SiDroForest: A comprehensive forest inventory of Siberian boreal forest investigations including drone-based point clouds, individually labelled trees, synthetically generated tree crowns and Sentinel-2 labelled image patches

Abstract. This data collection is an attempt to remedy the scarcity of tree level forest structure data in the circum-boreal region, whilst providing, as part of the data collection, adjusted and labelled tree level and vegetation plot level data for machine learning and upscaling practices. Publicly available comprehensive datasets on tree level forest structure are rare, due to the involvement of governmental agencies, public sectors, and private actors that all influence the availability of these datasets. We present datasets of vegetation composition and tree and plot level forest structure for two important vegetation transition zones in Siberia, Russia; the summergreen–evergreen transition zone in central Yakutia and the tundra–taiga transition zone in Chukotka (NE Siberia). The SiDroForest collection contains a variety of data mainly based on unmanned aerial vehicle (UAV) and field data collected from 64 vegetation plots during fieldwork jointly performed by the Alfred Wegener Institute for Polar and Marine Research (AWI) and the North-Eastern Federal University of Yakutsk (NEFU) during the Chukotka 2018 expedition to Siberia. The data collection consists of four separate datasets. The fieldwork locations are the anchors that bind the data types together based on the location of the vegetation plot. i) The first dataset (Kruse et al., 2021, https://doi.pangaea.de/10.1594/PANGAEA.933263) provides UAV-borne data products covering the 64 vegetation plots surveyed during fieldwork: including structure from motion (SfM) point clouds, point-cloud products such as Digital Elevation Model (DEM), Canopy Height Model (CHM), Digital Surface Model (DSM) and Digital Terrain Model (DTM) constructed from Red Green Blue (RGB) and Red Green Near Infrared (RGN) orthomosaics. Forest structure and vegetation composition data are crucial in the assessment of whether a forest is to act as a carbon sink under changing climate conditions. Fieldwork and UAV-products can provide such data in depth. ii) The second dataset contains spatial data in the form of points and polygon shape files of 872 labelled individual trees and shrubs that were recorded during fieldwork at the same vegetation plots with information on tree height, crown diameter, and species (van Geffen et al., 2021c, https://doi.pangaea.de/10.1594/PANGAEA.932821). These tree- and shrub-individual labelled point and polygon shape files were generated and are located on the UAV RGB orthoimages. The individual number links to the information collected during the expedition such as tree height, crown diameter and vitality provided in table format. This dataset can be used to link individual trees in the SfM point clouds, providing unique insights into the vegetation composition and also allows future monitoring of the individual trees and the contents of the recorded vegetation plots at large. iii) The third dataset contains a synthesis of 10 000 generated images and masks that have the tree crowns of two species of larch (Larix gmelinii and Larix cajanderi) automatically extracted from the RGB UAV images in the common objects in context (COCO) format (van Geffen et al., 2021a, https://doi.pangaea.de/10.1594/PANGAEA.932795). The synthetic dataset was created specifically to detect Siberian larch species. iv) If publicly available forest-structure datasets at tree level are rarely available for Siberia, even fewer ready-to-use tree and plot level data are available for machine learning approaches, for example optimised data formats containing annotated vegetation categories. The fourth set contains Sentinel-2 Level-2 bottom of atmosphere labelled image patches with seasonal information and annotated vegetation categories covering the vegetation plots (van Geffen et al., 2021b, https://doi.pangaea.de/10.1594/PANGAEA.933268). The dataset is created with the aim of providing a small ready-to use validation and training data set to be used in various vegetation-related machine-learning tasks. The SidroForest data collection serves a variety of user communities. First of all, the UAV-derived top of canopy structure information, orthomosaics and the detailed vegetation information in the labelled data set provide detailed information on forest type, structure and composition for scientific communities with ecological and biological applications. The detailed Land Cover and Vegetation structure information in the first two data sets are of use for the generation and validation of Land Cover remote sensing products in radar and optical remote sensing. In addition to providing information on forest structure and vegetation composition of the vegetation plots, parts of the SiDroForest dataset are prepared to be used as training and validation data for machine learning purposes. For example, the Synthetic tree crown dataset is generated from the raw UAV images and optimized to be used in neural networks. Furthermore, the fourth SiDroForest data set contains standardized Sentinel-2 labelled image patches that provide training data on vegetation class categories for machine learning classification with JSON labels provided. The SiDroForst data collective serves as a basis to add future data collected during expeditions performed by the Alfred Wegener Institute, creating a larger dataset in the upcoming years that can provide unique insights into remote hard to reach boreal regions of Siberia.

