CN117636169B - Forest patrol method and system - Google Patents
Forest patrol method and system Download PDFInfo
- Publication number
- CN117636169B CN117636169B CN202311653781.XA CN202311653781A CN117636169B CN 117636169 B CN117636169 B CN 117636169B CN 202311653781 A CN202311653781 A CN 202311653781A CN 117636169 B CN117636169 B CN 117636169B
- Authority
- CN
- China
- Prior art keywords
- forest
- report
- data
- module
- fire
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000012544 monitoring process Methods 0.000 claims abstract description 100
- 238000007726 management method Methods 0.000 claims abstract description 72
- 238000005516 engineering process Methods 0.000 claims abstract description 67
- 241001465754 Metazoa Species 0.000 claims abstract description 56
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 56
- 238000004458 analytical method Methods 0.000 claims abstract description 53
- 238000004140 cleaning Methods 0.000 claims abstract description 23
- 238000013135 deep learning Methods 0.000 claims abstract description 20
- 230000010354 integration Effects 0.000 claims abstract description 17
- 230000003068 static effect Effects 0.000 claims abstract description 14
- 238000013473 artificial intelligence Methods 0.000 claims abstract description 7
- 238000012545 processing Methods 0.000 claims description 73
- 230000002159 abnormal effect Effects 0.000 claims description 52
- 230000004044 response Effects 0.000 claims description 39
- 230000006378 damage Effects 0.000 claims description 24
- 230000000694 effects Effects 0.000 claims description 19
- 201000010099 disease Diseases 0.000 claims description 17
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 claims description 17
- 241000238631 Hexapoda Species 0.000 claims description 16
- 241000607479 Yersinia pestis Species 0.000 claims description 16
- 238000007405 data analysis Methods 0.000 claims description 14
- 238000001914 filtration Methods 0.000 claims description 11
- 239000002243 precursor Substances 0.000 claims description 11
- 238000001514 detection method Methods 0.000 claims description 10
- 238000012549 training Methods 0.000 claims description 10
- 230000009471 action Effects 0.000 claims description 9
- 238000005457 optimization Methods 0.000 claims description 9
- 239000000779 smoke Substances 0.000 claims description 8
- 238000006243 chemical reaction Methods 0.000 claims description 6
- 238000010191 image analysis Methods 0.000 claims description 6
- 238000005065 mining Methods 0.000 claims description 6
- 238000012300 Sequence Analysis Methods 0.000 claims description 4
- 230000005856 abnormality Effects 0.000 claims description 4
- 230000004888 barrier function Effects 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000007418 data mining Methods 0.000 claims description 4
- 238000013136 deep learning model Methods 0.000 claims description 4
- 230000002068 genetic effect Effects 0.000 claims description 4
- 230000010365 information processing Effects 0.000 claims description 4
- 238000010801 machine learning Methods 0.000 claims description 4
- 238000012706 support-vector machine Methods 0.000 claims description 4
- 238000012876 topography Methods 0.000 claims description 4
- 238000013526 transfer learning Methods 0.000 claims description 4
- 230000005540 biological transmission Effects 0.000 claims description 3
- 238000013527 convolutional neural network Methods 0.000 claims description 3
- 238000003066 decision tree Methods 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 238000003058 natural language processing Methods 0.000 claims description 3
- 238000003062 neural network model Methods 0.000 claims description 3
- 230000009467 reduction Effects 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 230000009286 beneficial effect Effects 0.000 abstract description 5
- 238000011156 evaluation Methods 0.000 abstract description 3
- 230000008569 process Effects 0.000 description 5
- 230000000007 visual effect Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012806 monitoring device Methods 0.000 description 2
- 230000002265 prevention Effects 0.000 description 2
- 241000894007 species Species 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000003306 harvesting Methods 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 230000000474 nursing effect Effects 0.000 description 1
- 230000008092 positive effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- -1 temperature jump Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/188—Vegetation
-
- 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
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/251—Fusion techniques of input or preprocessed data
-
- 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
- G06N3/096—Transfer learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/02—Agriculture; Fishing; Forestry; Mining
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/10—Image acquisition
- G06V10/12—Details of acquisition arrangements; Constructional details thereof
- G06V10/14—Optical characteristics of the device performing the acquisition or on the illumination arrangements
-
- 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
-
- 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
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/94—Hardware or software architectures specially adapted for image or video understanding
- G06V10/95—Hardware or software architectures specially adapted for image or video understanding structured as a network, e.g. client-server architectures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/17—Terrestrial scenes taken from planes or by drones
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B17/00—Fire alarms; Alarms responsive to explosion
- G08B17/005—Fire alarms; Alarms responsive to explosion for forest fires, e.g. detecting fires spread over a large or outdoors area
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B17/00—Fire alarms; Alarms responsive to explosion
- G08B17/12—Actuation by presence of radiation or particles, e.g. of infrared radiation or of ions
- G08B17/125—Actuation by presence of radiation or particles, e.g. of infrared radiation or of ions by using a video camera to detect fire or smoke
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B25/00—Alarm systems in which the location of the alarm condition is signalled to a central station, e.g. fire or police telegraphic systems
- G08B25/01—Alarm systems in which the location of the alarm condition is signalled to a central station, e.g. fire or police telegraphic systems characterised by the transmission medium
- G08B25/08—Alarm systems in which the location of the alarm condition is signalled to a central station, e.g. fire or police telegraphic systems characterised by the transmission medium using communication transmission lines
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B31/00—Predictive alarm systems characterised by extrapolation or other computation using updated historic data
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Computing Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Emergency Management (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Databases & Information Systems (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Evolutionary Biology (AREA)
- Biodiversity & Conservation Biology (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Marine Sciences & Fisheries (AREA)
- Animal Husbandry (AREA)
- Strategic Management (AREA)
- Agronomy & Crop Science (AREA)
- Mining & Mineral Resources (AREA)
- Primary Health Care (AREA)
- Tourism & Hospitality (AREA)
- Human Resources & Organizations (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Mathematical Physics (AREA)
Abstract
The invention relates to the technical field of forest patrol methods, in particular to a forest patrol method and a forest patrol system, which comprise the following steps: and carrying out omnibearing shooting on the forest by using the unmanned aerial vehicle or the static monitoring camera according to a set path and a set time sequence to acquire forest image data. According to the invention, the unmanned aerial vehicle and the monitoring camera are utilized to acquire forest image data, and the forest image data is processed through the deep learning and image recognition algorithm, so that forest management and protection can be performed in a large scale and efficiently, the coverage rate and accuracy of forest monitoring are improved, the artificial intelligence and cloud computing technology is utilized to provide data cleaning, integration and analysis, and real-time reporting is provided, so that a forest manager can more easily know and master the real-time state of the forest, the efficiency and accuracy of management decision are improved, and the wild animal monitoring is performed by combining the on-site sound monitoring equipment and the sound recognition technology, thereby being beneficial to forest ecological protection, and enriching the diversity evaluation and animal resource management of forest ecology.
