CN113744226A - Intelligent agricultural pest identification and positioning method and system - Google Patents
Intelligent agricultural pest identification and positioning method and system Download PDFInfo
- Publication number
- CN113744226A CN113744226A CN202110994521.3A CN202110994521A CN113744226A CN 113744226 A CN113744226 A CN 113744226A CN 202110994521 A CN202110994521 A CN 202110994521A CN 113744226 A CN113744226 A CN 113744226A
- Authority
- CN
- China
- Prior art keywords
- pest
- target detection
- agricultural
- module
- orchard
- 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
- 241000607479 Yersinia pestis Species 0.000 title claims abstract description 135
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000001514 detection method Methods 0.000 claims abstract description 69
- 239000002420 orchard Substances 0.000 claims abstract description 42
- 238000011156 evaluation Methods 0.000 claims abstract description 17
- 238000010606 normalization Methods 0.000 claims abstract description 9
- 238000012549 training Methods 0.000 claims abstract description 8
- 238000012545 processing Methods 0.000 claims abstract description 5
- 238000007781 pre-processing Methods 0.000 claims description 5
- 230000008569 process Effects 0.000 claims description 5
- 230000002776 aggregation Effects 0.000 claims description 3
- 238000004220 aggregation Methods 0.000 claims description 3
- 238000011176 pooling Methods 0.000 claims description 3
- 238000012360 testing method Methods 0.000 abstract description 8
- 230000005540 biological transmission Effects 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 241001481710 Cerambycidae Species 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- HPTJABJPZMULFH-UHFFFAOYSA-N 12-[(Cyclohexylcarbamoyl)amino]dodecanoic acid Chemical compound OC(=O)CCCCCCCCCCCNC(=O)NC1CCCCC1 HPTJABJPZMULFH-UHFFFAOYSA-N 0.000 description 1
- 241000254137 Cicadidae Species 0.000 description 1
- 241000254173 Coleoptera Species 0.000 description 1
- 241001641310 Cunea Species 0.000 description 1
- 241001243091 Gryllotalpa Species 0.000 description 1
- 241001243087 Gryllotalpidae Species 0.000 description 1
- 241000258937 Hemiptera Species 0.000 description 1
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 241000254023 Locusta Species 0.000 description 1
- 241000254022 Locusta migratoria Species 0.000 description 1
- 241000254043 Melolonthinae Species 0.000 description 1
- 241000254062 Scarabaeidae Species 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000033772 system development Effects 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
- 238000002054 transplantation Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/951—Indexing; Web crawling techniques
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4038—Image mosaicing, e.g. composing plane images from plane sub-images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- General Health & Medical Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Evolutionary Biology (AREA)
- Computational Linguistics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Databases & Information Systems (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to an agricultural pest intelligent identification and positioning method and system, wherein the method comprises the steps of collecting agricultural pest images by adopting a data crawler and an image real-time shooting mode; carrying out data normalization, data enhancement processing and data annotation on the collected agricultural pest images; establishing an orchard pest target detection model based on YOLOv 4; training and evaluating the precision in the orchard pest target detection model based on YOLOv4 to obtain an orchard pest target detection model based on YOLOv4 after evaluation; the method has the advantages that an agricultural pest image is arbitrarily selected and input into the orchard pest target detection model based on YOLOv4 after evaluation for agricultural pest identification and positioning, the method is high in agricultural pest detection precision and detection efficiency, the method is excellent in performance and robustness under various test scenes.
Description
Technical Field
The invention relates to the technical field of agricultural intellectualization and informatization, in particular to an agricultural pest intelligent identification and positioning method and system.
Background
Economic losses of agricultural pests to agriculture are high every year, and accurate and rapid identification and positioning of the agricultural pests are vital to improvement of crop yield, increase of agricultural income and improvement of living standard of people.
Most of the traditional agricultural pest screening methods rely on manual methods, so that the efficiency is low, the precision is difficult to meet the requirements, and the agricultural pest screening methods are difficult to popularize. The existing agricultural pest intelligent identification and positioning method mostly stays at an algorithm simulation level, namely, the method lacks systematic and integrated design, is mostly limited to a laboratory simulation test stage, and lacks performance evaluation and test under an actual application scene.
The artificial intelligence computer vision technology based on technical means such as convolutional neural network, deep learning develops rapidly, can show efficiency and the precision that promotes object discrimination and location, possesses higher reference meaning to agricultural pest intelligent identification and location field.