Download Full-text

SiDroForest Siberian Drone-mapped Forest inventory

10.5194/egusphere-egu21-15106 ◽

2021 ◽

Author(s):

Femke van Geffen ◽

Birgit Heim ◽

Ulrike Herzschuh ◽

Luidmila Pestryakova ◽

Evgenii Zakharov ◽

...

Keyword(s):

Machine Learning ◽

Satellite Data ◽

Forest Structure ◽

Vegetation Index ◽

Point Clouds ◽

Training Data ◽

Data Repository ◽

Surface Model ◽

Marine Research ◽

Sentinel 2

To gain a better understanding of global carbon storage and albedo feedback mechanisms it is important to have insights into high latitude vegetation change. Boreal forest compositions are changing in response to changes in climate, which in turn can lead to feedbacks in regional and global climate through altered carbon cycles and albedo dynamics. Circumpolar boreal forests represent close to 30% of all forested area on the planet, between 900 and 1,200 million ha. These forests are located primarily in Alaska, Canada, and Russia. Due to the remote location of these forests and the short seasons without snow, data collected on the boreal vegetation is limited.&#160;The proposed dataset is an attempt to remedy data scarcity whilst providing adjusted data for machine learning practices.We present a dataset containing diverse formats of forest structure information that covers two important vegetation transition zones in Siberia: the Evergreen - Summergreen transition zone in Central Yakutia and the northern treeline in Chukotka (NE Siberia).This dataset contains data from the locations covered by fieldwork was performed by the Alfred Wegener Institute for Polar and Marine research, (AWI) and the North-Eastern Federal University of Yakutsk&#8203; (NEFU). The fieldwork upscaled through the addition of Red Green Blue(RGB) UAV (Unmanned Aerial Vehicle) camera data and Sentinel-2 satellite data cropped to a 5 km radius around the fieldwork sites. The dataset is created with the aim of providing ground truth validation and training data to be used in various vegetation related machine learning tasks .The dataset contains:1.Labelled individual trees per 30x30 m plot assigned in field work with additional data on species, height, crown width, and biomass.2.Structure from Motion (SfM)point clouds that provide 3D information about the forest structure, included generated Canopy Height Model (CHM), Digital Elevation Model (DEM) and a Digital Surface Model (DSM) per 50x50 m.3.Multispectral Sentinel-2 satellite data (10 m ) cropped to a 5km radius with generated a NDVI(normalized difference vegetation index), available in three seasons: Early Summer, Peak Summer and Late Summer.4.Extracted tree crowns with species information and a synthetically generated large (10.000 samples) dataset for training machine leaning algorithms.The dataset will be made publicly available on&#160;the data repository&#160;PANGAEA.

Download Full-text

Detection of Very Small Tree Plantations and Tree-Level Characterization Using Open-Access Remote-Sensing Databases

Remote Sensing ◽

10.3390/rs12142276 ◽

2020 ◽

Vol 12 (14) ◽

pp. 2276

Author(s):

Laura Alonso ◽

Juan Picos ◽

Guillermo Bastos ◽

Julia Armesto

Keyword(s):

