Using high‐resolution aerial imagery and deep learning to detect tree spatio-temporal dynamics at the treeline

2021 ◽  
Author(s):  
Mirela Beloiu ◽  
Dimitris Poursanidis ◽  
Samuel Hoffmann ◽  
Nektarios Chrysoulakis ◽  
Carl Beierkuhnlein
Keyword(s):  
Deep Learning ◽  
High Resolution ◽  
Object Detection ◽  
Temporal Dynamics ◽  
Tree Density ◽  
Aerial Imagery ◽  
Aerial Images ◽  
Remote Areas ◽  
Spatio Temporal ◽  
Study Sites

<p>Recent advances in deep learning techniques for object detection and the availability of high-resolution images facilitate the analysis of both temporal and spatial vegetation patterns in remote areas. High-resolution satellite imagery has been used successfully to detect trees in small areas with homogeneous rather than heterogeneous forests, in which single tree species have a strong contrast compared to their neighbors and landscape. However, no research to date has detected trees at the treeline in the remote and complex heterogeneous landscape of Greece using deep learning methods. We integrated high-resolution aerial images, climate data, and topographical characteristics to study the treeline dynamic over 70 years in the Samaria National Park on the Mediterranean island of Crete, Greece. We combined mapping techniques with deep learning approaches to detect and analyze spatio-temporal dynamics in treeline position and tree density. We use visual image interpretation to detect single trees on high-resolution aerial imagery from 1945, 2008, and 2015. Using the RGB aerial images from 2008 and 2015 we test a Convolution Neural Networks (CNN)-object detection approach (SSD) and a CNN-based segmentation technique (U-Net). Based on the mapping and deep learning approach, we have not detected a shift in treeline elevation over the last 70 years, despite warming, although tree density has increased. However, we show that CNN approach accurately detects and maps tree position and density at the treeline. We also reveal that the treeline elevation on Crete varies with topography. Treeline elevation decreases from the southern to the northern study sites. We explain these differences between study sites by the long-term interaction between topographical characteristics and meteorological factors. The study highlights the feasibility of using deep learning and high-resolution imagery as a promising technique for monitoring forests in remote areas.</p>

scholarly journals Deep Learning Based Wildfire Event Object Detection from 4K Aerial Images Acquired by UAS

AI ◽  
2020 ◽  
Vol 1 (2) ◽  
pp. 166-179 ◽  
Author(s):  
Ziyang Tang ◽  
Xiang Liu ◽  
Hanlin Chen ◽  
Joseph Hupy ◽  
Baijian Yang
Keyword(s):  
Deep Learning ◽  
High Resolution ◽  
Object Detection ◽  
Aerial Images ◽  
Unmanned Aerial Systems ◽  
Computer Assisted ◽  
Visual Interpretation ◽  
Two Phase ◽  
Aerial Systems ◽  
Spot Fires

Unmanned Aerial Systems, hereafter referred to as UAS, are of great use in hazard events such as wildfire due to their ability to provide high-resolution video imagery over areas deemed too dangerous for manned aircraft and ground crews. This aerial perspective allows for identification of ground-based hazards such as spot fires and fire lines, and to communicate this information with fire fighting crews. Current technology relies on visual interpretation of UAS imagery, with little to no computer-assisted automatic detection. With the help of big labeled data and the significant increase of computing power, deep learning has seen great successes on object detection with fixed patterns, such as people and vehicles. However, little has been done for objects, such as spot fires, with amorphous and irregular shapes. Additional challenges arise when data are collected via UAS as high-resolution aerial images or videos; an ample solution must provide reasonable accuracy with low delays. In this paper, we examined 4K ( 3840 × 2160 ) videos collected by UAS from a controlled burn and created a set of labeled video sets to be shared for public use. We introduce a coarse-to-fine framework to auto-detect wildfires that are sparse, small, and irregularly-shaped. The coarse detector adaptively selects the sub-regions that are likely to contain the objects of interest while the fine detector passes only the details of the sub-regions, rather than the entire 4K region, for further scrutiny. The proposed two-phase learning therefore greatly reduced time overhead and is capable of maintaining high accuracy. Compared against the real-time one-stage object backbone of YoloV3, the proposed methods improved the mean average precision(mAP) from 0 . 29 to 0 . 67 , with an average inference speed of 7.44 frames per second. Limitations and future work are discussed with regard to the design and the experiment results.

