Improved Mask R-CNN for Rural Building Roof Type Recognition from UAV High-Resolution Images: A Case Study in Hunan Province, China

Accurate roof information of buildings can be obtained from UAV high-resolution images. The large-scale accurate recognition of roof types (such as gabled, flat, hipped, complex and mono-pitched roofs) of rural buildings is crucial for rural planning and construction. At present, most UAV high-resolution optical images only have red, green and blue (RGB) band information, which aggravates the problems of inter-class similarity and intra-class variability of image features. Furthermore, the different roof types of rural buildings are complex, spatially scattered, and easily covered by vegetation, which in turn leads to the low accuracy of roof type identification by existing methods. In response to the above problems, this paper proposes a method for identifying roof types of complex rural buildings based on visible high-resolution remote sensing images from UAVs. First, the fusion of deep learning networks with different visual features is investigated to analyze the effect of the different feature combinations of the visible difference vegetation index (VDVI) and Sobel edge detection features and UAV visible images on model recognition of rural building roof types. Secondly, an improved Mask R-CNN model is proposed to learn more complex features of different types of images of building roofs by using the ResNet152 feature extraction network with migration learning. After we obtained roof type recognition results in two test areas, we evaluated the accuracy of the results using the confusion matrix and obtained the following conclusions: (1) the model with RGB images incorporating Sobel edge detection features has the highest accuracy and enables the model to recognize more and more accurately the roof types of different morphological rural buildings, and the model recognition accuracy (Kappa coefficient (KC)) compared to that of RGB images is on average improved by 0.115; (2) compared with the original Mask R-CNN, U-Net, DeeplabV3 and PSPNet deep learning models, the improved Mask R-CNN model has the highest accuracy in recognizing the roof types of rural buildings, with F1-score, KC and OA averaging 0.777, 0.821 and 0.905, respectively. The method can obtain clear and accurate profiles and types of rural building roofs, and can be extended for green roof suitability evaluation, rooftop solar potential assessment, and other building roof surveys, management and planning.

Helipad on the building roof design features

The following items are considered: requirements, constructive decisions for helipads; covering plate design of the helipad located on the building roof; helipad modeling in LIRA CAD SP; calculation of a multi-storey building with a helipad on the roof, check for strength and rigidity; characteristics of the helipad on the building roof.

Experimental Study of Interference Effects of a High-Rise Building on the Snow Load on a Low-Rise Building with a Flat Roof

To explore the interference effects of a high-rise building on the snow load on a low-rise building with a flat roof, a series of wind tunnel tests were carried out with fine silica sand as a substitute for snow particles. The effects of the height of the interfering building and the distance between buildings on the snow distribution of the target building under three different wind directions were studied. The snow depth on the target building roof and the mass of particles blown off from the target building were measured during the wind tunnel tests, and the results showed that the snow distribution of the target building roof tends to be uniform when the interfering building is located upstream of the target building due to the shelter effect. When the interfering building is on the side of the target building, the snow distribution of the target building tends to be more uneven, because the interfering building increases the friction velocity on the target building roof near the interfering building. However, when the interfering building is located downstream of the target building, there will be an amplification effect of snow accumulation, and the snow distribution on the target building roof is nearly the same as that of the isolated condition. Under each wind direction, the interference effect of the snow load increases with the increase of the building height and the decrease of the building spacing. Therefore, the influence of the surrounding buildings on the snow distribution of the building roof cannot be ignored and should be considered in the structure design.

BUILDING ROOF VECTORIZATION WITH PPGNET

A challenge in data-based 3D building reconstruction is to find the exact edges of roof facet polygons. Although these edges are visible in orthoimages, convolution-based edge detectors also find many other edges due to shadows and textures. In this feasibility study, we apply machine learning to solve this problem. Recently, neural networks have been introduced that not only detect edges in images, but also assemble the edges into a graph. When applied to roof reconstruction, the vertices of the dual graph represent the roof facets. In this study, we apply the Point-Pair Graph Network (PPGNet) to orthoimages of buildings and evaluate the quality of the detected edge graphs. The initial results are promising, and adjusting the training parameters further improved the results. However, in some cases, additional work, such as post-processing, is required to reliably find all vertices.

A Deep Multi-Modal Learning Method and a New RGB-Depth Data Set for Building Roof Extraction

This study focuses on tackling the challenge of building mapping in multi-modal remote sensing data by proposing a novel, deep superpixel-wise convolutional neural network called DeepQuantized-Net, plus a new red, green, blue (RGB)-depth data set named IND. DeepQuantized-Net incorporated two practical ideas in segmentation: first, improving the object pattern with the exploitation of superpixels instead of pixels, as the imaging unit in DeepQuantized-Net. Second, the reduction of computational cost. The generated data set includes 294 RGB-depth images (256 training images and 38 test images) from different locations in the state of Indiana in the U.S., with 1024 × 1024 pixels and a spatial resolution of 0.5 ftthat covers different cities. The experimental results using the IND data set demonstrates the mean F1 scores and the average Intersection over Union scores could increase by approximately 7.0% and 7.2% compared to other methods, respectively.