Open Access ◽

Point Cloud ◽

Castanea Sativa ◽

Point Clouds ◽

Coefficient Of Determination ◽

National Database ◽

Tree Level ◽

Tree Plantations ◽

Support Vector ◽

Individual Trees

Highly fragmented land property hinders the planning and management of single species tree plantations. In such situations, acquiring information about the available resources is challenging. This study aims to propose a method to locate and characterize tree plantations in these cases. Galicia (Northwest of Spain) is an area where property is extremely divided into small parcels. European chestnut (Castanea sativa) plantations are an important source of income there; however, it is often difficult to obtain information about them due to their small size and scattered distribution. Therefore, we selected a Galician region with a high presence of chestnut plantations as a case study area in order to locate and characterize small plantations using open-access data. First, we detected the location of chestnut plantations applying a supervised classification for a combination of: Sentinel-2 images and the open-access low-density Light Detection and Ranging (LiDAR) point clouds, obtained from the untapped open-access LiDAR Spanish national database. Three classification algorithms were used: Random Forest (RF), Support Vector Machine (SVM), and XGBoost. We later characterized the plots at the tree-level using the LiDAR point-cloud. We detected individual trees and obtained their height applying a local maxima algorithm to a point-cloud-derived Canopy Height Model (CHM). We also calculated the crown surface of each tree by applying a method based on two-dimensional (2D) tree shape reconstruction and canopy segmentation to a projection of the LiDAR point cloud. Chestnut plantations were detected with an overall accuracy of 81.5%. Individual trees were identified with a detection rate of 96%. The coefficient of determination R2 value for tree height estimation was 0.83, while for the crown surface calculation it was 0.74. The accuracy achieved with these open-access databases makes the proposed procedure suitable for acquiring knowledge about the location and state of chestnut plantations as well as for monitoring their evolution.

Download Full-text

IMPROVING LULC CLASSIFICATION FROM SATELLITE IMAGERY USING DEEP LEARNING – EUROSAT DATASET

ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences ◽

10.5194/isprs-archives-xliii-b3-2021-369-2021 ◽

2021 ◽

Vol XLIII-B3-2021 ◽

pp. 369-376

Author(s):

H. Yassine ◽

K. Tout ◽

M. Jaber

Keyword(s):

Machine Learning ◽

Deep Learning ◽

Satellite Imagery ◽

Vegetation Index ◽

Near Infrared ◽

Spectral Feature ◽

Classification Performance ◽

Training Data ◽

Data Set ◽

Sentinel 2

Abstract. Machine learning (ML) has proven useful for a very large number of applications in several domains. It has realized a remarkable growth in remote-sensing image analysis over the past few years. Deep Learning (DL) a subset of machine learning were applied in this work to achieve a better classification of Land Use Land Cover (LULC) in satellite imagery using Convolutional Neural Networks (CNNs). EuroSAT benchmarking data set is used as training data set which uses Sentinel-2 satellite images. Sentinel-2 provides images with 13 spectral feature bands, but surprisingly little attention has been paid to these features in deep learning models. The majority of applications focused only on using RGB due to high availability of the RGB models in computer vision. While RGB gives an accuracy of 96.83% using CNN, we are presenting two approaches to improve the classification performance of Sentinel-2 images. In the first approach, features are extracted from 13 spectral feature bands of Sentinel-2 instead of RGB which leads to accuracy of 98.78%. In the second approach features are extracted from 13 spectral bands of Sentinel-2 in addition to calculated indices used in LULC like Blue Ratio (BR), Vegetation index based on Red Edge (VIRE) and Normalized Near Infrared (NNIR), etc. which gives a better accuracy of 99.58%.

Download Full-text

Comparison of Backpack, Handheld, Under-Canopy UAV, and Above-Canopy UAV Laser Scanning for Field Reference Data Collection in Boreal Forests

Remote Sensing ◽

10.3390/rs12203327 ◽

2020 ◽

Vol 12 (20) ◽

pp. 3327 ◽

Cited By ~ 1

Author(s):

Eric Hyyppä ◽

Xiaowei Yu ◽

Harri Kaartinen ◽

Teemu Hakala ◽

Antero Kukko ◽

...

Keyword(s):