scholarly journals Detecting methane ebullition on thermokarst lake ice using high resolution optical aerial imagery

2015 ◽  
Vol 12 (10) ◽  
pp. 7449-7490
Author(s):  
P. R. Lindgren ◽  
G. Grosse ◽  
K. M. Walter Anthony ◽  
F. J. Meyer
Keyword(s):  
Image Analysis ◽  
High Resolution ◽  
Aerial Imagery ◽  
Aerial Images ◽  
Accurate Estimation ◽  
Lake Ice ◽  
Thermokarst Lakes ◽  
Thermokarst Lake ◽  
Spatio Temporal ◽  
Methane Ebullition

Abstract. Thermokarst lakes are important emitters of methane, a potent greenhouse gas. However, accurate estimation of methane flux from thermokarst lakes is difficult due to their remoteness and observational challenges associated with the heterogeneous nature of ebullition (bubbling). We used multi-temporal high-resolution (9–11 cm) aerial images of an interior Alaskan thermokarst lake, Goldstream Lake, acquired 2 and 4 days following freeze-up in 2011 and 2012, respectively, to characterize methane ebullition seeps and to estimate whole-lake ebullition. Bubbles impeded by the lake ice sheet form distinct white patches as a function of bubbling rate vs. time as ice thickens. Our aerial imagery thus captured in a single snapshot the ebullition events that occurred before the image acquisition. Image analysis showed that low-flux A- and B-type seeps are associated with low brightness patches and are statistically distinct from high-flux C-type and Hotspot seeps associated with high brightness patches. Mean whole-lake ebullition based on optical image analysis in combination with bubble-trap flux measurements was estimated to be 174 ± 28 and 216 ± 33 mL gas m−2 d−1 for the years 2011 and 2012, respectively. A large number of seeps demonstrated spatio-temporal stability over our two-year study period. A strong inverse exponential relationship (R2 ≥ 0.79) was found between percent surface area of lake ice covered with bubble patches and distance from the active thermokarst lake margin. Our study shows that optical remote sensing is a powerful tool to map ebullition seeps on lake ice, to identify their relative strength of ebullition and to assess their spatio-temporal variability.

scholarly journals IMG2nDSM: Height Estimation from Single Airborne RGB Images with Deep Learning

2021 ◽  
Vol 13 (12) ◽  
pp. 2417
Author(s):  
Savvas Karatsiolis ◽  
Andreas Kamilaris ◽  
Ian Cole
Keyword(s):  
Deep Learning ◽  
State Of The Art ◽  
Aerial Imagery ◽  
Aerial Images ◽  
Surface Model ◽  
Large Area ◽  
Digital Terrain ◽  
Terrain Models ◽  
Architectural Features ◽  
Rgb Images

Estimating the height of buildings and vegetation in single aerial images is a challenging problem. A task-focused Deep Learning (DL) model that combines architectural features from successful DL models (U-NET and Residual Networks) and learns the mapping from a single aerial imagery to a normalized Digital Surface Model (nDSM) was proposed. The model was trained on aerial images whose corresponding DSM and Digital Terrain Models (DTM) were available and was then used to infer the nDSM of images with no elevation information. The model was evaluated with a dataset covering a large area of Manchester, UK, as well as the 2018 IEEE GRSS Data Fusion Contest LiDAR dataset. The results suggest that the proposed DL architecture is suitable for the task and surpasses other state-of-the-art DL approaches by a large margin.

scholarly journals Deep Learning-Based Object Detection for Unmanned Aerial Systems (UASs)-Based Inspections of Construction Stormwater Practices

