Forest Monitoring System Research Articles

UAV-based Forest Fire Exploration and Monitoring System

The goal of this work is to develop a system for monitoring forest fires in real time using an unmanned aerial vehicle (UAV). This study proposes a quadcopter-based forest fire monitoring system based on radio control from a ground station using a geolocation system. This approach ensures the effective use of UAVs both in automatic and manual modes with the help of onboard navigation equipment and controls. The hardware system is built using the ArduPilot controller and various sensors such as GPS, gyro compass, temperature sensor, distance sensor and digital cameras. Flight tests were carried out to assess the accuracy of the positioning system. The results show that the use of UAVs is an effective way to deal with a fire situation in addition to existing methods and technologies for solving operational problems of fire detection.

An optimization-based approach for an integrated forest fire monitoring system with multiple technologies and surveillance drones

Wildfires pose a significant threat to forests and the delicate natural balance. In this work, we present a novel optimization approach for designing forest fire monitoring systems that incorporate surveillance towers, monitoring balloons, and drone technology, which have gained popularity in recent years for monitoring, logistics, and civil applications. We develop a compact mixed-integer linear programming model that includes location decisions for multiple monitoring technologies and routing of drones among monitoring facilities to achieve the most extensive possible terrain coverage. Additionally, we propose a matheuristic algorithm that comprises six components: a solution procedure, perturbation procedures, local search procedures, a call to the general-purpose solver to solve a mixed-integer linear programming model, a global reset strategy, a local reset strategy, and an acceptance criterion. We test the proposed model and algorithm on a set of random instances and a real-life case study in Chile. While the proposed model can only solve small instances, the matheuristic can find good-quality solutions for all instances. This study can aid the government and private sector in designing an integrated fire monitoring system that leverages watchtowers, monitoring balloons, and drones.

Validation and limitations of large-scale forest condition indicators – An example from Hungary

As the multi-scale study of biodiversity is extremely resource-intensive, proxy indicators taken from various databases are often used to answer questions on a larger spatial scale. Considering differences in the scale and methods, the application of such indicators (especially from different monitoring systems) requires careful consideration and standard methods of validation.In order to demonstrate this, we validated the results of the MAES-HU (National Ecosystem Assessment of Hungary) forest condition assessment (based on the Hungarian NFD – National Forestry Database) with thematically richer, finer-scale field data. We also examined the relationships of some MAES-HU indicators with other variables significant for nature conservation, not currently included in the NFD. We found the MAES-HU scoring was similar to the fine-scale score results in the case of tree species composition indicators, however, less so for structural indicators. The MAES-HU assessment uses the values of the NFD, averaged for forest management units, and thus tends to underestimate structural variety. This highlights a potential loss of important conservation-related information. During the examination of relationships with other indicators that are not included in the large-scale MAES-HU assessment, we found that the presence of large-diameter and old trees correlates with tree-related microhabitats and large standing deadwood, but no relationship was found for other investigated indicators (game pressure, further deadwood indicators). This highlights the need for integrating some key conservation indicators (presence of old and large trees, quantity and quality of standing and lying deadwood) into existing forest monitoring systems in order to optimise the resources dedicated to multipurpose data collection. Our study also highlights that applying indicators as proxies requires the full knowledge of monitoring methods and validated indicator-indicanda relationships.

Preferred vector machine for forest fire detection

Machine learning-based fire detection/recognition is very popular in forest-monitoring systems. However, without considering the prior knowledge, e.g., equal attention on both classes of the fire and non-fire samples, fire miss-detected phenomena frequently appeared in the current methods. In this work, considering model’s interpretability and the limited data for model-training, we propose a novel pixel-precision method, termed as PreVM (Preferred Vector Machine). To guarantee high fire detection rate under precise control, a new L0 norm constraint is introduced to the fire class. Computationally, instead of the traditional L1 re-weighted techniques in L0 norm approximation, this L0 constraint can be converted into linear inequality and incorporated into the process of parameter selection. To further speed up model-training and reduce error warning rate, we also present a kernel-based L1 norm PreVM (L1-PreVM). Theoretically, we firstly prove the existence of dual representation for the general Lp (p≥1) norm regularization problems in RKHS (Reproducing Kernel Hilbert Space). Then, we provide a mathematical evidence for L1 norm kernelization to conquer the case when feature samples do not appear in pairs. The work also includes an extensive experimentation on the real forest fire images and videos. Compared with the-state-of-art methods, the results show that our PreVM is capable of simultaneously achieving higher fire detection rates and lower error warning rates, and L1-PreVM is also superior in real-time detection.