Data Collection ◽

Point Cloud ◽

Laser Scanning ◽

Reference Data ◽

Tree Height ◽

Tree Level ◽

Stem Volume ◽

Point Cloud Data ◽

Cloud Data ◽

Individual Trees

In this work, we compared six emerging mobile laser scanning (MLS) technologies for field reference data collection at the individual tree level in boreal forest conditions. The systems under study were an in-house developed AKHKA-R3 backpack laser scanner, a handheld Zeb-Horizon laser scanner, an under-canopy UAV (Unmanned Aircraft Vehicle) laser scanning system, and three above-canopy UAV laser scanning systems providing point clouds with varying point densities. To assess the performance of the methods for automated measurements of diameter at breast height (DBH), stem curve, tree height and stem volume, we utilized all of the six systems to collect point cloud data on two 32 m-by-32 m test sites classified as sparse (n = 42 trees) and obstructed (n = 43 trees). To analyze the data collected with the two ground-based MLS systems and the under-canopy UAV system, we used a workflow based on our recent work featuring simultaneous localization and mapping (SLAM) technology, a stem arc detection algorithm, and an iterative arc matching algorithm. This workflow enabled us to obtain accurate stem diameter estimates from the point cloud data despite a small but relevant time-dependent drift in the SLAM-corrected trajectory of the scanner. We found out that the ground-based MLS systems and the under-canopy UAV system could be used to measure the stem diameter (DBH) with a root mean square error (RMSE) of 2–8%, whereas the stem curve measurements had an RMSE of 2–15% that depended on the system and the measurement height. Furthermore, the backpack and handheld scanners could be employed for sufficiently accurate tree height measurements (RMSE = 2–10%) in order to estimate the stem volumes of individual trees with an RMSE of approximately 10%. A similar accuracy was obtained when combining stem curves estimated with the under-canopy UAV system and tree heights extracted with an above-canopy flying laser scanning unit. Importantly, the volume estimation error of these three MLS systems was found to be of the same level as the error corresponding to manual field measurements on the two test sites. To analyze point cloud data collected with the three above-canopy flying UAV systems, we used a random forest model trained on field reference data collected from nearby plots. Using the random forest model, we were able to estimate the DBH of individual trees with an RMSE of 10–20%, the tree height with an RMSE of 2–8%, and the stem volume with an RMSE of 20–50%. Our results indicate that ground-based and under-canopy MLS systems provide a promising approach for field reference data collection at the individual tree level, whereas the accuracy of above-canopy UAV laser scanning systems is not yet sufficient for predicting stem attributes of individual trees for field reference data with a high accuracy.

Download Full-text

Under-Canopy UAV Laser Scanning Providing Canopy Height and Stem Volume Accurately

Forests ◽

10.3390/f12070856 ◽

2021 ◽

Vol 12 (7) ◽

pp. 856

Author(s):

Juha Hyyppä ◽

Xiaowei Yu ◽

Teemu Hakala ◽

Harri Kaartinen ◽

Antero Kukko ◽

...

Keyword(s):

Data Collection ◽

Standard Error ◽

Laser Scanning ◽

Reference Data ◽

Laser Scanner ◽

Point Clouds ◽

Canopy Height ◽

Stem Volume ◽

Scanning System ◽

Individual Trees

The automation of forest field reference data collection has been an intensive research objective for laser scanning scientists ever since the invention of terrestrial laser scanning more than two decades ago. In this study, we demonstrated that an under-canopy UAV laser scanning system utilizing a rotating laser scanner can alone provide accurate estimates of canopy height and stem volume for the majority of trees in a boreal forest. We mounted a rotating laser scanner based on a Velodyne VLP-16 sensor onboard a manually piloted UAV. The UAV was commanded with the help of a live video feed from the onboard camera. Since the system was based on a rotating laser scanner providing varying view angles, all important elements such as treetops, branches, trunks, and ground could be recorded with laser hits. In an experiment including two different forest structures, namely sparse and obstructed canopy, we showed that our system can measure the heights of individual trees with a bias of −20 cm and a standard error of 40 cm in the sparse forest and with a bias of −65 cm and a standard error of 1 m in the obstructed forest. The accuracy of the obtained tree height estimates was equivalent to airborne above-canopy UAV surveys conducted in similar forest conditions or even at the same sites. The higher underestimation and higher inaccuracy in the obstructed site can be attributed to three trees with a height exceeding 25 m and the reduced point density of these tree tops due to occlusion and the limited ranging capacity of the scanner. Additionally, we used our system to estimate the stem volumes of individual trees with a standard error at the level of 10%. This level of error is equivalent to the error obtained when merging above-canopy UAV laser scanner data with terrestrial point cloud data. The results show that we do not necessarily need a combination of terrestrial point clouds and point clouds collected using above-canopy UAV systems in order to accurately estimate the heights and the volumes of individual trees in reference data collection.