Description
Technical Field
The invention relates to the technical field of forest patrol methods, in particular to a forest patrol method and a forest patrol system.
Background
The forest patrol method is a series of monitoring, patrol and intervention measures adopted for protecting and managing forest resources, and comprises the steps of periodically patrol forest areas, monitoring forest conditions in real time by using a monitoring technology, adopting fireproof measures, carrying out pest and disease management and striking illegal activities. Forest patrolling personnel check the state and potential problems of the forest through patrol and nursing, and take action in time. The monitoring technology provides information such as forest coverage, fire risk and the like, and helps to find problems. Fire protection measures include establishing fire lines, monitoring fire hazard areas, and extinguishing equipment to prevent and control fires. For the pest and disease damage problem, the patrolling personnel check and take proper management measures. The attack on illegal activities is the key to protecting forests, including stopping illegal felling, mishunting, harvesting plants, etc. Aims to protect the integrity and sustainable utilization of the forest ecosystem, maintain the environmental balance and protect the living environment of rare endangered species.
In the actual use process of the forest patrol method, the existing forest patrol method mainly depends on manpower for patrol, and has limited coverage area and low efficiency. The current method is mainly used for evaluating risks such as diseases and insect pests, fire disasters and the like through visual observation and collection of individual samples, and the method is easy to cause information omission to a certain extent, so that comprehensive and accurate information is difficult to achieve. The existing method often lacks support of real-time data in operation, cannot realize real-time supervision of forest resource conditions, and is also greatly limited in coping with emergency. In the existing method, the monitoring of wild animals mostly depends on traditional trace tracking and visual observation, and the biodiversity condition is difficult to fully understand.
Disclosure of Invention
The invention aims to solve the defects in the prior art and provides a forest patrol method and a forest patrol system.
In order to achieve the above purpose, the present invention adopts the following technical scheme: a forest patrol method, comprising the steps of:
S1: carrying out omnibearing shooting on a forest by using an unmanned aerial vehicle or a static monitoring camera according to a set path and a set time sequence to acquire forest image data;
s2: processing the forest image data by utilizing a deep learning and image recognition algorithm, detecting whether abnormal phenomena including fire risks and diseases and insect pests exist or not, and giving an abnormal report;
S3: analyzing satellite remote sensing images and geographic information data by combining an artificial intelligent algorithm, monitoring forest coverage rate and damage degree in real time, and outputting a forest state report;
S4: according to the forest status report and the abnormality report, fire risk precursors are monitored in real time through fire monitoring sensors deployed in the forest, and when a fire risk early warning signal is detected, an alarm is immediately sent to a related department and fire extinguishing actions are guided to generate a fire risk response report;
s5: sending the forest image data, the abnormal report, the forest state report and the fire hazard response report to a cloud computing platform for data cleaning, integration and analysis, and enabling the cloud computing platform to execute automatic processing, provide a patrol report and decision support and output a cloud processing report;
s6: based on the cloud processing report, processing and analyzing by utilizing a big data analysis technology, so as to optimize a forest resource management strategy and ecological protection measures and generate an ecological management strategy report;
S7: and combining on-site sound monitoring equipment and sound identification technology, and carrying out directional wild animal monitoring according to the ecological management strategy report to obtain a wild animal monitoring report.
As a further scheme of the invention, the unmanned aerial vehicle or the static monitoring camera is utilized to carry out omnibearing shooting on the forest according to a set path and a set time sequence, and the step of acquiring forest image data comprises the following steps:
s101: optimizing parameter settings of the unmanned aerial vehicle or the monitoring camera by adopting a genetic algorithm, including ISO, exposure and white balance, and generating a device configuration file;
S102: applying an A-algorithm or a Dijkstra algorithm, planning an optimal shooting path of the unmanned aerial vehicle or the monitoring camera according to the topography and the ecological barrier of the forest, and generating a path scheme;
S103: and ensuring the stability of the unmanned aerial vehicle or the camera based on the PID controller, automatically starting the equipment at the appointed time according to the equipment configuration file and the path scheme, and performing omnibearing shooting to generate forest image data.
As a further scheme of the invention, the forest image data is processed by utilizing a deep learning and image recognition algorithm, whether abnormal phenomena including fire danger and plant diseases and insect pests exist or not is detected, and the step of giving an abnormal report comprises the following specific steps:
s201: performing quality enhancement and cleaning on the forest image data by using a self-adaptive histogram equalization and noise filtering technology to obtain a preprocessed image data set;
s202: performing deep learning model training based on forest images on the preprocessed image dataset by adopting a transfer learning technology on the basis of a pre-trained ResNet or VGG network to obtain a training model;
s203: and identifying the abnormal phenomenon of the preprocessed image dataset by using a YOLO or SSD real-time target detection frame, and compiling an abnormal report according to the result.
As a further scheme of the invention, satellite remote sensing images and geographic information data are analyzed by combining an artificial intelligence algorithm, the forest coverage rate and the damage degree are monitored in real time, and the step of outputting a forest status report is specifically as follows:
S301: capturing the latest remote sensing image of a forest by utilizing Landsat satellite resources, and simultaneously carrying out data annotation to construct a remote sensing image database;
s302: analyzing the characteristics of a remote sensing image by adopting a U-Net structure, and carrying out learning and feature extraction on the remote sensing image database to form a remote sensing image analysis model;
S303: and inputting a real-time satellite remote sensing image into the remote sensing image analysis model by combining an edge calculation and a quick response technology, and obtaining the forest coverage rate and the damage degree in real time and generating a forest status report according to the forest coverage rate and the damage degree.
According to the forest status report and the abnormality report, the fire monitoring sensor deployed in the forest is used for monitoring the fire precursor in real time, and when a fire early warning signal is detected, an alarm is immediately sent to a relevant department and the fire extinguishing action is guided, so that the fire response report is generated by the following steps:
S401: determining key and inflammable areas of a forest by adopting a k-means algorithm, and disposing temperature and smoke fire monitoring sensors in a targeted manner to form a sensor network;
S402: applying a Kalman filtering algorithm to fuse and optimize real-time signals from a sensor network to generate a sensor data stream;
S403: and analyzing the sensor data flow according to the precursors including temperature jump and smoke concentration increase based on the classifier of the support vector machine, immediately sending an alarm to related departments once the fire hazard early warning is detected, and forming a fire hazard response report.