VI preferential pathways : rule or exception

Trichloroethylene (TCE) releases from leaks and spills next to a large government building occurred over several decades with the most recent event occurring 20 years ago. In response to a perceived conventional vapor intrusion (VI) issue a sub-slab depressurization system (SSDS) was installed 6 years ago. The SSDS is operating within design limits and has achieved building TCE vapor concentration reductions. However, subsequent periodic TCE vapor spikes based on daily HAPSITE™ measurements indicate additional source(s). Two rounds of smoke tests conducted in 2017 and 2018 involved introduction of smoke into a sanitary sewer and storm drain manholes located on effluent lines coming from the building until smoke was observed exiting system vents on the roof. Smoke testing revealed many leaks in both the storm sewer and sanitary sewer systems within the building. Sleuthing of the VI source term using a portable HAPSITE™ indicate elevated vapor TCE levels correspond with observed smoke emanation from utility lines. In some instances, smoke odors were perceived but no leak or suspect pipe was identified suggesting the odor originates from an unidentified pipe located behind or enclosed in a wall. Sleuthing activities also found building roof materials explain some of the elevated TCE levels on the 2nd floor. A relationship was found between TCE concentrations in the roof truss area, plenum space above 2nd floor offices, and breathing zone of 2nd floor offices. Installation of an external blower in the roof truss space has greatly reduced TCE levels in the plenum and office spaces. Preferential VI pathways and unexpected source terms may be overlooked mechanisms as compared to conventional VI.

A Building Roof Identification CNN Based on Interior-Edge-Adjacency Features Using Hyperspectral Imagery

Hyperspectral remote sensing can obtain both spatial and spectral information of ground objects. It is an important prerequisite for a hyperspectral remote sensing application to make good use of spectral and image features. Therefore, we improved the convolutional Neural Network (CNN) model by extracting interior-edge-adjacency features of building roof and proposed a new CNN model with a flexible structure: Building Roof Identification CNN (BRI-CNN). Our experimental results demonstrated that the BRI-CNN can not only extract interior-edge-adjacency features of building roof, but also change the weight of these different features during the training process, according to selected samples. Our approach was tested using the Indian Pines (IP) data set and our comparative study indicates that the BRI-CNN model achieves at least 0.2% higher overall accuracy than that of the capsule network model, and more than 2% than that of CNN models.

Exploring the Impacts and Temporal Variations of Different Building Roof Types on Surface Urban Heat Island

This study examined the impact of different types of building roofs on urban heat islands. This was carried out using building roof data from remotely sensed Landsat 8 Operational Land Imager (OLI) and Thermal Infrared Sensor (TIRS) imagery. The roofs captured included white surface, blue steel, dark metal, other dark material, and residential roofs; these roofs were compared alongside three natural land covers (i.e., forest trees, grassland, and water). We also collected ancillary data including building height, building density, and distance to the city center. The impacts of various building roofs on land surface temperature (LST) were examined by analyzing their correlation and temporal variations. First, we examined the LST characteristics of five building roof types and three natural land covers using boxplots and variance analysis with post hoc tests. Then, multivariate regression analysis was used to explore the impact of building roofs on LST. There were three key findings in the results. First, the mean LSTs for five different building roofs statistically differed from each other; these differences were more significant during the hot season than the cool season. Second, the impact of the five types of roofs on LSTs varied considerably from each other. Lastly, the contribution of the five roof types to LST variance was more substantial during the cool season. These findings unveil specific urban heat retention drivers, in which different types of building roofs are one such driver. The outcomes from this research may help policymakers develop more effective strategies to address the surface urban heat island phenomenon and its related health concerns.

Analisis Peningkatan Kualitas Permukiman Kumuh di Desa Loh Sumber Kabupaten Kutai Kartanegara

Population growth is a high primary need, one of which is the need for housing. The existence of slum settlements in the Kutai Kartanegara Regency has an indication of social and environmental problems. From the results of the location justification that the densest slum area is in Loh Sumber Village in RT. 02 and RT. 03 with a land area of ​​5.31 hectares. This study aims to provide input through the concept of improving the quality of slum settlements, with the research approach used is a rationalistic approach, based on truth. The research method is in the form of a qualitative descriptive study, with the answer to an ongoing problem. The research indicators and variables are (1) the physical condition of the building, with the variables of building density and building quality; and (2) The condition of facilities and infrastructure, with variables of road environmental quality, environmental drainage, quality of drinking water network, solid waste management, quality of wastewater and sanitation, and public street lighting. The results of the discussion are the concept of rejuvenation in the form of the use of building roof materials, and the use of wood types as building materials with strong resistance to air, repair of environmental roads with concrete and paving blocks, by changing the direction of the directions. as a street name marker, improve the quality and quantity of the system. clean water, manufacture of wastewater from IPAL / IPLT, determine the location of TPS, and public street lighting. The need for local government cooperation in implementing community needs and involving the active role of the community in maintaining a clean and safe residential environment.