Download Full-text

Multiscale assessment of northern forest characteristics based on ultra-high resolution data

Abstracts of the ICA ◽

10.5194/ica-abs-1-246-2019 ◽

2019 ◽

Vol 1 ◽

pp. 1-1

Author(s):

Andrey Medvedev ◽

Arseny Kudikov ◽

Natalia Telnova ◽

Olga Tutubalina ◽

Elena Golubeva ◽

...

Keyword(s):

Remote Sensing ◽

High Resolution ◽

Forest Ecosystems ◽

Point Clouds ◽

Tree Canopy ◽

Digital Terrain ◽

Dense Point ◽

Individual Trees ◽

Very High ◽

Sentinel 2

Abstract. The algorithms for quantitative estimates of various structural and functional parameters of forest ecosystems, particularly boreal forests, on high resolution remote sensing data are actively developing since the mid-2000s. For monitoring of forest ecosystems located at the Northern limit of distribution, effective not only lidar data but also the optical data obtained by unmanned aerial vehicles (UAV’s) with ultra-low altitude photography and derived products resulting from modern algorithms for the photogrammetric processing.High-detail remote sensing from UAV’s is a key level of monitoring of Northern forests at a large-scale level, ensuring the correct transition from sub - satellite ground-based studies to thematic products obtained from multi-time Hyper-and multispectral data of medium and relatively high resolution (MODIS, LANDSAT, Sentinel-2).When planning and conducting specific case studies based on UAV data, special attention should be paid to the justification of the survey methodology. In particular, the choice of a strictly defined high-altitude echelon of the survey determines the recognition of the objects of study and the possibility of reliable determination of its properties and features. To study the parameters of forest ecosystems at the level of individual trees and at the level of forest plantations, we selected two different-height echelons of survey from ultra-low altitudes: from 50 m, which allowed us to obtain ultra-high-detailed data for each sample area provided by detailed ground-based studies with sub-tree account, and from 100 m-to obtain derived characteristics of forest communities within the area equivalent to 3 pixels of thematic MODIS products with a spatial resolution of 250 m. The data of optical survey with UAV were obtained in July 2018 for 22 plots located in the central part of the Kola Peninsula and representative of different types of North taiga stands and their dynamics under climate change.At the stage of preprocessing images were obtained dense point clouds, characterizing both vertical and horizontal structure of stands. Digital terrain and terrain models and tree canopy models were obtained after cloud filtering and classification. Algorithms of automated segmentation and classification have been developed and tested to obtain such characteristics of stands as the height of individual trees, the area of crown projections, the projective cover of the tree-shrub layer. The obtained characteristics are aggregated by cells of a regular network with the dimension corresponding to the spatial resolution of Sentinel-2 and Landsat-8 data.The main results of the works are digital spatial datasets for 22 sample plots: raw data with very high resolution imagery (optical images with very high resolution, dense point clouds, RGB-orthophoto) and create based on a thematic derivative products (digital terrain model, topography, tree canopy cover; map of the heights and projections of the crowns of trees, percent cover of tree and shrub vegetation).

Download Full-text

Under-Canopy UAV Laser Scanning Providing Canopy Height and Stem Volume Accurately

10.20944/preprints202104.0003.v1 ◽

2021 ◽

Author(s):

Juha Hyyppä ◽

Xiaowei Yu ◽

Teemu Hakala ◽

Harri Kaartinen ◽

Antero Kukko ◽

...

Keyword(s):