As a further scheme of the invention, based on the forest image data, the abnormal report, the forest status report and the fire hazard response report, the forest image data, the abnormal report, the forest status report and the fire hazard response report are sent to a cloud computing platform for data cleaning, integration and analysis, the cloud computing platform executes automatic processing and provides a patrol report and decision support, and the step of outputting a cloud processing report specifically comprises the following steps:
s501: uploading the forest image data, the abnormal report, the forest state report and the fire hazard response report to a cloud computing platform by using a distributed uploading technology, and establishing an original data set;
S502: performing preliminary screening on error, missing and redundant data in the original data set by using a data quality detection algorithm, and processing the missing data by using an interpolation algorithm to obtain a cleaned data set;
S503: classifying and marking the data by using a data conversion and mapping technology, integrating the data in the cleaned data set according to time, place and type to generate an integrated data set;
s504: deep analysis is carried out on the integrated data set by utilizing a machine learning and statistics method, the data characteristics are analyzed and classified, potential problems and risks of forests are detected, and a preliminary analysis report is output;
S505: according to the preliminary analysis report, the cloud computing platform generates an optimal proposal and strategy for a decision maker by utilizing a decision tree or a neural network model, and outputs a patrol proposal report;
S506: and integrating the integrated data set and the patrol proposal report by using a report generating tool to be used as a cloud processing report.
As a further scheme of the invention, based on the cloud processing report, the big data analysis technology is utilized for processing and analyzing, thereby optimizing the forest resource management strategy and the ecological protection measures, and the steps for generating the ecological management strategy report are specifically as follows:
S601: analyzing and extracting key data and suggestions in the cloud processing report by using a natural language processing technology to generate key information data;
s602: based on the key information data, applying an association rule learning and clustering algorithm to carry out depth data mining, and outputting mining results;
s603: based on the mining result, optimizing a forest resource management strategy and ecological protection measures by utilizing a multi-objective optimization algorithm and combining current resources and constraints to generate an optimized strategy;
S604: and converting the optimized strategy into a specific operation guideline and suggestion by using a strategy generation tool, and generating an ecological management strategy report.
As a further scheme of the invention, by combining on-site sound monitoring equipment and sound recognition technology, the method for carrying out directional wild animal monitoring according to the ecological management strategy report comprises the following steps:
S701: according to the suggestions in the ecological management policy report, deploying sound equipment, periodically collecting sound data in a forest from the sound equipment by using a wireless transmission technology, and generating an original sound data set;
S702: using a fast Fourier transform and filtering technology to perform noise reduction and format conversion on the original sound data set to obtain a cleaned sound data set;
s703: analyzing the cleaned sound data set by utilizing a sound recognition technology, extracting and classifying sound characteristics by using a convolutional neural network in deep learning, and outputting an animal sound recognition result;
s704: based on the animal voice recognition result, carrying out statistics and trend prediction on animal activities by utilizing a time sequence analysis and pattern matching technology, and generating an animal activity analysis report;
S705: and generating a wild animal monitoring report by using a report generating tool and combining the animal voice recognition result and the animal activity analysis report.
The forest patrol system consists of an image acquisition module, an image recognition module, a remote sensing and GIS analysis module, a fire hazard early warning module, a cloud data processing module and an ecological management strategy module;
the image acquisition module shoots a forest according to a preset path and a preset time sequence by using the unmanned aerial vehicle and the static monitoring camera so as to acquire omnibearing forest image data;
the image recognition module: deep learning and image recognition processing are carried out on the omnibearing forest image data, abnormal phenomena are detected in real time, and an abnormal report is generated;
The remote sensing and GIS analysis module is combined with an artificial intelligence algorithm to analyze satellite remote sensing images and geographic information data, detect forest coverage rate and damage degree and output a forest status report;
The fire early warning module deploys fire monitoring sensors in the key areas according to the forest state report and the abnormal report, and when a fire early warning signal is detected, the fire early warning module gives an alarm to related departments and generates a fire response report;
The cloud data processing module uploads the omnibearing forest image data, the abnormal report, the forest status report and the fire hazard response report to the cloud computing platform for data integration, cleaning and analysis to generate a comprehensive patrol report and decision support information, and outputs a cloud processing report;
The ecological management strategy module is combined with cloud processing reporting, optimizes forest resource management strategies and ecological protection measures by using a big data analysis technology, and simultaneously monitors wild animals by combining with a sound monitoring technology to generate an ecological management strategy report and a wild animal monitoring report.
As a further scheme of the invention, the image acquisition module comprises an unmanned aerial vehicle sub-module, a static camera sub-module and a time sequence shooting control sub-module;
The image recognition module comprises a deep learning sub-module, a fire risk recognition sub-module and a plant disease and insect pest monitoring sub-module;
the remote sensing and GIS analysis module comprises a satellite remote sensing sub-module, a geographic information processing sub-module and a forest coverage analysis sub-module;
the fire hazard early warning module comprises a sensor deployment sub-module, a real-time fire hazard monitoring sub-module and an alarm sending sub-module;
the cloud data processing module comprises a data uploading sub-module, a data cleaning sub-module and a decision support generating sub-module;
The ecological management strategy module comprises a big data strategy optimization sub-module, a sound monitoring sub-module and a wild animal identification sub-module.
Compared with the prior art, the invention has the advantages and positive effects that:
According to the invention, the unmanned aerial vehicle and the monitoring camera are utilized to acquire forest image data, and the forest image data is processed through the deep learning and image recognition algorithm, so that forest management and protection can be performed in a large scale and high efficiency, and the coverage rate and accuracy of forest monitoring are improved. By using artificial intelligence and cloud computing technology, data cleaning, integration and analysis are provided, real-time reporting is provided, forest managers can more easily know and master the real-time state of the forest, and management decision efficiency and accuracy are improved. The wild animal is monitored by combining the on-site sound monitoring equipment and the sound recognition technology, which is helpful for protecting forest ecology and enriches diversity evaluation of forest ecology and animal resource management.
Drawings
FIG. 1 is a schematic workflow diagram of the present invention;
FIG. 2 is a S1 refinement flowchart of the present invention;
FIG. 3 is a S2 refinement flowchart of the present invention;
FIG. 4 is a S3 refinement flowchart of the present invention;
FIG. 5 is a S4 refinement flowchart of the present invention;
FIG. 6 is a S5 refinement flowchart of the present invention;
FIG. 7 is a S6 refinement flowchart of the present invention;
FIG. 8 is a S7 refinement flowchart of the present invention;
FIG. 9 is a system flow diagram of the present invention;
fig. 10 is a system block diagram of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
In the description of the present invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," and the like indicate orientations or positional relationships based on the orientation or positional relationships shown in the drawings, merely to facilitate describing the present invention and simplify the description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be configured and operated in a specific orientation, and therefore should not be construed as limiting the present invention. Furthermore, in the description of the present invention, the meaning of "a plurality" is two or more, unless explicitly defined otherwise.
