feature map (feature map size threshold) according to the density of pests in the picture, and the traditional feature extraction network sets the size of the corresponding finally extracted feature map according to the number of targets contained in the training set picture. The target prediction network is a multi-scale prediction structure convolutional neural network.
Preferably, the last layer of the convolutional neural network sets the number of neural nodes to be 50, so that for an input picture, the neural network will output the probability of corresponding 50 classes of classification. The number of ganglion points in the output layer of the conventional convolutional neural network is determined by the total number of different classes.
Specifically, as shown in fig. 2, an end-to-end target detection deep neural network is established, and the model can predict all pest categories in a picture and mark specific positions at one time only by inputting a picture to be detected. And because the network model is end-to-end, the classification is predicted without separating redundant foreground and background in the model prediction calculation, and the specific positions of the pests are directly classified and regressed, so that the algorithm speed is greatly improved, and the real-time target detection can be achieved. The model is divided into two parts: a feature extraction network and a target prediction network. The feature extraction network is a 43-layer deep convolutional neural network with a residual error structure, the deeper the number of the neural network layers is, the stronger the capability of feature extraction and classification identification is, but the greater the difficulty of training the network is, so the residual error structure is added into the network, the convergence speed of network training is accelerated, and the identification accuracy is improved.
The 43-layer convolutional neural network extracts various features through a convolutional structure of each layer, accelerates the training convergence speed by using residual connection, and finally outputs the classification result of feature extraction by adopting a softmax function.
The target prediction network is a convolutional neural network adopting a multi-scale prediction structure, and input data of the target prediction network is output of a previous feature extraction network. The target prediction network uses a 7-layer convolutional neural network for the last layer output of the feature extraction network, the sizes of the first, third and fifth layers of convolutional kernels are 1x1, and the number of convolutional channels is 512; the sizes of the second, fourth and sixth layers of convolution kernels are 3x3, and the number of convolution channels is 1024; the seventh layer of convolution kernel has a size of 3x3 and the number of convolution channels is 3x (classes +5), where classes is the number of pest categories to be predicted, thereby outputting the first target prediction feature map, and the output size is 19x19x3x (classes + 5). The first target prediction feature map is then upsampled using two-line new interpolation, and the above 7-layer convolution operation is repeated to obtain a second target prediction feature map, with an output size of 38x38x3x (classes + 5). And continuously performing upsampling on the second target prediction feature map to obtain a third target prediction feature map, wherein the output size is 76x76x3x (classes + 5). By calculating the Loss function of Loss, namely coordErr + iouErr + clsrr, the types and positions of all pests in the picture can be predicted at one time finally. Wherein coordErr is the error between the predicted coordinate and the real coordinate of the pest; the maximum overlapping proportion of the iouErr pest prediction frame and the real frame is 1; clsrer is the error between the predicted category and the actual category of the pest.
In this embodiment, the image input size 224x224 is obtained by using a 43-layer network of feature extraction in the class 1000 data training step in ImageNet for the migration learning pre-training. Combining the pre-trained feature extraction network with the target prediction network, training the whole network model by using the labeled pest data, and adjusting the input size of the image to 608x 608.
And step S40, importing the target pest image into the trained neural network to predict the pest type.
And step S50, counting the existing data information of the pest image and the pest early warning data information.
Specifically, the final artificial intelligence deep neural network can realize automatic prediction of pest species and give specific positions of pests in the image. According to the output of the neural network, the categories and the corresponding number of the detected pests, and the corresponding positions and sizes of the pests are counted. In the visual interface, the total number of the detected pests, the type, the position and the size of each pest can be clearly given, and the dynamic condition of the data information can be conveniently monitored in real time by presenting through the visual interface.
The pest early warning data information is used for predicting the pest situation of the analysis area, namely the future pest prediction situation in the area is estimated according to the data obtained by system analysis, and defense work is facilitated according to the prediction data.
As shown in fig. 3, another embodiment of the present invention provides a method for detecting field lesion leaves, which comprises the following steps:
step S1001, collecting a leaf picture, and marking out a pathological change leaf in the picture;
step S1002, inputting the pictures and the labeled data into a network for training;
step S1003, inputting the newly collected leaf picture into a trained network for detection and identification;
and step S1004, counting the ratio of the leaves with pathological changes, and giving an alarm when the ratio reaches a set threshold value.
Preferably, the method for intelligently identifying and early warning pests is also suitable for identifying various pests in the field, such as: stinkbug, white moth, black cutworm, cotton bollworm, white striped beetle, and cicada.
