Vision Foundation Models and Rule-Based Approaches for Roof Surface Segmentation and Photovoltaic Potential Analysis in Urban Areas
Abstract and Figures
This study presents two methods for rapidly and effectively determining the photovoltaic (PV) potential of building roofs in urban areas using aerial photographs and point cloud data. In the first method, the Segment Anything Model (SAM) and Contrastive Language Image Pre-Training (CLIP) models are used to detect roof surfaces and obstacles from aerial photographs. In the second method, the Random Sample Consensus (RANSAC) and Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithms are employed to identify roof surfaces from Light Detection and Ranging (LiDAR) point clouds. Through the first proposed method, the performance of current deep learning approaches in 2.5D PV potential analysis is investigated, while the second approach examines the performance of 3D PV potential analysis compared to the 2D approach. In PV potential analysis, the Photovoltaic Geographical Information System (PVGIS) Application Programming Interface (API) was utilized. The analysis is conducted based on roof parameters obtained through both proposed methods. In building detection, the first approach achieved an Intersection over Union (IoU) score of 94.29%, whereas the second approach attained an IoU score of 91.23%.
Figures - available via license: Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International
Content may be subject to copyright.
ResearchGate has not been able to resolve any citations for this publication.
As valued members of forest ecosystems, fast-growing tree species are of great importance for protecting natural forests, fulfilling the demand for raw wood material, and preserving biodiversity and ecosystem services. In this regard, multi-temporal monitoring and evaluation of poplar planted areas are essential for decision makers and planners to keep the inventory maps of planted areas up-to-date and track the temporal changes accurately. Using a spectral library, available spectral knowledge can be incorporated into inventory mapping and image classification processes. In this study, in situ spectral measurements of five Populus (four clones and one species) planted sites in Türkiye were collected at three phenological stages (early, mid, and late vegetation) and a novel spectral library was created after completing strict processing stages. Analysis of the spectral reflectances of the Populus species revealed that their spectral properties in the visible region are similar, whereas the discrimination in the near-infrared and shortwave infrared (SWIR) regions was noticeably higher. Two types of applications were conducted to effectively use the spectral library. First, the spectral library was used to validate the poplar class labels of the sites visited during the field campaign. The effectiveness of Worldview-3 imagery with eight SWIR bands was evaluated by pairwise comparison of the spectral curves of the sample fields with those of the spectral library. Second, the spectral library facilitated the extraction of poplar samples from Sentinel-2 imagery in an unvisited region, and a supervised classification process was conducted using the random forest algorithm. The results revealed that the Populus species (Samsun and I-214) could be delineated with high accuracy (>70%) using 10-band Sentinel-2 imagery, and object-based classification outperformed pixel-based classification by approximately 5% for the poplar types. The developed spectral library stands as a valuable asset for studies focused on identifying and classifying Populus species using high-resolution imagery.
In this work, a flower-shaped shading system with integrated tracking photovoltaic, suitable for sustainable extensive urban coverages, is designed. Detailed photovoltaic energy yield simulations with a single-diode model approach are performed to disclose the potential of the proposed tracking photovoltaic shading device (PVSD). Simulations are performed with reference to a case study. A double-layer space truss is used to house the innovative modular photovoltaic tracking system, and the first application is envisaged for the coverage of a public market area of a sunny municipality in Southern Italy. By comparing it with the traditional photovoltaic fixed system, the results of the simulations show a steadier energy generation of the new PVSD, and it also provides better coverage with renewable energy during the hours of the day when the traditional system produces low electric energy. Lastly, an early interactive prototype of the PVSD system is presented. The tracking mechanism is carefully designed, 3D-printed at a small scale and tested with a motorized dynamic system controlled by a microcontroller board. The realization of the physical prototype and the engineering of the movement mechanism confirmed the feasibility and the correct functioning of the conceived system opening to real-scale applications.