Alignment of National Forest Policy Frameworks with the EU Timber Regulation Requirements: Insights from Montenegro and the Republic of Srpska (Bosnia and Herzegovina)

The Western Balkans represent a priority area for improving forest legality monitoring systems in line with the European Union Timber Regulation (EUTR). However, research on EUTR implementation in Western Balkan countries is still sporadic with a limited geographical scope; therefore, the preparedness of forestry sector actors for the EUTR in the region is largely unknown. The main objective of this study is to determine to what extent the forest policy frameworks of Montenegro and the Republic of Srpska (Bosnia and Herzegovina) are aligned with the EUTR requirements. To achieve this aim, we applied a qualitative content analysis of policy documents identified via an expert-based approach. Our results show that both countries have well-developed policy frameworks addressing illegal logging and preventing illegal activities in forestry, especially through dedicated action plans. Key actors in both countries are public, including the ministries responsible for forestry, public forest enterprises, and forestry inspectorates. The forestry sector in Montenegro is facing significant changes due to the termination of forest concessions and the reorganization of the management of state forests, including forest certification. The Republic of Srpska has relatively well-established institutional bodies for EUTR implementation but, in some cases, insufficient exchange of information and cooperation among them. Our findings indicate that the forestry sectors in Montenegro and the Republic of Srpska (as well as in Serbia, Croatia, and Slovenia, as per previous research) are dynamic, undergoing various changes, so there is room for improvement in terms of capacities (e.g., human, technological, infrastructural), legal responsibilities, and information access and availability. With an increasing focus on “deforestation-free” commodities within the EU and global policy arena, a new, more demanding, and broader regulation is expected at the EU level, replacing the EUTR. The incoming regulation will expand existing EUTR requirements, likely posing severe challenges to many EU member countries. This could be even more challenging for countries with less developed or advanced systems to enforce legality requirements, including Western Balkan countries.

Realization of Forest Internet of Things Using Wireless Network Communication Technology of Low-Power Wide-Area Network.

This work implements an intelligent forest monitoring system using the Internet of things (IoT) with the wireless network communication technology of a low-power wide-area network (LPWAN), a long range (LoRa), and a narrow-band Internet of things (NB-IoT). A solar micro-weather station with LoRa-based sensors and communications was built to monitor the forest status and information such as the light intensity, air pressure, ultraviolet intensity, CO2, etc. Moreover, a multi-hop algorithm for the LoRa-based sensors and communications is proposed to solve the problem of long-distance communication without 3G/4G. For the forest without electricity, we installed solar panels to supply electricity for the sensors and other equipment. In order to avoid the problem of insufficient solar panels due to insufficient sunlight in the forest, we also connected each solar panel to a battery to store electricity. The experimental results show the implementation of the proposed method and its performance.

A tree-level analysis of baboon damage in commercial forest stands using deep learning techniques

Commercial forest plantations in South Africa are homogeneous monocultures of highly bred exotic species grown to deliver timber products of the best potential quality. As such, these stands are susceptible to adverse effects of biotic and abiotic factors, and therefore require intense management to mitigate these risks. A sustainable forest monitoring system that can detect real-time changes in the physiological state of these plantations is needed for timeous management intervention to reduce losses. The use of machine learning algorithms has recently become popular, with acceptable levels of success. This study explores the application of deep learning neural networks for early detection of damage caused by baboons in evergreen plantations of Pinus species. Using PlanetScope imagery (spectral band 590–860 nm), which is captured by a constellation of Dove nanosatellites, with a high temporal resolution available daily at 3 m spatial resolution, the study achieved an overall accuracy of 81.54%, with a kappa value of 0.69, using a deep neural network. In comparison, using a random-forest classifier produced 74.04% accuracy and a kappa value of 0.62. The study successfully mapped different levels of baboon damage within commercial pine forests. We provide a repeatable method for daily monitoring initiatives, and attest to the utility of higher-resolution imagery such as PlanetScope for mapping health and damage severity at the tree level.

