It is precisely because of the advantages of UAV technology, which is very consistent with some work in forestry work, that its application in forestry work gradually goes deep, greatly improves the efficiency and accuracy of forestry work, and reduces the work intensity of traditional manual operation.
1、 Application of UAV technology in forestry
1.1 application in forestry pest monitoring and control
Forest pest monitoring and control is one of the important contents in forestry work. With the continuous advancement of afforestation in China, the continuous increase of greening area and the influence of various factors, forestry diseases and pests show a trend of multiple and widespread occurrence. However, the traditional monitoring and control of forest diseases and pests mainly rely on manual operation, which is difficult to realize the timeliness and effective coverage of monitoring and control. In this context, the application of UAV technology has effectively improved the level of pest monitoring and control.
First, most forestry diseases and pests occur in areas with inconvenient transportation and complex terrain. Manual operation requires long journey, long time and high intensity, which is difficult to ensure the timeliness of monitoring and early warning. UAV remote sensing technology can carry out timely monitoring in a large range, comprehensively and timely reflect the occurrence dynamics of diseases and pests, with strong real-time and high security, effectively reduce the use of manpower, and timely feed back information, so that the prevention and control personnel of diseases and pests have enough time for emergency disposal. This greatly improves the timeliness of forestry pest monitoring and control, and is conducive to promoting the healthy development of forestry economy.
Second, the application of UAV technology for monitoring has improved the coverage of forestry pest monitoring. When monitoring forestry resources in high mountains and steep slopes that are difficult to reach by manpower, UAV low altitude remote sensing technology and aerial mapping technology have obvious advantages, which can quickly determine the occurrence and development dynamics of diseases and pests and obtain remote sensing data in time. At the same time, the efficiency of UAV is higher. According to practice, it can complete the accurate monitoring of forest diseases and pests of thousands of hectares in one day, which is unmatched by traditional manual operation.
Third, although UAV monitoring is greatly affected by meteorological conditions, its low altitude aerial photography technology greatly avoids the disadvantages of high-altitude monitoring means such as satellites, has higher accuracy, clearer image information, is conducive to marking the longitude and latitude of trees, reduces the location of manually positioning pests and even specific damaged trees, and effectively improves the timeliness and effectiveness of prevention and control work.
Fourth, in the link of pest control, UAV technology can be used as an operating machine to carry spraying equipment for agricultural spraying control under the guidance of reasonable control scheme. The first mock exam is effective in avoiding the direct contact between forestry workers and pesticides and effectively reducing labor costs. It is one of the effective measures to improve the safety of forestry production.
1.2 application in afforestation
Before afforestation, it is necessary to accurately obtain the relevant information of the afforestation area, so as to formulate a scientific and reasonable afforestation plan. In this link, the application of UAV technology can not only accurately obtain image information, and then analyze the influence of soil, terrain, meteorology, rivers and other factors, but also improve the quality of afforestation. In the afforestation stage, UAV technology can realize remote ultra-low fertilizer spraying and improve the afforestation efficiency.
1.3 application in forest fire prevention
Forest fire early warning and fighting is the top priority of forestry work. Because the occurrence of forest fire is not only difficult to put out, but also causes huge ecological losses, economic losses and casualties. The traditional forest fire prevention work mainly relies on staff to patrol on the spot, while the application of UAV technology has the advantages of low cost, strong timeliness, timely detection, timely early warning and safe fire fighting. At the same time, UAV has a variety of operation modes, which can effectively adapt to the sudden fire.
During the monitoring flight of the preset route, the UAV can transmit the monitoring data in real time through photographing images and thermal imaging technology, which greatly increases the possibility that the fire will be found in the initial stage, and determines the fire point and fire range, which is conducive to the staff to organize rescue in time. In case of forest fire, the fire site can be monitored without personnel arriving at the fire site, so that the staff can judge the fire situation in time and find the orientation of trapped personnel, so as to implement fire fighting and personnel help. In addition, it is more conducive to the scientific prediction of fire spread and timely organize the people to transfer. At the same time, it is quite dangerous to carry out fire fighting and help seeking work at the fire scene, and UAV can timely transport rescue materials and equipment, improve rescue efficiency, and transport equipment and materials for front-line firefighters to ensure the effective progress of fire fighting work.