Disclosure of Invention
The invention aims to solve the technical problem of providing an intelligent agricultural pest identification and positioning method which is high in precision and efficiency and can improve the agricultural economic income.
The invention adopts the technical scheme that an agricultural pest intelligent identification and positioning method comprises the following steps:
(1) collecting agricultural pest images by adopting a data crawler and an image real-time shooting mode;
(2) carrying out data normalization and data enhancement processing on the collected agricultural pest images, carrying out data annotation on the processed agricultural pest images, and forming an agricultural pest image data set by the annotated agricultural pest images;
(3) establishing an orchard pest target detection model based on YOLOv 4; the orchard pest target detection model based on YOLOv4 comprises an input module, a backbone network, a tack module and a head module;
(4) inputting the agricultural pest image data set in the step (2) into an orchard pest target detection model based on YOLOv4 for training to obtain a trained orchard pest target detection model based on YOLOv 4; the specific process is as follows:
(4-1) carrying out mosaic enhancement on the images in the agricultural pest image data set in the step (2) by using an input module of an orchard pest target detection model based on YOLOv4, then randomly selecting any four mosaic-enhanced agricultural pest images for random zooming, random cutting and random distortion treatment, splicing the four processed images to obtain an image sample, and randomly selecting for a plurality of times in a replacement mode to obtain a plurality of image samples;
(4-2) in the main network part, a CSPNet network is adopted to extract the characteristics of the image samples in the step (4-1), then a concat mode is adopted to carry out channel splicing to further obtain characteristic information, and then a 1 x 1 convolution and a 2 x2 pooling network layer are introduced to calculate the obtained characteristic information;
(4-3) in the sock module part, shallow content information and deep semantic information contained in the feature information are mined out in a manner of combining a feature pyramid network and a path aggregation network to obtain a prediction frame;
(4-4) in the head module part, obtaining the distance condition and the intersection condition between the predicted frame and the labeled frame by adopting a CIoU loss function;
(5) performing precision evaluation on the trained orchard pest target detection model based on YOLOv4 to obtain an evaluated orchard pest target detection model based on YOLOv 4;
(6) and randomly selecting an agricultural pest image, and inputting the agricultural pest image into the orchard pest target detection model based on YOLOv4 after evaluation for agricultural pest identification and positioning.
The invention has the beneficial effects that: according to the intelligent agricultural pest identification and positioning method, the detection precision of agricultural pests is high, the detection efficiency is high, the method is excellent in performance under various test scenes, and the robustness is good.
Preferably, in step (5), the precision of the orchard pest target detection model based on Yolov4 after training is evaluated by using a classification precision mAP and a positioning precision evaluation index IoU, wherein the classification precision mAP is an average value of detection precision of each single agricultural pest category, and the positioning precision evaluation index IoU is expressed mathematically as:
IoU, the larger the value, the higher the ratio of the overlapping area of the predicted frame and the labeled frame, and the higher the positioning accuracy of the orchard pest target detection model based on YOLOv 4.
The utility model provides an agricultural pest intelligent recognition and positioning system, includes image acquisition module, image preprocessing module, based on YOLOv4 orchard pest target detection model's pest target detection module and information display storage module, image acquisition module includes local data acquisition module and online data acquisition module, local data acquisition module gathers image data through the mode of web crawler, and online data acquisition module gathers real-time image data through industrial camera, pest target detection module includes orientation module and classification module, information display storage module includes user interface and information storage module. By adopting the intelligent agricultural pest identification and positioning system, the detection precision of agricultural pests is high, the detection efficiency is also high, and the system integration level is high.
Drawings
FIG. 1 is a flow chart of an intelligent agricultural pest identification and location method of the present invention;
FIG. 2 is a schematic structural diagram of an orchard pest target detection model based on YOLOv4 in the invention;
FIG. 3 is a schematic diagram of a positioning accuracy evaluation index IoU according to the present invention;
FIG. 4 is a schematic diagram of a framework of an intelligent agricultural pest identification and location system according to the present invention;
FIG. 5 is a schematic diagram of an actual test using an intelligent agricultural pest identification and location method according to the present invention;
Detailed Description
The invention is further described below with reference to the accompanying drawings in combination with specific embodiments so that those skilled in the art can practice the invention with reference to the description, and the scope of the invention is not limited to the specific embodiments.