Texture based prototypical network for few-shot semantic segmentation of forest cover: Generalizing for different geographical regions

Forest plays a vital role in reducing greenhouse gas emissions and mitigating climate change, besides maintaining the world’s biodiversity. The existing satellite-based forest monitoring system utilizes supervised learning approaches limited to a particular region and depends on manually annotated data to identify forest. This work envisages forest identification as a few-shot semantic segmentation task to achieve generalization across different geographical regions. The proposed few-shot segmentation approach incorporates a texture attention module in the prototypical network to highlight the texture features of the forest. Indeed, the forest exhibits a characteristic texture different from other classes, such as road, water, etc. In this work, the proposed approach is trained for identifying tropical forests of South Asia and adapted to determine the temperate forest of Central Europe with the help of a few (one image for 1-shot) manually annotated support images of the temperate forest. An IoU of 0.62 for forest class (1-way 1-shot) was obtained using the proposed method, which is significantly higher (0.46 for PANet) than the existing few-shot semantic segmentation approach. Besides, the experimental results demonstrate that the inclusion of the texture attention module in the existing prototypical few-shot segmentation methods (PFENet and ASGNet) results in a more accurate forest identification. These results indicate that the proposed approach can generalize across geographical regions for forest identification, creating an opportunity to develop a global forest cover identification tool.

3D Drone Mapping

A major portion of the earth remains unexplored today. There are many reasons for this. One of them is because those places are or maybe dangerous for the explorers. Also, Nepal and many other under-developed countries buy satellite images and maps of their territory from other countries. The problem with these maps is that they are not always up-to-date and are very low-resolution images. In order to solve these problems, this research work has designed a quadcopter (four propellers) drone with HD camera. This device is capable of capturing images with colour, intensity, and depth values at each pixel so that a 3D map can be constructed. Unmanned Aerial Vehicles (UAVs) also called Drones can be designed for numerous objectives and tasks. This article intends to design a drone capable of creating its own pathway and taking flight autonomously. During the flight, the UAV is programmed to capture high-resolution images at regular intervals with 60-80% overlap. The overlap between images is highly necessary to obtain numerous data of the same point, which will be beneficial in creating a finer 3D map. The 3D map is generated by using a process called photogrammetry, which takes pictures as input and displays a 3D map as output by utilizing the data present in pictures. This drone can be used to capture pictures and create 3D maps of unexplored areas like caves, volcano craters, etc. Also, high-resolution pictures can be merged and uploaded for satellite imaging applications. Besides, the same vehicle can be used for many other applications like forest fire monitoring systems, endangered wildlife monitoring, surveying, and so on with slight alterations.

Forest Segmentation with Spatial Pyramid Pooling Modules: A Surveillance System Based on Satellite Images

The global deforestation rate continues to worsen each year, and will eventually lead to various negative consequences for humans and the environment. It is essential to develop an effective forest monitoring system to detect any changes in forest areas, in particular, by monitoring the progress of forest conservation efforts. In general, changes in forest status are difficult to annotate manually, whereby the boundaries can be small in size or hard to discern, especially in areas that are bordering residential areas. The previously implemented forest monitoring systems were ineffective due to their use of low-resolution satellite images and the inefficiency of drone-based data that offer a limited field of view. Most government agencies also still rely on manual annotation, which makes the monitoring process time-consuming, tedious, and expensive. Therefore, the goal of this study is to overcome these issues by developing a forest monitoring system that relies on a robust deep semantic segmentation network that is capable of discerning forest boundaries automatically, so that any changes over the years can be tracked. The backbone of this system is based on satellite imaging supplied to a modified U-Net deep architecture to incorporate multi-scale modules to deliver the semantic segmentation output. A dataset of 6048 Landsat-8 satellite sub-images that were taken from eight land parcels of forest areas was collected and annotated, and then further divided into training and testing datasets. The novelty of this system is the optimal integration of the spatial pyramid pooling (SPP) mechanism into the base model, which allows the model to effectively segment forest areas regardless of their varying sizes, patterns, and colors. To investigate the impact of SPP on the forest segmentation system, a set of experiments was conducted by integrating several variants of SPP ranging from two to four parallel paths with different combinations of pooling kernel size, placed at the bottleneck layer of the U-Net model. The results demonstrated the effectiveness of the SPP module in improving the performance of the forest segmentation system by 2.57%, 6.74%, and 7.75% in accuracy (acc), intersection over union (IoU), and F1-score (F1score), respectively. The best SPP variant consists of four parallel paths with a combination of pooling kernel sizes of 2×2, 4×4, 6×6, and 8×8 pixels that produced the highest acc, IoU, and F1score of 86.71%, 75.59%, and 82.88%, respectively. As a result, the multi-scale module improved the proposed forest segmentation system, making it a highly useful system for government and private agencies in tracking any changes in forest areas.