Example 1
Referring to fig. 1, the present invention provides a technical solution: a forest patrol method, comprising the steps of:
S1: carrying out omnibearing shooting on a forest by using an unmanned aerial vehicle or a static monitoring camera according to a set path and a set time sequence to acquire forest image data;
S2: processing forest image data by utilizing a deep learning and image recognition algorithm, detecting whether abnormal phenomena including fire danger and plant diseases and insect pests exist or not, and giving an abnormal report;
S3: analyzing satellite remote sensing images and geographic information data by combining an artificial intelligent algorithm, monitoring forest coverage rate and damage degree in real time, and outputting a forest state report;
S4: according to the forest state report and the abnormal report, fire monitoring sensors deployed in the forest are used for monitoring the fire precursor in real time, and when a fire early warning signal is detected, an alarm is immediately sent to related departments and fire extinguishing actions are guided to generate a fire response report;
s5: sending the forest image data, the abnormal report, the forest state report and the fire hazard response report to a cloud computing platform for data cleaning, integration and analysis, and enabling the cloud computing platform to execute automatic processing, provide a patrol report and decision support and output a cloud processing report;
s6: based on cloud processing report, processing and analyzing by utilizing a big data analysis technology, so as to optimize a forest resource management strategy and ecological protection measures and generate an ecological management strategy report;
S7: and combining on-site sound monitoring equipment and sound identification technology, and carrying out directional wild animal monitoring according to the ecological management strategy report to obtain a wild animal monitoring report.
Through utilizing unmanned aerial vehicle or static surveillance camera to carry out the omnidirectional and shoot, acquire a large amount of forest image data to utilize degree of depth study and image recognition algorithm to handle, can detect out abnormal phenomena such as fire hazard, insect disease fast, provide timely early warning information. Meanwhile, satellite remote sensing images and geographic information data are analyzed by combining an artificial intelligent algorithm, the forest coverage rate and the damage degree are monitored in real time, and the method is beneficial to making and adjusting a proper protection plan and protecting a forest ecosystem.
The fire monitoring sensor and the sound monitoring equipment are deployed, and the fire disaster response report and the ecological management strategy report are combined, so that the fire disaster precursor and the activity condition of wild animals can be timely detected, and the fire disaster prevention and the ecological management can be carried out in a targeted manner. The data is sent to the cloud computing platform for data cleaning, integration and analysis, so that comprehensive patrol reports and decision support can be generated, forest resource management strategies are optimized, and beneficial ecological protection measures are provided.
In combination, the integration effect of the forest patrol method comprises efficient data collection and analysis, accurate anomaly detection and early warning, real-time forest state monitoring and fire hazard response, optimized resource management strategy and targeted wild animal monitoring. Through the comprehensive effect of the effects, the forest protection level can be improved, the risk of natural disasters is reduced, and beneficial support is provided for scientific forest resource management and ecological protection.
Referring to fig. 2, using an unmanned aerial vehicle or a static monitoring camera, according to a set path and a set time sequence, the steps of performing omnidirectional shooting on a forest to obtain forest image data are specifically as follows:
s101: optimizing parameter settings of the unmanned aerial vehicle or the monitoring camera by adopting a genetic algorithm, including ISO, exposure and white balance, and generating a device configuration file;
S102: applying an A-algorithm or a Dijkstra algorithm, planning an optimal shooting path of the unmanned aerial vehicle or the monitoring camera according to the topography and the ecological barrier of the forest, and generating a path scheme;
s103: and the stability of the unmanned aerial vehicle or the camera is ensured based on the PID controller, and equipment is automatically started at a designated time according to the equipment configuration file and the path scheme to carry out omnibearing shooting so as to generate forest image data.
Firstly, the optimal camera parameter setting can be obtained by optimizing equipment configuration through a genetic algorithm, and high-quality forest image data acquisition is ensured. This helps to improve the sharpness, contrast and color rendition of the image, providing a good basis for subsequent image processing and analysis. Secondly, an optimal path is planned by applying an A-scale algorithm or a Dijkstra algorithm, so that omnibearing and efficient forest coverage can be ensured. The planned path considers the topography and ecological barriers of the forest, so that the key area is covered to the greatest extent, repeated shooting is reduced, and the patrol efficiency is improved. In addition, the stability of unmanned aerial vehicle or camera has been guaranteed in the application of PID controller, ensures to start equipment automatically and shoot in appointed time. The stable shooting capability eliminates the problems of image blurring and shaking, and ensures the usability and quality of data.
Referring to fig. 3, the steps of processing forest image data by using a deep learning and image recognition algorithm, detecting whether abnormal phenomena including fire danger and insect diseases exist, and reporting the abnormal phenomena include:
s201: performing quality enhancement and cleaning on forest image data by using a self-adaptive histogram equalization and noise filtering technology to obtain a preprocessed image data set;
s202: performing deep learning model training based on forest images on the preprocessed image dataset by adopting a transfer learning technology on the basis of a pre-trained ResNet or VGG network to obtain a training model;
S203: and (3) using a YOLO or SSD real-time target detection frame to identify abnormal phenomena of the preprocessed image dataset, and compiling an abnormal report according to the result.
Firstly, the quality and the definition of forest image data are improved through preprocessing of the self-adaptive histogram equalization and noise filtering technology, and a more accurate basis is provided for subsequent analysis and identification. Secondly, training a deep learning model by adopting a transfer learning technology, fully utilizing the characteristics and the weights of the pre-training model, accelerating the model training process and improving the accuracy. The training can learn the characteristics of abnormal phenomena such as fire risks, diseases and insect pests in forest images, and provides reliable recognition capability for target detection. Finally, by using a real-time object detection framework, such as YOLO or SSD, abnormal objects in forest images can be rapidly and accurately identified, detailed abnormal reports can be compiled, and timely early warning information can be provided.
Referring to fig. 4, in combination with an artificial intelligence algorithm, satellite remote sensing images and geographic information data are analyzed, forest coverage and damage degree are monitored in real time, and a forest status report is output specifically as follows:
S301: capturing the latest remote sensing image of a forest by utilizing Landsat satellite resources, and simultaneously carrying out data annotation to construct a remote sensing image database;
s302: analyzing the characteristics of the remote sensing image by adopting a U-Net structure, and carrying out learning and feature extraction on a remote sensing image database to form a remote sensing image analysis model;
S303: and (3) inputting a real-time satellite remote sensing image into a remote sensing image analysis model by combining an edge calculation and a quick response technology, and obtaining the forest coverage rate and the damage degree in real time and generating a forest state report according to the forest coverage rate and the damage degree.
Firstly, by utilizing the latest satellite remote sensing image data, the real-time state of the forest can be comprehensively captured, timely and accurate information is provided, and the possible forest damage and change trend can be quickly identified. And secondly, by means of data labeling and construction of a remote sensing image database, the system can learn and extract the characteristics of the remote sensing image, so that a model with high analysis capability is established, different ground features and coverage types can be automatically identified, and forest coverage rate can be accurately estimated. Third, combining with the edge calculation and the quick response technology, the capability of the system for processing the real-time satellite remote sensing image is enhanced, the damage degree of the forest can be timely observed, and quick and accurate data support is provided for related departments and decision makers. Finally, the generated forest status report provides comprehensive forest information, including assessment of coverage rate and damage degree, provides decision support for forest resource management and ecological protection, and is beneficial to making effective protection strategies and taking timely measures.