Data Collection ◽

Standard Error ◽

Laser Scanning ◽

Laser Scanner ◽

Point Clouds ◽

Canopy Height ◽

Stem Volume ◽

Scanning System ◽

Individual Trees ◽

Laser Scanning System

Automation of forest field reference data collection has been an intensive research objective for laser scanning scientists ever since the invention of terrestrial laser scanning more than two decades ago. Recently, it has been proposed that such automated data collection providing both the tree heights and stem curves would require a combination of above-canopy UAV point clouds and terrestrial point clouds. In this study, we demonstrate that an under-canopy UAV laser scanning system utilizing a rotating laser scanner can alone provide accurate estimates of the canopy height and the stem volume for the majority of the trees in a boreal forest. To this end, we mounted a rotating laser scanner based on a Velodyne VLP-16 sensor onboard a manually piloted UAV. The UAV was commanded with the help of a live video feed from the onboard camera of the UAV. Since the system was based on a rotating laser scanner providing varying view angles, all important elements such as treetops, branches, trunks, and ground could be recorded with laser hits. In an experiment including two different forest structures, namely sparse and obstructed canopy, we showed that our system can measure the heights of individual trees with a bias of -20 cm and a standard error of 40 cm in the sparse forest and with a bias of -65 cm and a standard error of 1 m in the obstructed forest. The accuracy of the obtained tree height estimates was equivalent to airborne above-canopy UAV surveys conducted in similar forest conditions. The higher underestimation and higher inaccuracy in the obstructed site can be attributed to three trees with a height exceeding 25 m and the applied laser scanning system VLP-16 that had a limited height measurement capacity when it comes to trees taller than 25 m. Additionally, we used our system to estimate the stem volumes of individual trees with a standard error at the level of 10%. This level of error is equivalent to the error obtained when merging above-canopy UAV laser scanner data with terrestrial point cloud data. Future research is needed for testing new sensors, for implementing autonomous operation inside canopies through collision avoidance and navigation through canopies, and for developing robust methods that work also with more complex forest structure. The results show that we do not necessarily need a combination of terrestrial point clouds and point clouds collected using above-canopy UAV systems in order to accurately estimate the heights and the volumes of individual trees.

Download Full-text

Monitoring Selective Logging in a Pine-Dominated Forest in Central Germany with Repeated Drone Flights Utilizing A Low Cost RTK Quadcopter

Drones ◽

10.3390/drones4020011 ◽

2020 ◽

Vol 4 (2) ◽

pp. 11 ◽

Cited By ~ 2

Author(s):

Christian Thiel ◽

Marlin M. Müller ◽

Christian Berger ◽

Felix Cremer ◽

Clémence Dubois ◽

...

Keyword(s):

Low Cost ◽

Three Dimensional ◽

Point Clouds ◽

Selective Logging ◽

Ease Of Use ◽

Tree Level ◽

Unmanned Aerial Systems ◽

3D Point Clouds ◽

Individual Trees ◽

New Generation

There is no doubt that unmanned aerial systems (UAS) will play an increasing role in Earth observation in the near future. The field of application is very broad and includes aspects of environmental monitoring, security, humanitarian aid, or engineering. In particular, drones with camera systems are already widely used. The capability to compute ultra-high-resolution orthomosaics and three-dimensional (3D) point clouds from UAS imagery generates a wide interest in such systems, not only in the science community, but also in industry and agencies. In particular, forestry sciences benefit from ultra-high-structural and spectral information as regular tree level-based monitoring becomes feasible. There is a great need for this kind of information as, for example, due to the spring and summer droughts in Europe in the years 2018/2019, large quantities of individual trees were damaged or even died. This study focuses on selective logging at the level of individual trees using repeated drone flights. Using the new generation of UAS, which allows for sub-decimeter-level positioning accuracies, a change detection approach based on bi-temporal UAS acquisitions was implemented. In comparison to conventional UAS, the effort of implementing repeated drone flights in the field was low, because no ground control points needed to be surveyed. As shown in this study, the geometrical offset between the two collected datasets was below 10 cm across the site, which enabled a direct comparison of both datasets without the need for post-processing (e.g., image matching). For the detection of logged trees, we utilized the spectral and height differences between both acquisitions. For their delineation, an object-based approach was employed, which was proven to be highly accurate (precision = 97.5%; recall = 91.6%). Due to the ease of use of such new generation, off-the-shelf consumer drones, their decreasing purchase costs, the quality of available workflows for data processing, and the convincing results presented here, UAS-based data can and should complement conventional forest inventory practices.

Download Full-text

Using machine learning to model the distribution of Vegetation Types across Norway

10.5194/egusphere-egu21-15211 ◽

2021 ◽

Author(s):

Lasse Torben Keetz ◽

Anders Bryn ◽

Peter Horvath ◽

Olav Skarpaas ◽

Lena Merete Tallaksen ◽

...