MAPPING LAND COVER TIME SERIES USING LANDSAT-8 AND SENTINEL-1 IN SOUTH KALIMANTAN

Abstract. Remote sensing has been widely used for forest monitoring. However, most forest monitoring systems largely rely on optical images that limit temporal analyses due to cloud cover, particularly in the tropics. The current study used integration of optical Landsat-8 and Sentinel-1 SAR to produce individual year land cover classification maps for Kalimantan, Indonesia that differentiate between native forest and tree plantations, such as oil palm and rubber. We applied a Bayesian network to produce a time series of land cover classification that improved accuracy of individual year land cover maps. Accuracy assessment using a confusion matrix showed that final map had overall accuracy of 90%, while user's and producer's accuracy for each land cover class was above 85%, except non–forest, which had 76% producer's accuracy due to errors in the classification between young rubber plantations and non–forest. Improved maps will support Indonesia's national forest monitoring system and sustainable forest management.

Data Acquisition Filtering Focused on Optimizing Transmission in a LoRaWAN Network Applied to the WSN Forest Monitoring System.

Developing innovative systems and operations to monitor forests and send alerts in dangerous situations, such as fires, has become, over the years, a necessary task to protect forests. In this work, a Wireless Sensor Network (WSN) is employed for forest data acquisition to identify abrupt anomalies when a fire ignition starts. Even though a low-power LoRaWAN network is used, each module still needs to save power as much as possible to avoid periodic maintenance since a current consumption peak happens while sending messages. Moreover, considering the LoRaWAN characteristics, each module should use the bandwidth only when essential. Therefore, four algorithms were tested and calibrated along real and monitored events of a wildfire. The first algorithm is based on the Exponential Smoothing method, Moving Averages techniques are used to define the other two algorithms, and the fourth uses the Least Mean Square. When properly combined, the algorithms can perform a pre-filtering data acquisition before each module uses the LoRaWAN network and, consequently, save energy if there is no necessity to send data. After the validations, using Wildfire Simulation Events (WSE), the developed filter achieves an accuracy rate of 0.73 with 0.5 possible false alerts. These rates do not represent a final warning to firefighters, and a possible improvement can be achieved through cloud-based server algorithms. By comparing the current consumption before and after the proposed implementation, the modules can save almost 53% of their batteries when is no demand to send data. At the same time, the modules can maintain the server informed with a minimum interval of 15 min and recognize abrupt changes in 60 s when fire ignition appears.

Inter-comparison of optical and SAR-based forest disturbance warning systems in the Amazon shows the potential of combined SAR-optical monitoring