The invention relates to an intelligent identification and positioning method for agricultural pests, which comprises the following steps:
(1) collecting agricultural pest images by adopting a data crawler and an image real-time shooting mode; the wired equipment adopted for real-time image shooting is an industrial undistorted camera HF-869, and the data connection mode between the wired equipment and the computer host is USB 2.0; the adopted data Stream reading mode of the wireless image acquisition equipment reads the Real-Time video Stream by virtue of an RTSP (Real-Time streaming Protocol) Protocol; under the condition of simultaneously supporting two network modes of Wireless (WLAN) image data transmission and wired data transmission, various static and dynamic detection modes such as pictures and videos are supported;
(2) carrying out data normalization and data enhancement processing on the collected agricultural pest images, carrying out data annotation on the processed agricultural pest images, and forming an agricultural pest image data set by the annotated agricultural pest images; the data normalization process includes size normalization (320 × 320), channel number normalization (3 channels), and format normalization (. JPG); the data enhancement processing comprises a plurality of technical modes such as size rotation, content stretching, noise transformation, fuzzy transformation and the like; the data labeling covers the basic content of the image information (including image name, width w, height h, channel number, storage position information and the like), the category information of agricultural pests and the position coordinate information of the agricultural pests (center point coordinate x, center point coordinate y, positioning frame width information w and positioning frame height h);
(3) establishing an orchard pest target detection model based on YOLOv 4; the orchard pest target detection model based on YOLOv4 comprises an input module, a backbone network, a tack module and a head module;
(4) inputting the agricultural pest image data set in the step (2) into an orchard pest target detection model based on YOLOv4 for training to obtain a trained orchard pest target detection model based on YOLOv 4; the specific process is as follows:
(4-1) carrying out mosaic enhancement on the images in the agricultural pest image data set in the step (2) by using an input module of an orchard pest target detection model based on YOLOv4, then randomly selecting any four mosaic-enhanced agricultural pest images for random zooming, random cutting and random distortion treatment, splicing the four processed images to obtain an image sample, and randomly selecting for a plurality of times in a replacement mode to obtain a plurality of image samples;
(4-2) in the main network part, a CSPNet network is adopted to perform efficient feature extraction on a plurality of image samples in the step (4-1), then a concat mode is adopted to perform channel splicing to further obtain richer feature information, and then a 1 x 1 convolution and a 2 x2 pooling network layer are introduced to calculate the obtained feature information, so that the calculated amount can be reduced, and the calculation speed of the model can be improved;
(4-3) in the sock module part, shallow content information and deep semantic information contained in the feature information are mined out in a manner of combining a feature pyramid network and a path aggregation network to obtain a prediction frame;
(4-4) in the head module part, obtaining the distance condition and the intersection condition between the predicted frame and the labeled frame by adopting a CIoU loss function;
a total of 7 types of agricultural pests are set as target agricultural pests, namely, cicadas (hemiptera cicadae), mole crickets (crickelidae, Gryllotalpa spp), scarab beetles (coleoptera chafer superfamily, scarabaeaoidea), eastern asian migratory locust (locustaceae, Locusta migoritoria), longicorn beetles (coeptera longitaceae, Cerambycidae), gibba giralda (budestidae, coleoptera), and white moth (hyphalaris cupriaceae, hyphena cunea), and basic parameter settings are as follows during model training: the input image size is 320 × 320 × 4; the batch size is set to 64; sub-batch set to 16; network trigger momentum is set to 0.949; the maximum number of iterations is 14000; the learning rate strategy is steps, the initial value is 0.001, scale is set to be 0.1, and the values of two steps of the learning rate variation are 11200 and 12600 respectively; the angular rotation is set to 0;
(5) carrying out precision evaluation on the trained orchard pest target detection model based on YOLOv4 to obtain an orchard pest target detection model based on YOLOv 4;
(6) and randomly selecting an agricultural pest image, and inputting the agricultural pest image into the orchard pest target detection model based on YOLOv4 after testing to identify and position agricultural pests.
According to the intelligent agricultural pest identification and positioning method, the detection precision of agricultural pests is high, the detection efficiency is high, the method is excellent in performance under various test scenes, and the robustness is good.