Referring to fig. 5, according to a forest status report and an anomaly report, a fire monitoring sensor deployed in a forest is used to monitor a fire precursor in real time, and when a fire early warning signal is detected, an alarm is immediately sent to a relevant department to instruct fire extinguishing actions, and the steps of generating a fire response report are specifically as follows:
S401: determining key and inflammable areas of a forest by adopting a k-means algorithm, and disposing temperature and smoke fire monitoring sensors in a targeted manner to form a sensor network;
S402: applying a Kalman filtering algorithm to fuse and optimize real-time signals from a sensor network to generate a sensor data stream;
S403: and analyzing the sensor data flow according to the precursors including temperature jump and smoke concentration increase based on the classifier of the support vector machine, immediately sending an alarm to related departments once the fire hazard early warning is detected, and forming a fire hazard response report.
Firstly, a k-means algorithm is utilized to determine key and flammable areas of a forest, and fire monitoring sensors such as air temperature, smoke and the like are deployed in a targeted manner to form a distributed sensor network. Secondly, a Kalman filtering algorithm is applied to fuse and optimize real-time signals collected in the sensor network, so that a sensor data stream is generated, and the accuracy and stability of data are improved. And then, analyzing the sensor data stream by a classifier based on a support vector machine, and judging according to fire precursors including temperature jump, smoke concentration increase and the like. Upon detection of the fire alarm signal, the system immediately sends an alarm to the relevant department while providing advice to direct the action of extinguishing the fire. Finally, the system generates a fire hazard response report which comprises detailed information of fire hazard early warning, corresponding measures taken, effect evaluation and the like. By comprehensively applying the technologies and steps, the effects of fire monitoring and prevention and control can be greatly improved, the loss and harm caused by fire can be effectively reduced, and the safety of forest resources and ecological environment can be protected.
Referring to fig. 6, based on forest image data, an anomaly report, a forest status report, and a fire hazard response report, the forest status report and the fire hazard response report are sent to a cloud computing platform for data cleaning, integration and analysis, the cloud computing platform performs automatic processing and provides a patrol report and decision support, and the step of outputting a cloud processing report specifically includes:
S501: uploading forest image data, an abnormal report, a forest state report and a fire hazard response report to a cloud computing platform by using a distributed uploading technology, and establishing an original data set;
s502: performing preliminary screening on error, missing and redundant data in the original data set by using a data quality detection algorithm, and processing the missing data by using an interpolation algorithm to obtain a cleaned data set;
s503: classifying and marking the data by using a data conversion and mapping technology, integrating the data in the cleaned data set according to time, place and type, and generating an integrated data set;
s504: deep analysis is carried out on the integrated data set by utilizing a machine learning and statistics method, the data characteristics are analyzed and classified, potential problems and risks of forests are detected, and a preliminary analysis report is output;
S505: according to the preliminary analysis report, the cloud computing platform generates an optimal proposal and strategy for a decision maker by utilizing a decision tree or a neural network model, and outputs a patrol proposal report;
S506: and integrating the integrated data set and the patrol proposal report by using a report generating tool to be used as a cloud processing report.
Firstly, through data cleaning and integration, the accuracy and the integrity of the data can be improved, and the influences of errors, missing and redundant data are eliminated, so that the analysis result is more reliable and accurate. And secondly, deep analysis is performed by using a machine learning and statistical method, so that potential problems and risks in the forest can be found, and key decision support is provided. Third, the rapid computing and storage capacity of the cloud computing platform enables the processes of data processing and decision support to be more efficient and real-time, and can timely respond to emergency situations such as forest state changes and fire early warning. In addition, by generating the patrol report and the decision support, comprehensive forest information and flexible measure suggestions can be provided for related departments and decision makers, and effective forest resource management and ecological protection are promoted.
Referring to fig. 7, based on the cloud processing report, the big data analysis technology is used for processing and analyzing, so as to optimize the forest resource management policy and the ecological protection measures, and the steps for generating the ecological management policy report are specifically as follows:
S601: analyzing and extracting key data and suggestions in the cloud processing report by using a natural language processing technology to generate key information data;
S602: based on the key information data, applying association rule learning and clustering algorithm to carry out depth data mining, and outputting mining results;
s603: based on the mining result, optimizing a forest resource management strategy and ecological protection measures by utilizing a multi-objective optimization algorithm and combining current resources and constraints to generate an optimized strategy;
S604: and converting the optimized strategy into a specific operation guideline and suggestion by using a strategy generation tool, and generating an ecological management strategy report.
Firstly, by analyzing and extracting key information data, important information in the cloud processing report can be rapidly acquired, and time and energy are saved. And secondly, carrying out depth data mining by utilizing association rule learning and clustering algorithm, revealing rules and trends hidden in the data, and providing accurate insight and analysis results for decision makers. Thirdly, based on a multi-objective optimization algorithm, comprehensively considering resources and constraint conditions, realizing optimization of a forest resource management strategy and ecological protection measures, and promoting sustainable development and ecological balance. In addition, the optimized strategy is converted into specific operation guidelines and suggestions, and implementation guidance and decision support can be provided for a decision maker. The cloud computing and big data analysis technology is comprehensively utilized, and the ecological management strategy report is integrally generated, so that the scientificity and the accuracy of decision making are improved, and sustainable forest management and ecological protection are promoted.
Referring to fig. 8, in combination with the on-site sound monitoring device and the sound recognition technology, the method for monitoring the wild animals in a direction according to the report of the ecological management policy specifically includes the following steps:
s701: according to suggestions in the ecological management policy report, deploying sound equipment, periodically collecting sound data in a forest from the sound equipment by using a wireless transmission technology, and generating an original sound data set;
S702: using a fast Fourier transform and filtering technology to perform noise reduction and format conversion on the original sound data set to obtain a cleaned sound data set;
S703: analyzing the cleaned sound data set by utilizing a sound recognition technology, extracting and classifying sound characteristics by using a convolutional neural network in deep learning, and outputting an animal sound recognition result;
S704: based on animal voice recognition results, carrying out statistics and trend prediction on animal activities by using a time sequence analysis and pattern matching technology, and generating an animal activity analysis report;
s705: and generating a wild animal monitoring report by using a report generating tool and combining the animal voice recognition result and the animal activity analysis report.
Firstly, through deploying sound equipment and collecting sound data, the sound information of the wild animals can be obtained in real time, and the accuracy and timeliness of monitoring are greatly improved. And secondly, the voice data is analyzed and classified by utilizing the voice recognition technology, so that different kinds of wild animal voices can be accurately identified, and a basis is provided for species recognition and quantity statistics. Thirdly, through time sequence analysis and trend prediction technology, the activity rule and migration trend of the wild animals can be known, and scientific reference is provided for protection and management. In addition, the generated wild animal monitoring report contains voice recognition results, animal activity analysis and other important findings, and provides detailed information and decision support for decision makers and related departments.