ABSTRACT More than half a decade after the launch of the Sentinel-1A C-band SAR satellite, several near real-time forest disturbances detection systems based on backscattering time series analysis have been developed and made operational. Every system has its own particular approach to change detection. Here, we have compared the performance of the main SAR-based near real-time operational forest disturbance detection systems produced by research agencies (INPE, in Brazil, CESBIO, in France, JAXA, in Japan, and Wageningen University, in the Netherlands), and compared them to the state-of-the-art optical algorithm, University of Maryland’s GLAD-S2. We implemented an innovative validation protocol, specially conceived to encompass all the analysed systems, which measured every system’s accuracy and detection speed in four different areas of the Amazon basin. The results indicated that, when parametrized equally, all the Sentinel-1 SAR methods outperformed the reference optical method in terms of sample-count F1-Score, having comparable results among them. The GLAD-S2 optical method showed superior results in terms of user’s accuracy (UA), issuing no false detections, but had a lower producer accuracy (PA, 84.88%) when compared to the Sentinel-1 SAR-based systems (PA90%). Wageningen University’s system, RADD, proved to be relatively faster, especially in heavily clouded regions, where RADD warnings were issued 41 days before optical ones, and the one that better performs on small disturbed patches (0.25 ha) with a UA of 70.11%. Of all the high-resolution SAR methods, CESBIO’s had the best results regarding UA (99.0%). Finally, we tested the potential of three hypothetical combined optical-SAR systems. The results show that these combined systems would have excellent detection capabilities, exceeding largely the producer’s accuracy of all the tested methods at the cost of a slightly diminished user’s accuracy, and constitute a promising and feasible approach for the forthcoming forest monitoring systems.

Optimizing Forest Fire Prevention: Intelligent Scheduling Algorithms for Drone-Based Surveillance System

Given the importance of forests and their role in maintaining the ecological balance, which directly affects the planet, the climate, and the life on this planet, this research presents the problem of forest fire monitoring using drones. The forest monitoring process is performed continuously to track any changes in the monitored region within the forest. During fires, drones’ capture data is used to increase the follow-up speed and enhance the control process of these fires to prevent their spread. The time factor in such problems determines the success rate of the fire extinguishing process, as appropriate data at the right time may be the decisive factor in controlling fires, preventing their spread, extinguishing them, and limiting their losses. Therefore, this research presented the problem of monitoring task scheduling for drones in the forest monitoring system. This problem is solved by developing several algorithms with the aim of minimizing the total completion time required to carry out all the drones’ assigned tasks. System performance is measured by using 990 instances of three different classes. The performed experimental results indicated the effectiveness of the proposed algorithms and their ability to act efficiently to achieve the desired goal. The algorithm RID achieved the best performance with a percentage rate of up to 90.3% with a time of 0.088 seconds.

Retracted: An Early Warning Intelligent Algorithm System for Forest Resource Management and Monitoring.

[This retracts the article DOI: 10.1155/2022/4250462.].

Effect of Deforestation on Land Surface Temperature in the Chiquitania Region, Bolivia

Neotropical forests offer alternatives to surface cooling and their conservation is an effective solution for mitigating the effects of climate change. Little is known about the importance of tropical dry forests for temperature regulation in Chiquitania, a region with increasing deforestation rates. The impact that deforestation processes are having on the surface temperature in Chiquitania remains an open question. This study evaluated trends in forest cover loss based on land surface temperatures (°C) in forested and deforested areas in Chiquitania. We hypothesized a positive relationship between higher deforestation and a temperature increase, which would decrease the resilience of highly disturbed Chiquitano forests. We evaluated ten sampling sites (10 × 10 km), including five in forested areas with some type of protection and the other five in areas with populated centers and accelerated forest loss. We developed scripts on the Google Earth Engine (GEE) platform using information from the Normalized Difference Vegetation Index (NDVI, MOD13A2) and the daytime and nighttime Land Surface Temperature (LST, MYD11A1) from MODIS products for the period 2001–2021. The statistical significance of the trends of the time series averages of the MODIS products was analyzed using a nonparametric Mann–Kendall test and the degree of the relationship between the variables was determined using the Pearson statistic. Our results based on NDVI analysis showed consistent vegetation growth in forested areas across the study period, while the opposite occurred in deforested lands. Regarding surface temperature trends, the results for daytime LST showed a positive increase in the four deforested areas. Comparatively, daytime LST averages in deforested areas were warmer than those in forested areas, with a difference of 3.1 °C. Additionally, correlation analyses showed a significant relationship between low NDVI values due to deforestation in three sites and an increase in daytime LST, while for nighttime LST this phenomenon was registered in two deforested areas. Our results suggest a significant relationship between the loss of forest cover and the increase in land surface temperature in Chiquitania. This study could be the first step in designing and implementing an early climate–forest monitoring system in this region.