1.4 application in forest resources investigation
Forest resources investigation is an important task in forestry work. The main methods of forest resources survey include remote sensing technology survey, image survey and sampling survey, which require a large number of staff to carry out field survey and sampling. In the traditional investigation work, it needs a lot of manpower and material resources, and the investigation efficiency is relatively low. The application of UAV technology can quickly obtain the spatial remote sensing information of relevant regions through UAV remote sensing technology, and obtain and analyze the forest resources in the region with high precision, which greatly improves the timeliness and accuracy of the investigation and reduces the cost. For example, in terms of monitoring wildlife resources, UAV technology can not only realize the real-time monitoring of species population distribution and growth, but also facilitate tracking and monitoring, and reduce the disturbance to wildlife.
1.5 application in vegetation cover monitoring
Vegetation coverage is one of the important indicators of forestry ecological environment evaluation. Therefore, accurate monitoring of vegetation coverage is very important. Traditional monitoring methods have large workload and low accuracy, which are difficult to meet the needs of current work. The application of UAV remote sensing technology can greatly reduce the cost and work intensity of manual monitoring, improve the efficiency of digital image acquisition, expand the coverage of monitoring, obtain image data with higher resolution, improve the accuracy of vegetation coverage monitoring, and improve more scientific reference data for forestry development.
2、 Forest information management platform
The platform has an internal development team of 150 people, and the platform technology has been supported by the National 863 technology support and the National Innovation Fund for small and medium-sized enterprises. It has developed a comprehensive solution for forest fire prevention, ecology, resource monitoring and early warning, which has been widely used in forestry departments. The integrated forest information management platform includes three parts: intelligent forest fire prevention early warning system, intelligent forest ecological monitoring system and intelligent forest resource supervision system. Secondly, there are operation and maintenance management, data management, video management, fire dispatching, big data analysis, app and other subsystem functions.
Forest fire prevention grid monitoring and early warning cloud platform is an intelligent video monitoring system that is unattended in the front section, real-time monitoring all day, intelligent alarm and really suitable for forestry characteristics. The system integrates heat source detection, data collection, video linkage, early warning and prediction, emergency command and other functions. Based on narrowband Internet of things technology, the system is based on grid heat source monitoring, supplemented by video monitoring, fire alarm as the core function, and expands additional functions such as geographic information system and emergency command and decision-making. The system can fully realize the functions of monitoring, fire risk early warning and information release, and realize the scientization, informatization, standardization and efficiency of forest fire prevention, so as to find and solve the disaster early. The platform introduces dozens of UAVs and multiple power delta wing equipment, which are equipped with orthophoto and tilt photography dual sensor systems. Medium and low altitude remote sensing service centers are established in Northeast China, Guangxi, Tibet and Xinjiang respectively. UAVs have 24-hour arrival service capability in most parts of the country, and can provide customized, multi-type and diversified aviation remote sensing data services for users.
2.1 system characteristics
(1) Automatic scanning: the platform provides zigzag scanning operation scheme, supports panoramic puzzles, and can set automatic cruise scripts in time periods to achieve real-time monitoring within the early warning range.
(2) Accurate recognition: using thermal imaging pyrotechnic recognition technology, core intelligent recognition algorithm, spectral analysis, spatial geometric feature analysis and motion trajectory analysis of pyrotechnics, the accuracy of pyrotechnic recognition is more than 95%, and there is no missing report.
(3) Timely alarm: once the system recognizes fireworks, the monitoring room of the command center will sound an alarm. At the same time, the platform and app will notify relevant personnel with alarm prompt icons to alarm in various ways at the first time.
(4) Accurate positioning: after finding a suspected fire, quickly and accurately locate the GPS position of the fire point according to the GPS position of the equipment in the front section, PTZ orientation information, fire source pixel coordinates and 3D GIS elevation data of the map, and mark the fire point position on the electronic map, with a positioning accuracy of up to tens of meters.