Numerous deep learning techniques have been explored in pursuit of achieving precise road segmentation; nonetheless, this task continues to present a significant challenge. Exposing shadows and the obstruction of objects are the most important difficulties associated with road segmentation using optical image data alone. By incorporating additional data sources, such as LiDAR data, the accuracy of road segmentation can be improved in areas where optical images are insufficient to segment roads properly. The missing information in spectral data due to the object blockage and shadow effect can be compensated by the integration of 2D and 3D information. This study proposes a feature-wise fusion strategy of optical images and point clouds to enhance the road segmentation performance of a deep learning model. For this purpose, high-resolution satellite images and airborne LiDAR point cloud collected over Florida, USA, were used. Eigenvalue-based and geometric 3D property-based features were calculated based on the LiDAR data. These optical images and LiDAR-based features were used together to train, end-to-end, a deep residual U-Net architecture. In this strategy, the high-level features generated from optical images were concatenated with the LiDAR-based features before the final convolution layer. The consistency of the proposed strategy was evaluated using ResNet backbones with a different number of layers. According to the obtained results, the proposed fusion strategy improved the prediction capacity of the U-Net models with different ResNet backbones. Regardless of the backbone, all models showed enhancement in prediction statistics by 1% to 5%. The combination of optical images and LiDAR point cloud in the deep learning model has increased the prediction performance and provided the integrity of road geometry in woodland and shadowed areas.
The main purpose of this study is to monitor the impacts of seasonal urban heat island based on the relationship between seasonal land surface temperature and Land Use/ Cover (LU/LC) fluctuations in Kabul city, the capital of Afghanistan in 2020.To do this, areas affected by Urban Heat Island (UHI) were detected. The relationship between UHI spatiotemporal fluctuations and LULC classes was also evaluated. The composition of Synthetic Aperture Radar (SAR) and optical Landsat 8 OLI images were used to extract the LU/LC type of the study area with higher accuracy for all seasons. Random forest supervised classifier was used to classify Kabul city into four different LU/LC classes (vegetation, water body, built-up, and bare land). Landsat 8 thermal bands were used to derive land surface temperature (LST) change patterns. The outputs revealed that the UHIs created in Kabul city have distinct causes depending on the type of LU/LC classes. The mean land surface temperature of the four LU/LC classes in different seasons demonstrated that the distribution mode of UHIs in the city was in line with the type of LU/LC. Based on overall assessments, the results revealed that nine districts in Kabul city (D-19, D-15, D-21, D-16, D-20, D-17, D-6, D-9, and D-1) are highly affected by urban heat islands. The UHIaffected regions in Kabul city, mostly consist of industrial parks, factories, the airport, industrial complexes, brick kilns, high volume of daily traffic congestion, densely populated, crush plant stations, and residential settlements. To validate the results, daytime and nighttime MODIS LST maps were generated. The daytime LST map similarly demonstrated that the spatial distribution of UHIs is related to the types of LU/LC. On the other hand, the results from nighttime LST maps revealed that nighttime heat islands are more intense than daytime heat islands, with the main concentrations of UHIs spreading towards the most populated areas in the central and eastern portion of Kabul city. This is due to the absorption of solar radiation throughout the day and emission during the nighttime. Remote sensing techniques were shown to be productive and economic, particularly in terms of minimizing the time for the evaluation of urban heat island variations based on the relationship between land surface temperature and LU/LC changes. Precise and updated remote sensing assessments of UHIs will pave the way for urban planners and land management organizations in addressing challenges related to the phenomenon.
A comprehensive understanding of failure modes of solar photovoltaic (PV) modules is key to extending their operational lifetime in the field. In this review, first, specific failure modes associated with mature PV technologies, such as crystalline silicon (c‐Si), copper indium gallium selenide (CIGS) and cadmium telluride (CdTe), are framed by sources of specific failure modes, their development from the early‐developmental stages onwards and their impact upon long term performance of PV modules. These failure modes are sorted by both PV technology and location of occurrence in PV modules, such as substrate, encapsulant, front and rear electrode, absorber and interlayers. The second part of the review is focused on emerging PV technologies, such as perovskites solar cells, dye sensitised and organic PVs, where due to their low to medium technology readiness levels, specific long‐term degradation mechanisms have not fully emerged, and most mechanisms are only partially understood. However, an in‐depth summary of the known stability challenges associated with each emerging PV technology is presented. Finally, in this paper, lessons learned from mature PV technologies are reviewed, and considerations are given in to how these might be applied to the further development of emerging technologies. Namely, any emerging PV technology must eventually pass industry‐standard qualification tests, while warranties for the lifetime of modern c‐Si‐based modules might be extended beyond the existing warranted life of 25 years.