In the step (5), precision evaluation is performed on the orchard pest target detection model based on YOLOv4 after training by adopting a classification precision mAP and a positioning precision evaluation index IoU, wherein the classification precision mAP is an average value of detection precision rates of agricultural pests of each single category, and the positioning precision evaluation index IoU is expressed mathematically as:
IoU, the larger the value, the higher the ratio of the overlapping area of the predicted frame and the labeled frame, and the higher the positioning accuracy of the orchard pest target detection model based on YOLOv 4.
The agricultural pest detection precision index in the agricultural pest intelligent identification and positioning method needs to be more than 90%.
The performance of the orchard pest target detection model based on Yolov4 was tested under the following three test scenarios: agricultural pests are of the same category and are single targets; the agricultural pests are of the same category and have multiple targets; and the agricultural pests have various categories, multiple targets and the like.
The utility model provides an agricultural pest intelligent recognition and positioning system, includes image acquisition module, image preprocessing module, based on YOLOv4 orchard pest target detection model's pest target detection module and information display storage module, image acquisition module includes local data acquisition module and online data acquisition module, local data acquisition module gathers image data through the mode of web crawler, and online data acquisition module gathers real-time image data through industrial camera, pest target detection module includes orientation module and classification module, information display storage module includes user interface and information storage module. By adopting the intelligent agricultural pest identification and positioning system, the detection precision of agricultural pests is high, the detection efficiency is also high, and the system integration level is high.
GPU operation is adopted to replace CPU operation so as to realize efficient establishment and training of orchard pest target detection models based on YOLOv4, and development environments comprise AMD Ryzen 53600, NVIDIA RTX2070 Super, 16GB, Windows 10, cuDNN 7.6, Visual Studio 2017, CUDA 10.0, OpenCV 3.4 and C + +. The interface application program development framework is QT 5.9.6, and cross-platform operation (Windows, Linux, embedded equipment and the like) is supported, so that the modularization degree is high, and the reusability is good; the document resources are rich.
Above-mentioned agricultural pest intelligent recognition and positioning system adopts the modularized design theory, divide into four module component parts with entire system: the agricultural pest control system comprises an image acquisition part, a data preprocessing part, an agricultural pest intelligent identification and positioning part and an information storage and display part. Each part is independent, can be designed separately, is convenient for system development and maintenance. Wherein, the transplantation and the embedding of the agricultural pest intelligent identification and positioning model are realized by adopting a dynamic link library (dynamic link library) compiling mode. The information storage and display part mainly comprises a window display area, a function selection area and other page areas.
The image acquisition module comprises two data contents of local data and online data, and supports a plurality of common picture formats such as jpg, png, bmp and the like and a plurality of common video formats such as mp4, avi and the like. The data transmission mode of the wired data acquisition equipment is USB 2.0, the acquisition equipment of the wireless image data is a Haokawav video camera DS-2SC3Q120IY-T/W, and the transmission mode is RTSP real-time video stream. In the data preprocessing module, in order to improve the detection effect of the agricultural pest intelligent identification and positioning system, image data are preprocessed in the modes of data normalization, cutting, brightness adjustment and the like. In the intelligent agricultural pest identifying and positioning algorithm part, in the compiling process of the dynamic link library, the version of Windows SDK is 10.0.17763.0, and the compiling mode is Release x 64. After compiling is completed, obtaining a dynamic link library file of a YOLO algorithm: yolo _ cpp _ dll.lib, yolo _ cpp _ dll.dll, pthread gc2.dll, pthread vc2.dll and opencv _ work 340.dll. The memory requirement is lower; the dynamic link library file and the execution file are mutually independent, and the expansibility and the maintainability of the system are better; the modularization degree is higher. In the information storage and display module, the menu bar includes three items: detecting demand, storing directory and version information. The detection requirements are as follows: the local images, the local videos and the local wired cameras are used for realizing online video detection, and the version information of online video detection software is realized by calling the IP camera through a local area network, so that the contents of the version number, the release time, the detection precision of the current algorithm model, the detection category, the detection target range and the like of the intelligent orchard pest detection system are recorded.