Measurement uncertainty in a national forest inventory: results from the northern region of the USA

Statistical confidence in estimates of timber volume, carbon storage, and other forest attributes depends, in part, on the uncertainty in field measurements. Surprisingly, measurement uncertainty is rarely reported, even though national forest inventories routinely repeat field measurements for quality assurance. We compared measurements made by field crews and quality assurance crews in the Forest Inventory and Analysis program of the U.S. Forest Service, using data from 2790 plots and 51 740 trees and saplings across the 24 states of the Northern Region. We characterized uncertainty in 12 national core tree-level variables; seven tree crown variables used in forest health monitoring; three variables describing seedlings; and 11 variables describing the site, such as elevation, slope, and distance from a road. Discrepancies in measurement were generally small but were higher for some variables requiring judgment, such as tree class, decay class, and cause of mortality. When scaled up to states, forest types, or the region, uncertainties in basal area, timber volume, and aboveground biomass were negligible. Understanding all sources of uncertainty is important to designing forest monitoring systems, managing the conduct of the inventory, and assessing the uncertainty of forest attributes required for making regional and national forest policy decisions.

Large-scale high-resolution yearly modeling of forest growing stock volume and above-ground carbon pool

Within the Paris Agreement's Enhanced Transparency Framework, consistent data collections are the prerequisite for a successful reporting of GHG emissions. For such purposes, NFIs are usually the primary source of information, even if they are frequently not designed for producing estimations on a yearly basis and in the form of wall-to-wall high-resolution maps. In this framework, we present a new spatial model to produce yearly growing stock volume (GSV), above-ground biomass (AGB), and carbon stock wall-to-wall estimates. We tested the model in Italy for the period 2005–2018, obtaining a time-series of yearly maps at 23 m spatial resolution. Results were validated against the 2015 Italian NFI reaching an average RMSE% of 19% for aggregated areas. Results were also compared against data reported by the Italian GHG inventory, reaching an RMSE% of 28% and 20% for GSV and carbon stock respectively.We demonstrated that the modeling approach can be successfully used for setting up a forest monitoring system to meet the interests of governments in inventories of GHG emissions and private entities in carbon offset investments.

High-Precision Real-Time Forest Fire Video Detection Using One-Class Model

Machine learning-base fire detection methods play a vital role in the current forest fire monitoring systems. In the literature, fire detection is usually viewed as a two-class (or multi-class) classification task. However, this violates the most foundational hypothesis in machine learning, e.g., independent and identical distribution (i.i.d.), especially for the non-fire samples drawn from a complex forest background. Moreover, for omni-directional video -monitoring, the background is also always changing, which leads this violation to a worse situation. In this work, by relaxing the i.i.d. of non-fire samples, we aim to learn a one-class model that just relies on the fire samples. Considering the requirements of high-precision and real-time detection, training samples are directly constructed on the fire pixels without a complex feature transformation. Additionally, we also provide a batch decision-making strategy to speed up fire detection. This work also includes an extensive experimental comparison on the public forest fire videos, obtained by ground- or unmanned aerial vehicle (UAV)-monitoring cameras. Compared with the state-of-the-art methods, the results show the superiority of our proposal in terms of a high-fire detection rate, low-error warning rate, accurate fire location positioning, and real-time detection.

An Early Warning Intelligent Algorithm System for Forest Resource Management and Monitoring.

The development of remote sensing technology has passed an effective means for forest resource management and monitoring, but remote sensing technology is limited by sensor hardware equipment, and the quality of remote sensing image data is low, which is difficult to meet the needs of forest resource change monitoring. This paper presents a remote sensing image classification method based on the combination of the SSIF algorithm and wavelet denoising. Forest information is extracted from PALSAR/PALSAR-2 radar remote sensing data. The forest distribution map is generated by pixel level fusion algorithm, and the accuracy of the forest distribution map is evaluated by a confusion matrix. The remote sensing image is spatio-temporal fused by the SSIF algorithm to capture more details of forest distribution. The simulation analysis shows that the overall accuracy of the forest classification results obtained by the fusion algorithm is 96% ± 1, and the kappa coefficient is 0.66. The accuracy of forest recognition meets the requirements.