As shown in fig. 4, the present invention provides a system for intelligently identifying and early warning pests, which specifically comprises:
the data acquisition unit is used for collecting pest image data and enhancing the pest image data;
the data labeling unit is used for denoising and labeling the pest image data;
the pest image data are marked to mark the specific position and the category of each pest in the picture;
specifically, the pest image data are manually marked, and the specific position and the category of each pest in the picture are manually marked. And after the marking is finished, generating a marking file with an xml format, wherein the file records the category, position coordinates and width and height of each pest in the picture. One picture corresponds to one label file, and finally the original picture and the label file are used for training the neural network.
The denoising treatment specifically comprises the following steps: and removing the particle and salt and pepper noises through median filtering.
Pest image data before being sent to training and transfer learning needs to be subjected to denoising processing;
specifically, the denoising processing of the pest image data before training and transfer learning is specifically to remove particles and salt and pepper noises through median filtering.
That is, filtering and denoising are firstly performed on a picture to be predicted, median filtering is used for removing particles and salt and pepper noises, and Gaussian filtering is used for denoising so that the picture is smoother. The filtered and denoised pest picture is sent into a trained deep convolutional neural network, the network can automatically predict the pest type and position, and finally the type of each pest and the specific position in the picture are given
The training and transferring unit is used for training and transferring and learning the pest image data through a neural network;
the prediction unit is used for leading the target pest image into the trained neural network to predict the pest category;
and the statistic and early warning unit is used for counting the existing data information and the early warning data information of the pests for generating the pest images.
Specifically, the final artificial intelligence deep neural network can realize automatic prediction of pest species and give specific positions of pests in the image. According to the output of the neural network, the categories and the corresponding number of the detected pests, and the corresponding positions and sizes of the pests are counted. In the visual interface, the total number of the detected pests, the type, the position and the volume size of each pest can be clearly given.
As shown in fig. 5, the system further includes:
the image transformation module is used for carrying out data transformation on the pest image;
specifically, trained image data are collected through a pest trapping platform and an automatic image acquisition platform, data preprocessing and enhancement are carried out on an original image data sample through an image transformation algorithm (such as turning, mirroring, rotating, properly increasing noise, properly blurring, affine transformation and transmission transformation), and the original data are expanded by more than ten times.
Specifically, the end-to-end deep neural network comprises a feature extraction network and a target prediction network; the characteristic extraction network is a deep convolutional neural network with 43 layers and a residual error structure; the target prediction network is a multi-scale prediction structure convolutional neural network.
Through the steps of the intelligent pest identification and early warning method, and the functional units and functional modules of the system, compared with the traditional operation method, the method has the advantages of good identification effect, good prediction capability on small targets in the graph, different pest detection effects, capability of training a network model with good prediction capability under the condition of few samples, effectively accelerated training speed, and capability of refining pest types, thereby improving the identification of pest types (more than 5000 types).
Preferably, pictures are collected through the continuous lens, pest species are automatically identified, pest quantity is automatically counted, and time and labor are saved. Through different data set training, the method can be expanded to be used for detecting field plants, detecting insects, detecting pest videos and classifying in real time; according to the actual application requirements, the last output classes number of the model is modified, namely the last output network model can be modified, and the amplification of the prediction classes is realized, so that the identification classes are increased in the later period.
The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.
Claims (8)
1. A method for intelligently identifying and early warning pests is characterized by comprising the following steps:
step S10, pest image data are collected and enhanced;
step S20, denoising and labeling the pest image data;
step S30, training and transfer learning are carried out on pest image data through a neural network;
step S40, leading the target pest image into the trained neural network to predict the pest category;
and step S50, counting the existing data information of the pest image and the pest early warning data information.
2. The method for intelligently identifying and warning pests as claimed in claim 1, further comprising, before step S10, the steps of:
step S100, data transformation is carried out on the pest image.
3. The method for intelligently identifying and warning pests according to claim 1, wherein the pest image data is labeled to mark the specific position and the category of each pest in the picture;
the denoising treatment specifically comprises the following steps: and removing the particle and salt and pepper noises through median filtering.
4. The method for intelligently identifying and warning pests according to claim 1, wherein the neural network in step S30 is a convolutional neural network;
the convolutional neural network is specifically as follows: an end-to-end deep neural network;
the end-to-end deep neural network comprises a feature extraction network and a target prediction network;
the characteristic extraction network is a deep convolutional neural network with 43 layers and a residual error structure;
the target prediction network is a multi-scale prediction structure convolutional neural network.
5. A system for intelligently identifying and early warning pests, the system comprising:
the data acquisition unit is used for collecting pest image data and enhancing the pest image data;
the data labeling unit is used for denoising and labeling the pest image data;
the training and transferring unit is used for training and transferring and learning the pest image data through a neural network;
the prediction unit is used for leading the target pest image into the trained neural network to predict the pest category;
and the statistic and early warning unit is used for counting the existing data information and the early warning data information of the pests for generating the pest images.
6. The system for intelligently identifying and warning pests according to claim 5, further comprising:
and the image transformation module is used for carrying out data transformation on the pest image.