Referring to fig. 9, a forest patrol system is composed of an image acquisition module, an image recognition module, a remote sensing and GIS analysis module, a fire hazard early warning module, a cloud data processing module and an ecological management strategy module;
The image acquisition module shoots a forest according to a preset path and a preset time sequence by using the unmanned aerial vehicle and the static monitoring camera so as to acquire omnibearing forest image data;
An image recognition module: deep learning and image recognition processing are carried out on the omnibearing forest image data, abnormal phenomena are detected in real time, and an abnormal report is generated;
The remote sensing and GIS analysis module is combined with an artificial intelligent algorithm to analyze satellite remote sensing images and geographic information data, detect forest coverage rate and damage degree and output a forest status report;
the fire early warning module deploys fire monitoring sensors in the key areas according to the forest state report and the abnormal report, and when a fire early warning signal is detected, the fire early warning module gives an alarm to related departments and generates a fire response report;
The cloud data processing module uploads the omnibearing forest image data, the abnormal report, the forest status report and the fire hazard response report to the cloud computing platform for data integration, cleaning and analysis to generate a comprehensive patrol report and decision support information, and outputs a cloud processing report;
The ecological management strategy module is combined with cloud processing reporting, optimizes forest resource management strategies and ecological protection measures by using a big data analysis technology, and simultaneously monitors wild animals by combining with a sound monitoring technology to generate an ecological management strategy report and a wild animal monitoring report.
Firstly, the image acquisition module is used for acquiring the omnibearing forest image data, so that a comprehensive and accurate visual angle can be provided, forest abnormal conditions such as fire, plant diseases and insect pests and the like can be found timely, and measures can be taken rapidly for processing. And secondly, the image recognition module processes the image data by utilizing deep learning and image recognition technology, detects abnormal phenomena in real time and generates an abnormal report, and provides instant abnormal condition recognition and report generation capability.
Further, the remote sensing and GIS analysis module is combined with an artificial intelligence algorithm to analyze satellite remote sensing images and geographic information data, can detect forest coverage rate and damage degree, and generate a forest status report, so that accurate forest health information and scientific management strategy establishment basis are provided for a manager. Meanwhile, the fire hazard early warning module deploys fire monitoring sensors in the key areas according to the forest state report and the abnormal report, timely sends out fire hazard alarms and generates fire hazard response reports, and fire hazard early warning and emergency response capabilities are improved.
And the cloud data processing module uploads the data to the cloud computing platform for integration, cleaning and analysis, and generates a comprehensive patrol report and decision support information. Through cloud processing, the data are comprehensively analyzed, scientific patrol report and management strategy support are provided for a decision maker, and sustainable management and protection of forest resources are promoted.
And finally, the ecological management strategy module combines cloud processing reporting, and optimizes forest resource management strategies and ecological protection measures by utilizing a big data analysis technology. And (3) combining with a sound monitoring technology, performing wild animal monitoring, and generating an ecological management strategy report and a wild animal monitoring report. This helps to provide scientific basis and decision support, pushing the integrity and sustainable development of the forest ecosystem.
Referring to fig. 10, the image acquisition module includes a unmanned aerial vehicle sub-module, a still camera sub-module, and a time sequence shooting control sub-module;
the image recognition module comprises a deep learning sub-module, a fire risk recognition sub-module and a plant disease and insect pest monitoring sub-module;
The remote sensing and GIS analysis module comprises a satellite remote sensing sub-module, a geographic information processing sub-module and a forest coverage analysis sub-module;
The fire hazard early warning module comprises a sensor deployment sub-module, a real-time fire hazard monitoring sub-module and an alarm sending sub-module;
the cloud data processing module comprises a data uploading sub-module, a data cleaning sub-module and a decision support generating sub-module;
the ecological management strategy module comprises a big data strategy optimization sub-module, a sound monitoring sub-module and a wild animal identification sub-module.
Firstly, the image acquisition module utilizes the unmanned aerial vehicle sub-module, the static camera sub-module and the time sequence shooting control sub-module to realize the comprehensive and high-resolution forest image acquisition, provides a comprehensive and accurate visual angle, and helps to discover forest abnormal conditions in time. And secondly, the image recognition module monitors abnormal phenomena in image data, such as fire, plant diseases and insect pests and the like in real time through the deep learning sub-module, the fire risk recognition sub-module and the plant diseases and insect pests monitoring sub-module, generates an abnormal report, and provides a basis for quick response and processing for patrolling personnel. In addition, the remote sensing and GIS analysis module is combined with the satellite remote sensing sub-module, the geographic information processing sub-module and the forest coverage analysis sub-module, so that the coverage rate and the damage degree of the forest can be accurately analyzed, a forest state report is output, and scientific data support and decision basis are provided for a manager. The fire hazard early warning module monitors and early warns fire hazard risks in advance through the sensor deployment sub-module, the real-time fire hazard monitoring sub-module and the alarm sending sub-module, timely sends out an alarm and generates a fire hazard response report, and occurrence and spread of the fire hazard are greatly reduced. On the other hand, the cloud data processing module utilizes a data uploading sub-module, a data cleaning sub-module and a decision support generating sub-module to realize data integration, cleaning and analysis in all aspects. The cloud processing can improve the efficiency and accuracy of data processing, generate comprehensive patrol reports and decision support information, and provide scientific management strategies and decision support for decision makers. The ecological management strategy module is combined with the big data strategy optimization sub-module, the sound monitoring sub-module and the wild animal identification sub-module, optimizes the forest resource management strategy through the big data analysis technology, and combines with the sound monitoring and the wild animal identification technology to carry out ecological monitoring and protection, thereby realizing comprehensive ecological management and protection measures.
Working principle: the forest patrol system uses an unmanned aerial vehicle or a static monitoring camera to carry out omnibearing shooting on a forest on a preset path so as to acquire image data. Deep learning and image recognition algorithms are then applied to process these image data, detect anomalies such as fire hazards and insect pests, and generate corresponding anomaly reports. Meanwhile, the remote sensing and GIS analysis module analyzes satellite remote sensing images and geographic information data by using an artificial intelligent algorithm, monitors the coverage rate and the damage degree of the forest in real time, and generates a forest state report. According to the reports, the fire hazard early warning module arranges fire hazard monitoring sensors in the key areas, and immediately sends an alarm to related departments once the sensors detect fire hazard, and generates a fire hazard response report to guide fire extinguishing actions. All data, including image data, anomaly reports, forest status reports and fire response reports, are uploaded to the cloud data processing module for data cleaning, integration and analysis. The cloud computing platform can execute automatic processing, provide patrol reports and decision support and output related processing reports. In addition, the ecological management strategy module is also used for further processing and analyzing the data by utilizing a big data analysis technology in combination with cloud processing results, so that the forest resource management strategy and ecological protection measures are optimized. There are also voice monitoring devices and voice recognition techniques for wildlife monitoring and generating ecological management policy reports and wildlife monitoring reports.
The present invention is not limited to the above embodiments, and any equivalent embodiments which can be changed or modified by the technical disclosure described above can be applied to other fields, but any simple modification, equivalent changes and modification made to the above embodiments according to the technical matter of the present invention will still fall within the scope of the technical disclosure.