Figure fire source identification and location
2.2 system advantages
1. Tailored to the special environment and needs of forestry, the application technology is mature and the operation is stable (1) the protection grade of front-end equipment is IP67, which is suitable for all kinds of bad weather and environment; (2) It can realize uninterrupted transmission, uninterrupted power supply and uninterrupted remote maintenance; (3) The monitoring radius is 3-15km, which can meet the monitoring needs of different forest areas and different environments. 2. It can realize early detection, accurate positioning and accurate alarm of fire. (1) the internationally leading high-precision PTZ and multispectral dynamic fireworks identification algorithm can accurately identify fire; (2) Mature embedded dynamic fireworks identification technology can realize rapid and accurate fire discrimination and early warning; (3) High precision positioning and pointing, linked with 3D GIS information, can quickly, accurately and locate fire points. 3. It is easy to operate, can realize multi business linkage, and has strong compatibility and extensibility. (1) it closely conforms to the forest fire prevention early warning and command business, and the operation is simple and easy to use; (2) It has the functions of forest fire risk prediction, fire intelligent monitoring, fire emergency auxiliary decision-making, fire loss assessment, information release, etc; (3) Integrate a variety of equipment interfaces, which can be seamlessly integrated with the Internet of things monitoring system. 4. Multi level deployment, information interconnection, remote intelligent management and low operation and maintenance cost (1) the system supports the multi-level platform structure deployment of the state, province, city and county (Forestry Bureau), and can realize information interconnection and multi-level collaborative work; (2) It has remote management function to reduce operation and maintenance cost. 5. The comprehensive investment is less, the construction investment is saved, the number of monitoring stations is reduced through the product coverage radius, and the saved capital construction cost is used to improve the safe operation degree of a single station, so as to save the construction cost and maintenance cost on the basis of ensuring the monitoring and early warning effect of the system.
2.3 core technology – intelligent image fireworks recognition algorithm
1. The original data analysis algorithm uses the uncompressed original data for identification, which can quickly distinguish the smoke characteristics in the image background, so that the whole algorithm system can distinguish one more level, improve the accuracy level of the algorithm and eliminate missing reports. 2. The two-level discrimination of dynamic analysis and static filtering adopts the industry’s unique motion recognition and image enhancement technology. The ball table can identify the fire in dynamic cruise, which improves the image processing time, and the coverage area of a single point is much larger than that of traditional video fire prevention equipment. 3. Infrared coordinate mapping and visible light aided analysis infrared algorithm can calculate the position of each pixel on the infrared image corresponding to the visible light image, and then use the visible light algorithm for auxiliary recognition to effectively avoid false positives. 4. The deep learning algorithm based on big data and the artificial intelligence algorithm based on imitating human neural network are adopted in the fireworks recognition system, Learn from the massive fire data, memorize all kinds of fireworks features, improve the algorithm model, strengthen the adaptability of the algorithm, and automatically improve the fire recognition rate.
In addition, the intelligent forestry ecological monitoring system and intelligent forestry resource supervision system can meet the needs of automatic Forest Farm felling, illegal construction inspection and supervision, automatic ecological data collection, data display and analysis, forestry vegetation information management and so on.
(1) Automatic Forest Farm felling, illegal construction inspection and supervision
China’s forest farms generally have a large area and inconvenient transportation and communication, which increases great difficulty for the regulation of deforestation and illegal construction of forest farms. At present, most forest farms adopt the way of patrol team survey to supervise the forest farms, which has low patrol efficiency, long cycle, untimely information feedback and high missed inspection rate.
The intelligent forestry resources supervision system provides spatio-temporal data comparison technology to conduct regular inspection of the forest farm, compare and automatically analyze the remote sensing map data at different times, and automatically analyze the logging and illegal construction of the forest farm through the results of map differentiation.
(2) Automatic collection, display and analysis of forest farm ecological data
The forest farm has rich ecological data, including meteorology, environment, soil quality, water quality, etc. most of the forest farms have backward equipment. The data collection is still carried out by traditional personnel field collection. There is no unified platform for display and management of all kinds of data, let alone data analysis and utilization.
The intelligent forestry ecological monitoring system automatically collects, centrally manages and analyzes various ecological data of the forest farm through sensors in hydrology, soil quality, environment and meteorology, which is not only conducive to the daily management and maintenance of the forest farm, but also important to the information development of forestry ecology.
(3) Forestry Vegetation Information Management
The division and management of vegetation in the forest farm still stays on the drawings, so the management of a wide range of forest farm is not intuitive and accurate, and all attribute information of each vegetation can not be recorded. It is not convenient for managers to grasp the overall macro of Forest Farm vegetation.
The forest resources supervision system realizes the three-dimensional display, query and statistics of vegetation through vegetation information management. All attribute information of vegetation can be displayed by clicking different vegetation layers, and the statistics of area and quantity can be realized.
The beautiful natural environment has been people’s yearning since ancient times. People are closely connected with nature. Through information technology, we have a “dialogue” with every tree and forest, accurately “perceive” their “joys and sorrows”, understand their “clothing, food, housing and transportation”, and take good care of every shade. The attempt of cloud sensing Internet of things is just the beginning. Dare to innovate, dare to advance to the “no man’s land” for the application of new technologies, speed up the construction of “smart forestry”, and the ecological forest will usher in infinite scenery.
Responsible editor: GT