7. The utility model provides a platform of intelligent recognition and early warning pest which characterized in that includes:
the system comprises a processor, a memory and a platform control program for intelligently identifying and early warning pests;
wherein the processor executes the platform control program, the platform control program for intelligent pest identification and warning being stored in the memory, the platform control program for intelligent pest identification and warning implementing the method steps of intelligent pest identification and warning as claimed in any one of claims 1 to 4.
8. A computer-readable storage medium, characterized in that the computer-readable storage medium stores a platform control program for intelligently identifying and early warning pests, which implements the method steps of intelligently identifying and early warning pests according to any one of claims 1 to 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810794841.2A CN110659659A (en) | 2018-07-19 | 2018-07-19 | Method and system for intelligently identifying and early warning pests |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810794841.2A CN110659659A (en) | 2018-07-19 | 2018-07-19 | Method and system for intelligently identifying and early warning pests |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110659659A true CN110659659A (en) | 2020-01-07 |
Family
ID=69028514
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810794841.2A Pending CN110659659A (en) | 2018-07-19 | 2018-07-19 | Method and system for intelligently identifying and early warning pests |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110659659A (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111507314A (en) * | 2020-06-05 | 2020-08-07 | 张立 | Artificial intelligence image data acquisition system of insect pest control facility |
| CN111611889A (en) * | 2020-05-12 | 2020-09-01 | 安徽大学 | Miniature insect pest recognition device in farmland based on improved convolutional neural network |
| CN111696127A (en) * | 2020-06-02 | 2020-09-22 | 中国联合网络通信集团有限公司 | Locust plague analysis and early warning system and method |
| CN111883205A (en) * | 2020-07-14 | 2020-11-03 | 云南省烟草农业科学研究院 | Prediction method for selecting harmful components of tobacco based on whole genome and application |
| CN111887473A (en) * | 2020-07-14 | 2020-11-06 | 云南省烟草农业科学研究院 | Prediction method for smoke harmful component release amount based on whole genome selection and application |
| CN111914914A (en) * | 2020-07-21 | 2020-11-10 | 上海理想信息产业（集团）有限公司 | Method, device, equipment and storage medium for identifying plant diseases and insect pests |
| CN112001290A (en) * | 2020-08-18 | 2020-11-27 | 南京工程学院 | Rice planthopper migration flight path prediction method based on YOLO algorithm |
| CN112861707A (en) * | 2021-02-03 | 2021-05-28 | 重庆市风景园林科学研究院 | Harmful organism visual identification method, device, equipment and readable storage medium |
| CN113076873A (en) * | 2021-04-01 | 2021-07-06 | 重庆邮电大学 | Crop disease long-tail image identification method based on multi-stage training |
| CN114429556A (en) * | 2020-10-15 | 2022-05-03 | 中移动信息技术有限公司 | Picture auditing method and device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017194276A1 (en) * | 2016-05-13 | 2017-11-16 | Basf Se | System and method for detecting plant diseases |
| CN107463937A (en) * | 2017-06-20 | 2017-12-12 | 大连交通大学 | A kind of tomato pest and disease damage automatic testing method based on transfer learning |
| CN108090417A (en) * | 2017-11-27 | 2018-05-29 | 上海交通大学 | A kind of method for detecting human face based on convolutional neural networks |
| KR20180058049A (en) * | 2016-11-23 | 2018-05-31 | 세종대학교산학협력단 | Apparatus and method for diagonising disease and insect pest of crops |
| CN108111614A (en) * | 2017-12-27 | 2018-06-01 | 北京工业大学 | Agriculture and forestry pest and disease monitoring manages system |
-
2018
- 2018-07-19 CN CN201810794841.2A patent/CN110659659A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017194276A1 (en) * | 2016-05-13 | 2017-11-16 | Basf Se | System and method for detecting plant diseases |
| KR20180058049A (en) * | 2016-11-23 | 2018-05-31 | 세종대학교산학협력단 | Apparatus and method for diagonising disease and insect pest of crops |
| CN107463937A (en) * | 2017-06-20 | 2017-12-12 | 大连交通大学 | A kind of tomato pest and disease damage automatic testing method based on transfer learning |
| CN108090417A (en) * | 2017-11-27 | 2018-05-29 | 上海交通大学 | A kind of method for detecting human face based on convolutional neural networks |
| CN108111614A (en) * | 2017-12-27 | 2018-06-01 | 北京工业大学 | Agriculture and forestry pest and disease monitoring manages system |
Non-Patent Citations (2)
| Title |
|---|
| EUSEBIO L. MIQUE等: "Rice Pest and Disease Detection Using Convolutional Neural Network", 《ICISS "18: PROCEEDINGS OF THE 2018 INTERNATIONAL CONFERENCE ON INFORMATION SCIENCE AND SYSTEM》 * |