Claims (3)
1. The forest patrol method is characterized by comprising the following steps of:
Carrying out omnibearing shooting on a forest by using an unmanned aerial vehicle or a static monitoring camera according to a set path and a set time sequence to acquire forest image data;
processing the forest image data by utilizing a deep learning and image recognition algorithm, detecting whether abnormal phenomena including fire risks and diseases and insect pests exist or not, and giving an abnormal report;
Analyzing satellite remote sensing images and geographic information data by combining an artificial intelligent algorithm, monitoring forest coverage rate and damage degree in real time, and outputting a forest state report;
According to the forest status report and the abnormality report, fire risk precursors are monitored in real time through fire monitoring sensors deployed in the forest, and when a fire risk early warning signal is detected, an alarm is immediately sent to a related department and fire extinguishing actions are guided to generate a fire risk response report;
Sending the forest image data, the abnormal report, the forest state report and the fire hazard response report to a cloud computing platform for data cleaning, integration and analysis, and enabling the cloud computing platform to execute automatic processing, provide a patrol report and decision support and output a cloud processing report;
Based on the cloud processing report, processing and analyzing by utilizing a big data analysis technology, so as to optimize a forest resource management strategy and ecological protection measures and generate an ecological management strategy report;
Combining on-site sound monitoring equipment and sound identification technology, and carrying out directional wild animal monitoring according to the ecological management strategy report to obtain a wild animal monitoring report;
the method comprises the following steps of carrying out omnibearing shooting on a forest by using an unmanned aerial vehicle or a static monitoring camera according to a set path and a set time sequence, and obtaining forest image data:
Optimizing parameter settings of the unmanned aerial vehicle or the monitoring camera by adopting a genetic algorithm, including ISO, exposure and white balance, and generating a device configuration file;
application A An algorithm or Dijkstra algorithm plans an optimal shooting path of the unmanned aerial vehicle or the monitoring camera according to the topography and ecological barriers of the forest, and generates a path scheme;
The stability of the unmanned aerial vehicle or the camera is ensured based on the PID controller, equipment is automatically started at a designated time according to the equipment configuration file and the path scheme, and the omnidirectional shooting is carried out to generate forest image data;
Processing the forest image data by utilizing a deep learning and image recognition algorithm, and detecting whether abnormal phenomena including fire risks and diseases and insect pests exist or not, wherein the step of giving an abnormal report specifically comprises the following steps:
Performing quality enhancement and cleaning on the forest image data by using a self-adaptive histogram equalization and noise filtering technology to obtain a preprocessed image data set;
Performing deep learning model training based on forest images on the preprocessed image dataset by adopting a transfer learning technology on the basis of a pre-trained ResNet or VGG network to obtain a training model;
Using a YOLO or SSD real-time target detection frame to identify abnormal phenomena of the preprocessed image dataset, and compiling an abnormal report according to the result;
the satellite remote sensing image and the geographic information data are analyzed by combining an artificial intelligent algorithm, the forest coverage rate and the damage degree are monitored in real time, and the step of outputting a forest state report is specifically as follows:
Capturing the latest remote sensing image of a forest by utilizing Landsat satellite resources, and simultaneously carrying out data annotation to construct a remote sensing image database;
Analyzing the characteristics of a remote sensing image by adopting a U-Net structure, and carrying out learning and feature extraction on the remote sensing image database to form a remote sensing image analysis model;
inputting a real-time satellite remote sensing image into the remote sensing image analysis model by combining an edge calculation and a quick response technology, and obtaining forest coverage rate and damage degree in real time and generating a forest status report according to the forest coverage rate and damage degree;
According to the forest status report and the abnormality report, the fire risk precursor is monitored in real time by a fire monitoring sensor deployed in the forest, and when a fire risk early warning signal is detected, an alarm is immediately sent to a relevant department and the fire extinguishing action is guided, and the step of generating a fire risk response report specifically comprises the following steps:
Determining key and inflammable areas of a forest by adopting a k-means algorithm, and disposing temperature and smoke fire monitoring sensors in a targeted manner to form a sensor network;
Applying a Kalman filtering algorithm to fuse and optimize real-time signals from a sensor network to generate a sensor data stream;
The classifier based on the support vector machine analyzes the sensor data stream according to the precursors including temperature jump and smoke concentration increase, and immediately gives an alarm to the relevant departments once the fire hazard early warning is detected, and forms a fire hazard response report;
Sending the forest image data, the abnormal report, the forest state report and the fire hazard response report to a cloud computing platform for data cleaning, integration and analysis, wherein the cloud computing platform executes automatic processing, provides a patrol report and decision support, and outputs cloud processing report:
uploading the forest image data, the abnormal report, the forest state report and the fire hazard response report to a cloud computing platform by using a distributed uploading technology, and establishing an original data set;
performing preliminary screening on error, missing and redundant data in the original data set by using a data quality detection algorithm, and processing the missing data by using an interpolation algorithm to obtain a cleaned data set;
Classifying and marking the data by using a data conversion and mapping technology, integrating the data in the cleaned data set according to time, place and type to generate an integrated data set;
Deep analysis is carried out on the integrated data set by utilizing a machine learning and statistics method, the data characteristics are analyzed and classified, potential problems and risks of forests are detected, and a preliminary analysis report is output;
according to the preliminary analysis report, the cloud computing platform generates an optimal proposal and strategy for a decision maker by utilizing a decision tree or a neural network model, and outputs a patrol proposal report;
integrating the integrated data set and the patrol proposal report by using a report generating tool to be used as a cloud processing report;
Based on the cloud processing report, the big data analysis technology is utilized for processing and analyzing, so that a forest resource management strategy and ecological protection measures are optimized, and the ecological management strategy report is generated specifically by the following steps:
analyzing and extracting key data and suggestions in the cloud processing report by using a natural language processing technology to generate key information data;
Based on the key information data, applying an association rule learning and clustering algorithm to carry out depth data mining, and outputting mining results;
Based on the mining result, optimizing a forest resource management strategy and ecological protection measures by utilizing a multi-objective optimization algorithm and combining current resources and constraints to generate an optimized strategy;
converting the optimized strategy into a specific operation guide and suggestion by using a strategy generation tool, and generating an ecological management strategy report;
And combining on-site sound monitoring equipment and sound identification technology, and carrying out directional wild animal monitoring according to the ecological management strategy report, wherein the step of giving a wild animal monitoring report comprises the following steps:
According to the suggestions in the ecological management policy report, deploying sound equipment, periodically collecting sound data in a forest from the sound equipment by using a wireless transmission technology, and generating an original sound data set;
using a fast Fourier transform and filtering technology to perform noise reduction and format conversion on the original sound data set to obtain a cleaned sound data set;
Analyzing the cleaned sound data set by utilizing a sound recognition technology, extracting and classifying sound characteristics by using a convolutional neural network in deep learning, and outputting an animal sound recognition result;
Based on the animal voice recognition result, carrying out statistics and trend prediction on animal activities by utilizing a time sequence analysis and pattern matching technology, and generating an animal activity analysis report;
And generating a wild animal monitoring report by using a report generating tool and combining the animal voice recognition result and the animal activity analysis report.