Sensors ◽  
2021 ◽  
Vol 21 (8) ◽  
pp. 2834
Author(s):  
Billur Kazaz ◽  
Subhadipto Poddar ◽  
Saeed Arabi ◽  
Michael A. Perez ◽  
Anuj Sharma ◽  
...  
Keyword(s):  
Deep Learning ◽  
Object Detection ◽  
Pollution Prevention ◽  
Aerial Images ◽  
Unmanned Aerial Systems ◽  
Construction Sites ◽  
State Regulations ◽  
Novel Approach ◽  
Model Training ◽  
Aerial Systems

Construction activities typically create large amounts of ground disturbance, which can lead to increased rates of soil erosion. Construction stormwater practices are used on active jobsites to protect downstream waterbodies from offsite sediment transport. Federal and state regulations require routine pollution prevention inspections to ensure that temporary stormwater practices are in place and performing as intended. This study addresses the existing challenges and limitations in the construction stormwater inspections and presents a unique approach for performing unmanned aerial system (UAS)-based inspections. Deep learning-based object detection principles were applied to identify and locate practices installed on active construction sites. The system integrates a post-processing stage by clustering results. The developed framework consists of data preparation with aerial inspections, model training, validation of the model, and testing for accuracy. The developed model was created from 800 aerial images and was used to detect four different types of construction stormwater practices at 100% accuracy on the Mean Average Precision (MAP) with minimal false positive detections. Results indicate that object detection could be implemented on UAS-acquired imagery as a novel approach to construction stormwater inspections and provide accurate results for site plan comparisons by rapidly detecting the quantity and location of field-installed stormwater practices.

scholarly journals Dynamics of Erosion and Deposition in a Partially Restored Valley-Bottom Gully

Land ◽  
2021 ◽  
Vol 10 (1) ◽  
pp. 62
Author(s):  
Alberto Alfonso-Torreño ◽  
Álvaro Gómez-Gutiérrez ◽  
Susanne Schnabel
Keyword(s):  
High Resolution ◽  
Sediment Deposition ◽  
Aerial Images ◽  
Valley Bottom ◽  
Erosion And Deposition ◽  
Channel Dynamics ◽  
Check Dams ◽  
Before And After ◽  
Multi Temporal ◽  
Spatio Temporal

Gullies are sources and reservoirs of sediments and perform as efficient transfers of runoff and sediments. In recent years, several techniques and technologies emerged to facilitate monitoring of gully dynamics at unprecedented spatial and temporal resolutions. Here we present a detailed study of a valley-bottom gully in a Mediterranean rangeland with a savannah-like vegetation cover that was partially restored in 2017. Restoration activities included check dams (gabion weirs and fascines) and livestock exclosure by fencing. The specific objectives of this work were: (1) to analyze the effectiveness of the restoration activities, (2) to study erosion and deposition dynamics before and after the restoration activities using high-resolution digital elevation models (DEMs), (3) to examine the role of micro-morphology on the observed topographic changes, and (4) to compare the current and recent channel dynamics with previous studies conducted in the same study area through different methods and spatio-temporal scales, quantifying medium-term changes. Topographic changes were estimated using multi-temporal, high-resolution DEMs produced using structure-from-motion (SfM) photogrammetry and aerial images acquired by a fixed-wing unmanned aerial vehicle (UAV). The performance of the restoration activities was satisfactory to control gully erosion. Check dams were effective favoring sediment deposition and reducing lateral bank erosion. Livestock exclosure promoted the stabilization of bank headcuts. The implemented restoration measures increased notably sediment deposition.

Analysis Of High Resolution Aerial Images For Object Detection

1989 ◽  
Author(s):  
Mohan M. Trivedi ◽  
Amol G. Bokil ◽  
Mourad B. Takla ◽  
George B. Maksymonko ◽  
J. Thomas Broach
Keyword(s):  
High Resolution ◽  
Object Detection ◽  
Aerial Images

scholarly journals Potato crop stress identification in aerial images using deep learning‐based object detection

2021 ◽  
Author(s):  
Sujata Butte ◽  
Aleksandar Vakanski ◽  
Kasia Duellman ◽  
Haotian Wang ◽  
Amin Mirkouei
Keyword(s):  
Deep Learning ◽  
Object Detection ◽  
Potato Crop ◽  
Aerial Images