| JOSEPH REDMON等: "YOLOv3:An Incremental Improvement", 《COMPUTER VISION AND PATTERN RECOGNITION》 * |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111611889A (en) * | 2020-05-12 | 2020-09-01 | 安徽大学 | Miniature insect pest recognition device in farmland based on improved convolutional neural network |
| CN111611889B (en) * | 2020-05-12 | 2023-04-18 | 安徽大学 | Miniature insect pest recognition device in farmland based on improved convolutional neural network |
| CN111696127A (en) * | 2020-06-02 | 2020-09-22 | 中国联合网络通信集团有限公司 | Locust plague analysis and early warning system and method |
| CN111507314A (en) * | 2020-06-05 | 2020-08-07 | 张立 | Artificial intelligence image data acquisition system of insect pest control facility |
| CN111887473B (en) * | 2020-07-14 | 2022-07-01 | 云南省烟草农业科学研究院 | Prediction method for smoke harmful component release amount based on whole genome selection and application |
| CN111883205A (en) * | 2020-07-14 | 2020-11-03 | 云南省烟草农业科学研究院 | Prediction method for selecting harmful components of tobacco based on whole genome and application |
| CN111887473A (en) * | 2020-07-14 | 2020-11-06 | 云南省烟草农业科学研究院 | Prediction method for smoke harmful component release amount based on whole genome selection and application |
| CN111883205B (en) * | 2020-07-14 | 2023-10-20 | 云南省烟草农业科学研究院 | Prediction method for selecting harmful ingredients of tobacco based on whole genome and application |
| CN111914914A (en) * | 2020-07-21 | 2020-11-10 | 上海理想信息产业（集团）有限公司 | Method, device, equipment and storage medium for identifying plant diseases and insect pests |
| CN112001290A (en) * | 2020-08-18 | 2020-11-27 | 南京工程学院 | Rice planthopper migration flight path prediction method based on YOLO algorithm |
| CN112001290B (en) * | 2020-08-18 | 2023-10-27 | 南京工程学院 | Rice planthopper migration path prediction method based on YOLO algorithm |
| CN114429556A (en) * | 2020-10-15 | 2022-05-03 | 中移动信息技术有限公司 | Picture auditing method and device |
| CN112861707A (en) * | 2021-02-03 | 2021-05-28 | 重庆市风景园林科学研究院 | Harmful organism visual identification method, device, equipment and readable storage medium |
| CN113076873A (en) * | 2021-04-01 | 2021-07-06 | 重庆邮电大学 | Crop disease long-tail image identification method based on multi-stage training |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110659659A (en) | Method and system for intelligently identifying and early warning pests | |
| CN107123131B (en) | Moving target detection method based on deep learning | |
| Kalsotra et al. | Background subtraction for moving object detection: explorations of recent developments and challenges | |
| CN111178183B (en) | Face detection method and related device | |
| CN110059642B (en) | Face image screening method and device | |
| CN109086811B (en) | Multi-label image classification method and device and electronic equipment | |
| CN110009556A (en) | Image background weakening method, device, storage medium and electronic equipment | |
| CN101179713A (en) | Method of detecting single moving target under complex background | |
| CN108416780B (en) | Object detection and matching method based on twin-region-of-interest pooling model | |
| KR101884611B1 (en) | Method of extracting objects-of-Interest out of CCTV video using metadate filtering of moving objects | |
| CN111723773B (en) | Method and device for detecting carryover, electronic equipment and readable storage medium | |
| CN109685045A (en) | A kind of Moving Targets Based on Video Streams tracking and system | |
| CN106815576B (en) | Target tracking method based on continuous space-time confidence map and semi-supervised extreme learning machine | |
| CN104616002A (en) | Facial recognition equipment used for judging age groups | |
| CN112381104A (en) | Image identification method and device, computer equipment and storage medium | |
| Cheng et al. | Advanced background subtraction approach using Laplacian distribution model | |
| Shafiee et al. | Embedded motion detection via neural response mixture background modeling | |
| Abdul Rahman et al. | Enhancement of background subtraction techniques using a second derivative in gradient direction filter | |
| CN109831634A (en) | The density information of target object determines method and device | |
| CN113128368A (en) | Method, device and system for detecting character interaction relationship | |
| CN108932465B (en) | Method and device for reducing false detection rate of face detection and electronic equipment | |
| CN113947744A (en) | Fire image detection method, system, equipment and storage medium based on video | |
| Fan et al. | Two-stage salient region detection by exploiting multiple priors | |
| CN110321808B (en) | Method, apparatus and storage medium for detecting carry-over and stolen object | |
| CN104809438B (en) | A kind of method and apparatus for detecting electronic eyes |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200107 |
|
| RJ01 | Rejection of invention patent application after publication |