Keyword(s):

Machine Learning ◽

Vegetation Index ◽

Normalized Difference Vegetation Index ◽

Cross Validation ◽

Google Earth ◽

Vegetation Types ◽

Environmental Predictors ◽

Distribution Models ◽

Data Set ◽

Sentinel 2

Machine learning (ML) provides a powerful set of tools that can improve the accuracy of distribution models by automatically representing underlying ecological relationships empirically captured from large sets of data. However, more recent methodological advances in ML that have been less frequently applied, e.g. in the field of deep learning, may yield additional potential for Distribution Modeling. In this project, we use two ML algorithms, Random Forest (RF) and multi-layer feed-forward artificial neural networks (ANN), to predict the occurrences of Vegetation Types (VT) across Norway. Accurate predictions may support environmental management or the validation of earth system models. The VT data (derived from the AR18x18 data set; n=31 classes) covers the entire spatial scope in 0.9 ha plots on a systematically sampled 18 km grid (n = 1,081 plots, n = 22,154 observations). It was obtained through a field-based survey by a group of trained experts between 2004 and 2014. We use the cloud-based platform "Google Earth Engine" to generate a set of remotely sensed predictor variables based on SENTINEL-2 satellite imagery (i.e. surface reflectance from 12 spectral bands and six vegetation indices). These are then combined with ancillary environmental rasters used previously to model Norwegian VT distribution, e.g. representing climate, land cover, or geological properties (n=55, before one-hot encoding). Preliminary results suggest that in both modeling approaches, the generated SENTINEL-2 variables, particularly the Normalized Difference Vegetation Index (NDVI), have the highest predictive power as measured by permutation importance. The mean overall accuracy using 5-fold cross-validation shows only minor differences between the two methods (approx. 0.45 for ANN vs. 0.44 for RF; the respective F1-scores are 0.35 for ANN and 0.34 for RF; most frequent class baseline accuracy = 0.136). The modeling challenges we currently face include a class imbalance in the VT data set, reconciling the different spatial resolutions of the environmental predictors, and discrepancies in the timing of data acquisition. The next steps in the project will be to incorporate spatial cross-validation into the workflow and to analyze the differences between the ML methods in detail (e.g. regarding the ability to model rare VTs or differences in variable importance). Moreover, we will evaluate the possibility to include additional satellite data sources. This work is a contribution to the Strategic Research Initiative &#8216;Land Atmosphere Interaction in Cold Environments&#8217; (LATICE) of the University of Oslo.

Download Full-text

Object detection for automotive radar point clouds – a comparison

AI Perspectives ◽

10.1186/s42467-021-00012-z ◽

2021 ◽

Vol 3 (1) ◽

Author(s):

Nicolas Scheiner ◽

Florian Kraus ◽

Nils Appenrodt ◽

Jürgen Dickmann ◽

Bernhard Sick

Keyword(s):

Machine Learning ◽

Object Detection ◽

Point Clouds ◽

Machine Learning Algorithms ◽

Future Research ◽

Automotive Radar ◽

Data Set ◽

Radar Sensors ◽

Adverse Weather ◽

Algorithm Comparison

AbstractAutomotive radar perception is an integral part of automated driving systems. Radar sensors benefit from their excellent robustness against adverse weather conditions such as snow, fog, or heavy rain. Despite the fact that machine-learning-based object detection is traditionally a camera-based domain, vast progress has been made for lidar sensors, and radar is also catching up. Recently, several new techniques for using machine learning algorithms towards the correct detection and classification of moving road users in automotive radar data have been introduced. However, most of them have not been compared to other methods or require next generation radar sensors which are far more advanced than current conventional automotive sensors. This article makes a thorough comparison of existing and novel radar object detection algorithms with some of the most successful candidates from the image and lidar domain. All experiments are conducted using a conventional automotive radar system. In addition to introducing all architectures, special attention is paid to the necessary point cloud preprocessing for all methods. By assessing all methods on a large and open real world data set, this evaluation provides the first representative algorithm comparison in this domain and outlines future research directions.

Download Full-text