The objective of this paper is to detect the type of vegetation so that a more accurate Digital Terrain Model (DTM) can be generated by excluding the vegetation from the Digital Surface Model (DSM) based on the vegetation type (such as trees). This way, many different inpainting methods can be applied subsequently to restore the terrain information from the removed vegetation pixels from DSM and obtain a more accurate DTM. We trained three DeepLabV3+ models with three different datasets that are collected at different resolutions. Among the three DeepLabV3+ models, the model trained with the dataset that has an image resolution close to the test data images provided the best performance and the semantic segmentation results with this model looked highly promising.
To accurately extract digital terrain model (DTM), it is necessary to remove heights due to vegetation such as trees and shrubs and other manmade structures such as buildings, bridges, etc. from the digital surface model (DSM). The resulting DTM can then be used for construction planning, land surveying, etc. Normally, the process of extracting DTM involves two steps. First, accurate land cover classification is required. Second, an image inpainting process is needed to fill in the missing pixels due to trees, buildings, bridges, etc. In this paper, we focus on the second step of using image inpainting algorithms for terrain reconstruction. In particular, we evaluate seven conventional and deep learning based inpainting algorithms in the literature using two datasets. Both objective and subjective comparisons were carried out. It was observed that some algorithms yielded slightly better performance than others.
Current advancements on photogrammetric software along with affordability and wide spreading of Autonomous Unmanned Aerial Vehicles (AUAV), allow for rapid, timely and accurate 3D modelling and mapping of small to medium sized areas. Although the importance of flight patterns and large overlaps in aerial triangulation and Digital Surface Model (DSM) production from large format aerial cameras is well documented in literature, this is not the case for AUAV photography. This paper assess DSM accuracy of models created using different flight patterns and compares them against check points and Lidar data. Three UAV flights took place, with 70%-65% forward and side overlaps, with West-East (W-E), North-South (N-S) and Northwest-Southeast (NW-SE) directions. Blocks with different flight patterns were created and processed to create raster DSM with 0.25m ground pixel size using Multi View Stereo (MVS). Using Lidar data as reference, difference maps and statistics were calculated for each block, in order to evaluate their overall accuracy. The combined scenario performed slightly better that the rest. Because of their lower spatial resolution, Lidar data prove to be an inadequate reference data set, although according to their internal vertical precision they are superior to UAV DSM. Point cloud noise from MVS, is considerable in contrast to Lidar data. A Lidar data set from a lower flying platform such as helicopter might have been a better match to low flying UAV data.
Semi-distributed physically-based models are well established and widely used for hydrological modeling due to their ability to capture the spatial variability of the watershed among land use, soil types and topographic characteristics; and to characterize distributed inputs in different areas within the watershed. They offer a more realistic watershed representation, allowing for better predictions of the behavior of a hydrologic system, based on novel climatic inputs. Watershed subdivision and the question of an optimum discretization level is an important issue in distributed hydrological modeling as it affects the setup of hydrologic models and has the potential to affect model output. Soil and Water Assessment Tool (SWAT), a semi-distributed physically-based hydrologic model, divides the watershed into smaller subwatersheds which are further subdivided into HRUs consisting of homogeneous land use, soil, slope and management characteristics. The number and size of HRUs is calculated based on user-specified land use, soil and slope thresholds. This study investigates the impact of the slope threshold in the HRU definition on flow predictions and hydrologic mass balance, applied on three subwatersheds of the Evrotas River Basin (1348km2), a mountainous catchment in Peloponnesus, Greece. The catchment is delineated using a 90m DEM and then divided into 150 subwatersheds. The model was calibrated, and simulations were performed on three subwatersheds using a range of 5%- 30% slope thresholds for the HRU definition while land use and soil thresholds remained the same. Results showed that the coarser delineation (13 HRUs) produced a very accurate hydrologic mass balance and satisfactory flow predictions (RSR, PBIAS, NSE) while, finer delineations (21 HRUs) produces inaccurate hydrologic mass balance (54.49% lower surface runoff) but more accurate flow predictions (RSR, PBIAS, NSE).
In recent years, Autonomous Unmanned Aerial Vehicles (AUAV) became popular among researchers across disciplines because they combine many advantages. One major application is monitoring and mapping. Their ability to fly beyond eye sight autonomously, collecting data over large areas whenever, wherever, makes them excellent platform for monitoring hazardous areas or disasters. In both cases rapid mapping is needed while human access isn’t always a given. Indeed, current automatic processing of aerial photos using photogrammetry and computer vision algorithms allows for rapid orthophomap production and Digital Surface Model (DSM) generation, as tools for monitoring and damage assessment. In such cases, control point measurement using GPS is either impossible, or time consuming or costly. This work investigates accuracies that can be attained using few or none control points over areas of one square kilometer, in two test sites; a typical block and a corridor survey. On board GPS data logged during AUAV’s flight are being used for direct georeferencing, while ground check points are being used for evaluation. In addition various control point layouts are being tested using bundle adjustment for accuracy evaluation. Results indicate that it is possible to use on board single frequency GPS for direct georeferencing in cases of disaster management or areas without easy access, or even over featureless areas. Due to large numbers of tie points in the bundle adjustment, horizontal accuracy can be fulfilled with a rather small number of control points, but vertical accuracy may not.
Land Surface Temperature (LST) is an extremely important parameter that controls the exchange of long wave radiation between surface and atmosphere. It is a good indicator of the energy balance at the Earth’s surface and it is one of the key parameters in the physics of land-surface processes on regional as well as global scale. This paper utilizes monthly night and day averaged LST MODIS imagery over Cyprus for a 9 year period. Fourier analysis and Least squares estimation fitting are implemented to analyze mean daily data over Cyprus in an attempt to investigate possible temperature tenancy over these years and possible differences among areas with different land cover and land use, such as Troodos Mountain and Nicosia, the main city in the center of the island. The analysis of data over a long time period, allows questions such as whether there is a tenancy to temperature increase, to be answered in a statistically better way, provided that ‘noise’ is removed correctly. Dealing with a lot of data, always provides a more accurate estimation, but on the other hand, more noise in implemented on the data, especially when dealing with temperature which is subject to daily and annual cycles. A brief description over semi-automated data acquisition and standardization using object-oriented programming and GIS-based techniques, will be presented. The paper fully describes the time series analysis implemented, the Fourier method and how it was used to analyze and filter mean daily data with high frequency. Comparison of mean monthly daily LST against day and night LSTs is also performed over the 9 year period in order to investigate whether use of the extended data series provide significant advantage over short.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.