Claims (3)
1. An intelligent agricultural pest identification and positioning method is characterized by comprising the following steps: the method comprises the following steps:
(1) collecting agricultural pest images by adopting a data crawler and an image real-time shooting mode;
(2) carrying out data normalization and data enhancement processing on the collected agricultural pest images, carrying out data annotation on the processed agricultural pest images, and forming an agricultural pest image data set by the annotated agricultural pest images;
(3) establishing an orchard pest target detection model based on YOLOv 4; the orchard pest target detection model based on YOLOv4 comprises an input module, a backbone network, a tack module and a head module;
(4) inputting the agricultural pest image data set in the step (2) into an orchard pest target detection model based on YOLOv4 for training to obtain a trained orchard pest target detection model based on YOLOv 4; the specific process is as follows:
(4-1) carrying out mosaic enhancement on the images in the agricultural pest image data set in the step (2) by using an input module of an orchard pest target detection model based on YOLOv4, then randomly selecting any four mosaic-enhanced agricultural pest images for random zooming, random cutting and random distortion treatment, splicing the four processed images to obtain an image sample, and randomly selecting for a plurality of times in a replacement mode to obtain a plurality of image samples;
(4-2) in the main network part, a CSPNet network is adopted to extract the characteristics of the image samples in the step (4-1), then a concat mode is adopted to carry out channel splicing to further obtain characteristic information, and then a 1 x 1 convolution and a 2 x2 pooling network layer are introduced to calculate the obtained characteristic information;
(4-3) in the sock module part, shallow content information and deep semantic information contained in the feature information are mined out in a manner of combining a feature pyramid network and a path aggregation network to obtain a prediction frame;
(4-4) in the head module part, obtaining the distance condition and the intersection condition between the predicted frame and the labeled frame by adopting a CIoU loss function;
(5) performing precision evaluation on the trained orchard pest target detection model based on YOLOv4 to obtain an evaluated orchard pest target detection model based on YOLOv 4;
(6) and randomly selecting an agricultural pest image, and inputting the agricultural pest image into the orchard pest target detection model based on YOLOv4 after evaluation for agricultural pest identification and positioning.
2. The intelligent agricultural pest identification and positioning method according to claim 1, wherein the method comprises the following steps: in step (5), classification accuracy mAP and positioning accuracy evaluation index IoU are adopted to detect the trained orchard pest target detection model based on YOLOv4And performing precision evaluation, wherein the classification precision mAP is an average value of the detection precision rates of the agricultural pests of the single categories, and the positioning precision evaluation index IoU is expressed by the following mathematics:
IoU, the larger the value, the higher the ratio of the overlapping area of the predicted frame and the labeled frame, and the higher the positioning accuracy of the orchard pest target detection model based on YOLOv 4.
3. An intelligent agricultural pest identification and positioning system for implementing the intelligent agricultural pest identification and positioning method of any one of claims 1 to 2, wherein the intelligent agricultural pest identification and positioning system comprises: the orchard pest target detection system comprises an image acquisition module, an image preprocessing module, a pest target detection module based on an orchard pest target detection model of YOLOv4 and an information display storage module, wherein the image acquisition module comprises a local data acquisition module and an online data acquisition module, the local data acquisition module acquires image data in a network crawler mode, the online data acquisition module acquires real-time image data through an industrial camera, the pest target detection module comprises a positioning module and a classification module, and the information display storage module comprises a user interface and an information storage module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110994521.3A CN113744226A (en) | 2021-08-27 | 2021-08-27 | Intelligent agricultural pest identification and positioning method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110994521.3A CN113744226A (en) | 2021-08-27 | 2021-08-27 | Intelligent agricultural pest identification and positioning method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113744226A true CN113744226A (en) | 2021-12-03 |
Family
ID=78733310
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110994521.3A Pending CN113744226A (en) | 2021-08-27 | 2021-08-27 | Intelligent agricultural pest identification and positioning method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113744226A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114926720A (en) * | 2022-05-12 | 2022-08-19 | 中国农业大学 | Method for identifying multiple agricultural pests based on target detection technology and related equipment |