Spatio-temporal dynamics of erosion and deposition in a partially restored valley-bottom gully

2021 ◽  
Author(s):  
Alberto Alfonso-Torreño ◽  
Álvaro Gómez-Gutiérrez ◽  
Susanne Schnabel
Keyword(s):  
Temporal Dynamics ◽  
Land Use Changes ◽  
Degradation Process ◽  
Gully Erosion ◽  
Aerial Images ◽  
Negative Consequences ◽  
Valley Bottom ◽  
Erosion And Deposition ◽  
Check Dams ◽  
Spatio Temporal

<p>Soil erosion by water is a frequent soil degradation process in rangelands of SW Spain. The two main erosive processes in these areas are sheetwash erosion in hillslopes and gully erosion due to concentrated flow in valley bottoms. Land use changes and overgrazing play a key role in the genesis and development of gullies and gully erosion is a frequent process with negative consequences at the valley bottoms of these landscapes.</p><p>The development of new techniques allows monitoring of gully dynamics with an increase at spatial and temporal resolutions. Here we present a detailed study of a valley-bottom gully in a Mediterranean rangeland with a savannah-like vegetation cover that was partially restored in February 2017. Restoration activities included check dams (gabion weirs and fascines) and livestock exclosure by fencing. The objectives of this study were: (1) to analyze the effectiveness of the restoration measures, (2) to study erosion and deposition dynamics before and after the restoration activities, (3) to examine the role of micro-morphology on the observed topographic changes and (4) to compare the current and recent channel dynamics with previous studies conducted in the same study area through different methods and spatio-temporal scales, quantifying medium-term changes. Topographic changes were estimated using multi-temporal high-resolution DEMs produced using Structure-from-Motion (SfM) photogrammetry and aerial images acquired by a fixed-wing Unmanned Aerial Vehicle (UAV). DEMs and orthophotographs with a Ground Sampling Distance of 0.02 m were produced by means of SfM photogrammetric techniques. The average Root Mean Square Error (RMSE) estimated during the SfM processing was 0.03 m.</p><p>The performance of the restoration activities was satisfactory to control gully erosion. Check dams were effective favoring sediment deposition and reducing lateral bank erosion. Nevertheless, erosion was observed immediately downstream in 9% of the check dams. Livestock exclosure in the most degraded area promoted the stabilization of bank headcuts and revegetation. The sediments retained behind check dams reduced the longitudinal slope gradient of the channel bed and established a positive feedback mechanism for channel revegetation.</p><p><strong>Keywords</strong>: gully erosion, restoration, topographic change, UAV+SfM, rangeland.</p>

scholarly journals Object Detection in Very High-Resolution Aerial Images Using One-Stage Densely Connected Feature Pyramid Network

Sensors ◽  
2018 ◽  
Vol 18 (10) ◽  
pp. 3341 ◽  
Author(s):  
Hilal Tayara ◽  
Kil Chong
Keyword(s):  
High Resolution ◽  
Object Detection ◽  
Computation Time ◽  
Aerial Images ◽  
Feature Maps ◽  
Two Stage ◽  
One Stage ◽  
Wide Range ◽  
Feature Pyramid ◽  
Very High

Object detection in very high-resolution (VHR) aerial images is an essential step for a wide range of applications such as military applications, urban planning, and environmental management. Still, it is a challenging task due to the different scales and appearances of the objects. On the other hand, object detection task in VHR aerial images has improved remarkably in recent years due to the achieved advances in convolution neural networks (CNN). Most of the proposed methods depend on a two-stage approach, namely: a region proposal stage and a classification stage such as Faster R-CNN. Even though two-stage approaches outperform the traditional methods, their optimization is not easy and they are not suitable for real-time applications. In this paper, a uniform one-stage model for object detection in VHR aerial images has been proposed. In order to tackle the challenge of different scales, a densely connected feature pyramid network has been proposed by which high-level multi-scale semantic feature maps with high-quality information are prepared for object detection. This work has been evaluated on two publicly available datasets and outperformed the current state-of-the-art results on both in terms of mean average precision (mAP) and computation time.

Sign in / Sign up

Export Citation Format

Share Document