2. A forest patrol system for realizing the forest patrol method according to claim 1, which is characterized by comprising an image acquisition module, an image recognition module, a remote sensing and GIS analysis module, a fire hazard early warning module, a cloud data processing module and an ecological management strategy module;
the image acquisition module shoots a forest according to a preset path and a preset time sequence by using the unmanned aerial vehicle and the static monitoring camera so as to acquire omnibearing forest image data;
the image recognition module: deep learning and image recognition processing are carried out on the omnibearing forest image data, abnormal phenomena are detected in real time, and an abnormal report is generated;
The remote sensing and GIS analysis module is combined with an artificial intelligence algorithm to analyze satellite remote sensing images and geographic information data, detect forest coverage rate and damage degree and output a forest status report;
The fire early warning module deploys fire monitoring sensors in the key areas according to the forest state report and the abnormal report, and when a fire early warning signal is detected, the fire early warning module gives an alarm to related departments and generates a fire response report;
The cloud data processing module uploads the omnibearing forest image data, the abnormal report, the forest status report and the fire hazard response report to the cloud computing platform for data integration, cleaning and analysis to generate a comprehensive patrol report and decision support information, and outputs a cloud processing report;
The ecological management strategy module is combined with cloud processing reporting, optimizes forest resource management strategies and ecological protection measures by using a big data analysis technology, and simultaneously monitors wild animals by combining with a sound monitoring technology to generate an ecological management strategy report and a wild animal monitoring report.
3. The forest patrol system according to claim 2, wherein the image acquisition module comprises an unmanned aerial vehicle sub-module, a still camera sub-module, a time sequence shooting control sub-module;
The image recognition module comprises a deep learning sub-module, a fire risk recognition sub-module and a plant disease and insect pest monitoring sub-module;
the remote sensing and GIS analysis module comprises a satellite remote sensing sub-module, a geographic information processing sub-module and a forest coverage analysis sub-module;
the fire hazard early warning module comprises a sensor deployment sub-module, a real-time fire hazard monitoring sub-module and an alarm sending sub-module;
the cloud data processing module comprises a data uploading sub-module, a data cleaning sub-module and a decision support generating sub-module;
The ecological management strategy module comprises a big data strategy optimization sub-module, a sound monitoring sub-module and a wild animal identification sub-module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311653781.XA CN117636169B (en) | 2023-12-05 | 2023-12-05 | Forest patrol method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311653781.XA CN117636169B (en) | 2023-12-05 | 2023-12-05 | Forest patrol method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117636169A CN117636169A (en) | 2024-03-01 |
| CN117636169B true CN117636169B (en) | 2024-08-20 |
Family
ID=90016055
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311653781.XA Active CN117636169B (en) | 2023-12-05 | 2023-12-05 | Forest patrol method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117636169B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117911882B (en) * | 2024-03-20 | 2024-06-04 | 北京新兴科遥信息技术有限公司 | Monitoring method and system for forestry protection planning dynamic data |
| CN118229081A (en) * | 2024-04-08 | 2024-06-21 | 广东经纬天地科技有限公司 | Forest safety monitoring method, system, equipment and medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112435207A (en) * | 2020-12-07 | 2021-03-02 | 深圳航天智慧城市系统技术研究院有限公司 | Forest fire monitoring and early warning method based on sky-ground integration |
| CN113204736A (en) * | 2021-05-06 | 2021-08-03 | 中国科学技术大学 | Method, system, medium and equipment for real-time monitoring and prediction of forest fire danger |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102280005B (en) * | 2011-06-09 | 2014-10-29 | 广州飒特红外股份有限公司 | Early warning system for fire prevention of forest based on infrared thermal imaging technology and method |
| WO2022104736A1 (en) * | 2020-11-20 | 2022-05-27 | 周连惠 | Method and system for managing forest resources |
| CN115348247A (en) * | 2022-08-09 | 2022-11-15 | 四川科瑞源信息科技有限公司 | Forest fire detection early warning and decision-making system based on sky-ground integration technology |
-
2023
- 2023-12-05 CN CN202311653781.XA patent/CN117636169B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112435207A (en) * | 2020-12-07 | 2021-03-02 | 深圳航天智慧城市系统技术研究院有限公司 | Forest fire monitoring and early warning method based on sky-ground integration |
| CN113204736A (en) * | 2021-05-06 | 2021-08-03 | 中国科学技术大学 | Method, system, medium and equipment for real-time monitoring and prediction of forest fire danger |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117636169A (en) | 2024-03-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117636169B (en) | Forest patrol method and system | |
| KR101984983B1 (en) | System and method for monitoring wild animals | |
| Stephens et al. | Management by proxy? The use of indices in applied ecology | |
| KR101598898B1 (en) | The system for precaution for paralichthys olivaceus disease using and an analysis of image and sound | |
| CN107888887A (en) | A kind of video monitoring method for early warning and system for monitoring gas pipeline damage from third-party | |
| CN111639825B (en) | Forest fire indication escape path method and system based on A-Star algorithm | |
| CN111686392A (en) | Artificial intelligence fire extinguishing system is surveyed to full scene of vision condition | |
| CN114665608B (en) | Intelligent sensing inspection system and method for transformer substation | |
| CN111652128B (en) | High-altitude power operation safety monitoring method, system and storage device | |
| Kinaneva et al. | Application of artificial intelligence in UAV platforms for early forest fire detection | |
| CN115631421A (en) | Intelligent protection method and system for cultivated land | |
| Wildemeersch et al. | Modelling the multi-scaled nature of pest outbreaks | |
| US20240193946A1 (en) | Bird detection and species determination | |
| CN116052082A (en) | Power distribution station room anomaly detection method and device based on deep learning algorithm | |
| CN116543347A (en) | Intelligent insect condition on-line monitoring system, method, device and medium | |
| CN117094177B (en) | Track generation method and system based on solenopsis invicta prevention and control | |
| Saravanan et al. | IoT alert reflexion of forbidden deforestation regions with drone observation | |
| CN116863410B (en) | Data acquisition and processing method and system based on pest control | |
| Mekala et al. | Forest Fire Probability Prediction based on Humidity and Temperature | |
| Krishnaveni et al. | IoT and AI based Forest Fire Prediction and Animal Monitoring System | |
| Srivastava et al. | Iot Alert Observation of Prohibited Deforestation Regions with Drone Surveillance | |
| CN116899158A (en) | Linkage method and system for forming multistage fire extinguishing force based on data acquisition | |
| CN118071126B (en) | Intelligent cooperation method and system based on digital agricultural comprehensive law enforcement platform technology | |
| KR102581941B1 (en) | A monitering system for wearing muzzles of dog using deep learning and monitering method | |
| Bazzi et al. | Fruit fly electronic monitoring system. |
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 | ||
| GR01 | Patent grant |