Deep learning architectures are widely used for road segmentation studies. Data sets reflect the characteristics of the study region in which they are generated. The models that are trained with the data sets are unable to accurately forecast the region outside the area covered by the data sets. That is why local data sets must be generated to meet the requirements of navigation, military activities, urban planning, and health management applications where road data are indispensable. In this study, satellite images and corresponding mask images of Istanbul city at different zoom levels were generated via Google Map Platform. Trained models were produced using Deep Residual U-Net. This study consists of three parts. The first part includes training each zoom level, from 14 to 17, separately. The recall statistics of models were found as 95%, 94%, 92%, 86%, and corresponding F1 - Score statistics were calculated as 95%, 95%, 92%, 86%, respectively. Next, consecutive training with priory weight and L0 regularization on satellite images were tested to increase the prediction results. The consecutive model and L0 regularization methods resulted in an increase in both recall and F1-Score by 2% - 3%. In the third part, a new model was generated by combining all images which were produced at different levels. The combined model was found successful for the prediction of roads in all zoom levels. Finally, the performance of these models trained with the Istanbul data set was compared with well-known DeepGlobe and Massachusetts road data sets. Hence, this study illustrated that generating all data sets specifically for the area of interest is crucial.
Decentralized solar PhotoVoltaic (PV) is one of the most promising energy sources for cities and individuals pursuing energy self-sufficiency. Especially, the already available rooftop surfaces are a major contributor to push for rooftop mounted PV systems. However, accurate PV potential estimation of individual buildings is still a challenging task since many parameters must be considered such as meteorological factors, panel technology, geographical location, available roof surface area, surface azimuth and tilt angle. In this study, we created an efficient approach that can be used for roof surface's PV potential estimation based on point cloud data and capable of processing various scales from single building to city scale. In the proposed approach, each roof surface's features were utilized for PV potential estimation by employing the PVGIS database. PV potential estimation was carried out on daily, monthly, and annual periods to provide a better estimation. Also, we developed a flexible and easy to use open-source plugin based on the QGIS software for rooftop mounted PV potential estimation capable of estimating every roof surface's PV potential. The method was tested on 80 buildings selected from ROOFN3D dataset. The proposed approach achieved an overall accuracy of 84% and an F1 score of 0.92.
Solar photovoltaic (PV) system, as one kind of the most promising renewable energy technologies, plays a key role in reducing carbon emissions to achieve the targets of global net zero carbon. In the past few decades, PV installations have seen a rapid growth. Predicting the installed amount and the capacity of solar PV systems is therefore useful for formulating effective carbon reduction policies in the related area. In the present study, the methods of identifying PV installation based on satellite and aerial images have been reviewed. Suggestions have been put forward to optimize the identification process and to predict the potential of rooftop PV installation. The results show that the specific purposes of PV identification can be categorized as image classification, object detection and semantic segmentation. The available identification methods encompass pixel-based analysis method (PBIA), object-based analysis method (OBIA) and deep learning. Deep learning has a high accuracy in segmentation for all sizes of PV systems, with precision and recall of rooftop PV segmentation in the range of 41-98.9% and 54.5-95.8%, respectively. OBIA has the best accuracy in detecting centralized PV systems with relatively low-resolution multispectral images. Furthermore, a grading segmentation strategy for PV segmenta-tion in the large region is presented, combining the three identification methods and the images with different resolutions. In addition, the potential of rooftop PV installation can be predicted by segmenting the available roof area in the images. After considering the shading effects, upper structure and other uses, the roof availability coefficient tends to be in the range of 0.25-0.46. It is also suggested to combine PV and roof segmentation to estimate the installation potential more accurately, in the context of rapid growth of the rooftop PV.
The classification and mapping accuracy of urban land cover and land use has always been a critical topic and several auxiliary data have been used to improve the classification accuracy. However, to the best of our knowledge, there is limited knowledge of the addition of airborne Light Detection and Ranging (lidar)-Digital Elevation Model (DEM) and Topographic Position Index (TPI) for urban land cover and land use classification and mapping. The aim of this study was to explore the addition of airborne lidar-DEM and derived TPI to reflect data of Landsat Operational Land Imager (OLI) in improving the classification accuracy of urban land cover and land use map- ping. Specifically, this study explored the mapping accuracies of urban land cover and land use classifications derived using: 1) standalone Landsat OLI satellite data; 2) Landsat OLI with acquired airborne lidar-DEM ; 3) Landsat OLI with TPI ; and 4) Landsat OLI with airborne lidar-DEM and derived TPI. The results showed that the addition of airborne lidar-DEM and TPI yielded the best overall urban land cover and land use classification accuracy of about 88%. The findings in this study demonstrated that both lidar-DEM and TPI had a positive impact in improving urban land cover and land use classification.