| CN116485796A (en) * | 2023-06-19 | 2023-07-25 | 闽都创新实验室 | Pest detection method, pest detection device, electronic equipment and storage medium |
| CN117237814A (en) * | 2023-11-14 | 2023-12-15 | 四川农业大学 | Large-scale orchard insect condition monitoring method based on attention mechanism optimization |
| CN117649310A (en) * | 2023-11-21 | 2024-03-05 | 江苏一心寰宇生物科技有限公司 | Insect pest information monitoring method and system based on Internet of things technology |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112241762A (en) * | 2020-10-19 | 2021-01-19 | 吉林大学 | Fine-grained identification method for pest and disease damage image classification |
| CN112668490A (en) * | 2020-12-30 | 2021-04-16 | 浙江托普云农科技股份有限公司 | Yolov 4-based pest detection method, system, device and readable storage medium |
| CN112861767A (en) * | 2021-02-26 | 2021-05-28 | 北京农业信息技术研究中心 | Small-volume pest detection method and system on pest sticking plate image |
-
2021
- 2021-08-27 CN CN202110994521.3A patent/CN113744226A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112241762A (en) * | 2020-10-19 | 2021-01-19 | 吉林大学 | Fine-grained identification method for pest and disease damage image classification |
| CN112668490A (en) * | 2020-12-30 | 2021-04-16 | 浙江托普云农科技股份有限公司 | Yolov 4-based pest detection method, system, device and readable storage medium |
| CN112861767A (en) * | 2021-02-26 | 2021-05-28 | 北京农业信息技术研究中心 | Small-volume pest detection method and system on pest sticking plate image |
Non-Patent Citations (1)
| Title |
|---|
| ALEXEY BOCHKOVSKIY: "YOLOv4: Optimal Speed and Accuracy of Object Detection", 《ARXIV》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114926720A (en) * | 2022-05-12 | 2022-08-19 | 中国农业大学 | Method for identifying multiple agricultural pests based on target detection technology and related equipment |
| CN116485796A (en) * | 2023-06-19 | 2023-07-25 | 闽都创新实验室 | Pest detection method, pest detection device, electronic equipment and storage medium |
| CN116485796B (en) * | 2023-06-19 | 2023-09-08 | 闽都创新实验室 | Pest detection method, pest detection device, electronic equipment and storage medium |
| CN117237814A (en) * | 2023-11-14 | 2023-12-15 | 四川农业大学 | Large-scale orchard insect condition monitoring method based on attention mechanism optimization |
| CN117237814B (en) * | 2023-11-14 | 2024-02-20 | 四川农业大学 | Large-scale orchard insect condition monitoring method based on attention mechanism optimization |
| CN117649310A (en) * | 2023-11-21 | 2024-03-05 | 江苏一心寰宇生物科技有限公司 | Insect pest information monitoring method and system based on Internet of things technology |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113744226A (en) | Intelligent agricultural pest identification and positioning method and system | |
| CN111709489B (en) | Citrus identification method based on improved YOLOv4 | |
| Chen et al. | Detecting citrus in orchard environment by using improved YOLOv4 | |
| CN113065558A (en) | Lightweight small target detection method combined with attention mechanism | |
| Deng et al. | Deep learning-based automatic detection of productive tillers in rice | |
| CN108596102B (en) | RGB-D-based indoor scene object segmentation classifier construction method | |
| CN110223349A (en) | A kind of picking independent positioning method | |
| CN113191334B (en) | Plant canopy dense leaf counting method based on improved CenterNet | |
| CN114359727A (en) | Tea disease identification method and system based on lightweight optimization Yolo v4 | |
| CN110827312A (en) | Learning method based on cooperative visual attention neural network | |
| CN115661544B (en) | Spinach seedling water stress grade classification system and method based on N-MobileNetXt | |
| CN115984698A (en) | Litchi fruit growing period identification method based on improved YOLOv5 | |
| CN114140665A (en) | Dense small target detection method based on improved YOLOv5 | |
| CN113516102A (en) | Deep learning parabolic behavior detection method based on video | |
| CN113435254A (en) | Sentinel second image-based farmland deep learning extraction method | |
| Shuai et al. | An improved YOLOv5-based method for multi-species tea shoot detection and picking point location in complex backgrounds | |
| CN117830788A (en) | Image target detection method for multi-source information fusion | |
| CN112084913B (en) | End-to-end human body detection and attribute identification method | |
| Zhang et al. | A tea bud segmentation, detection and picking point localization based on the MDY7-3PTB model | |
| CN117079125A (en) | Kiwi fruit pollination flower identification method based on improved YOLOv5 | |
| CN116805415A (en) | Cage broiler health status identification method based on lightweight improved YOLOv5 | |
| Chin et al. | Plant Disease Detection and Classification Using Deep Learning Methods: A Comparison Study | |
| CN111223125A (en) | Python environment-based target motion video tracking method | |
| CN114066920B (en) | Harvester visual navigation method and system based on improved Segnet image segmentation | |
| CN116189076A (en) | Observation and identification system and method for bird observation station |
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 |