Roof Segments Research Articles

From roofs to renewables: Deep learning and geographic information systems insights into a comprehensive urban solar photovoltaic assessment for Stonehaven

As urban solar photovoltaic (PV) construction emerges as a leading renewable energy technology, there is a growing focus on its implementation. However, the challenges of scarce, low-resolution, and inaccurate PV-related data sources hinder accurate assessments of urban PV potentials and are not conducive to efficient and rational smart city planning. This study tackles these challenges by introducing a mature, detailed, and accurate assessment process, taking Stonehaven as an example, aimed at leveraging limited data to mine more data and geographic information useful for guiding urban PV planning. Initially, utilise existing Digital Surface Model (DSM) and optical image data, combined with deep learning techniques and a PV potential assessment model, to comprehensively assess the PV power generation potential of the Stonehaven area. Our results demonstrate that integrating DSM significantly enhances the accuracy of roof segmentation. Furthermore, compared with DeeplabV3, U-Net performs better in roof segmentation. Additionally, the solar radiation potential (SRP) map generated by DSM highlights the superior solar radiation receiving capacity of south-facing and flat roofs. We provide a detailed PV power generation potential (PPGP) map of individual building roofs, revealing the substantial potential of this area for generating up to 1.12 × 10^7 kWh of electricity per year. Detailed and fine-grained PPGP can also help to optimise PV siting and electricity resource allocation. Furthermore, our return-on-investment period (ROIP) analysis indicates that most Stonehaven roofs have ROIPs between 8.1 and 11.3 years. The detailed ROIP distribution map can help people make informed PV investment decisions. Future research directions include enhancing data quality, refining segmentation algorithms, and exploring assisted urban energy planning analysis for smarter urban planning.

The correlation between left atrium voltage mapping and 3D late gadolinium enhancement cardiac magnetic resonance imaging findings in patients with known atrial fibrillation

Abstract Introduction The non-invasive cardiac magnetic resonance imaging-based late gadolinium enhancement (LGE-CMR) of the left atrium (LA) fibrosis distribution and degree can be utilized for preprocedural planning and real-time guidance of AF ablation. There is an ongoing discussion with somewhat conflicting results regarding the accuracy of LGE-CMR of the LA as compared with the invasive electroanatomical mapping (EAM). Earlier data mainly encompassed post-ablation patients with an EAM involving a relatively small number of acquired points. Aim To evaluate the accuracy of LGE-CMR-based LA fibrosis using the ADAS 3D LA software among ablation-naive AF patients as compared with EAM using the CARTO-3 navigation system. Methods Patients with non-permanent AF, naive to previous ablation/cardiac surgery, who underwent CARTO-based AF ablation with preceding LGE-CMR scan were selected for the study. The 3D substrate atrial maps were generated based on the LGE-CMR (CMR-fib) using the ADAS 3D LA software and the CARTO EAM (EAM-fib). A fibrosis threshold of 1.2-1.32 was established for the CMR-fib using the Image Intensity Ratio whereas a threshold of 0.2-0.5mV was applied for the EAM-fib. The two 3D-maps were assessed for fibrosis using anatomically synchronized quantitative point by point comparison. The EAM was set as the gold standard. Results The study included 26,777 points acquired from 17 patients (aged 65.8±7.92 yrs with 6 females and 10 paroxysmal AF cases). CMR-fib showed 22.0% of LGE, while the EAM-fib demonstrated 48.8% of low voltage areas. The CMR-fib vs. EAM-fib point by point comparison concluded with an agreement of 73.1% and interrater reliability of 0.45 (p<0.001). The sensitivity and specificity were 45.0% and 99.9%, respectively, while the PPV and NPV were 99.8% and 65.6%, respectively. The Area Under the Curve (AUC) was calculated as 72.4. Increase of the EAM-fib threshold for scar (to 0.5-1.5mV) led to a further decrease in the assessment diagnostic performance and the interrater reliability. The lowest agreement was tested in the LA roof segments and the higher in the inferoseptal and posterior walls. Conclusion The LGE-CMR underestimates the presence of scar as presented by the EAM, however, those detected are mostly accurate. The low sensitivity that was demonstrated between the two methods may be attributed to the lower spatial resolution of the CMR as well as the superficial subendocardial layer mapped by EAM as compared with the averaged transmural mapping of the CMR-fib. A follow-up larger-scale study is required to assess the accuracy of LGE-CMR-based assessment of fibrosis, considering various wall thicknesses and establishing the scar threshold on the ADAS software, particularly for ablation-naive patients.

An Attention-Based Full-Scale Fusion Network for Segmenting Roof Mask from Satellite Images

Accurately segmenting building roofs from satellite images is crucial for evaluating the photovoltaic power generation potential of urban roofs and is a worthwhile research topic. In this study, we propose an attention-based full-scale fusion (AFSF) network to segment a roof mask from the given satellite images. By developing an attention-based residual ublock, the channel relationship of the feature maps can be modeled. By integrating attention mechanisms in multi-scale feature fusion, the model can learn different weights for features of different scales. We also design a ladder-like network to utilize weakly labeled data, thereby achieving pixel-level semantic segmentation tasks assisted by image-level classification tasks. In addition, we contribute a new roof segmentation dataset, which is based on satellite images and uses the roof as the segmentation target rather than the entire building to further promote the algorithm research of estimating roof area using satellite images. The experimental results on the new roof segmentation dataset, WHU dataset, and IAIL dataset demonstrate the effectiveness of the proposed network.

Improving Semantic Segmentation of Roof Segments Using Large-Scale Datasets Derived from 3D City Models and High-Resolution Aerial Imagery

Advances in deep learning techniques for remote sensing as well as the increased availability of high-resolution data enable the extraction of more detailed information from aerial images. One promising task is the semantic segmentation of roof segments and their orientation. However, the lack of annotated data is a major barrier for deploying respective models on a large scale. Previous research demonstrated the viability of the deep learning approach for the task, but currently, published datasets are small-scale, manually labeled, and rare. Therefore, this paper extends the state of the art by presenting a novel method for the automated generation of large-scale datasets based on semantic 3D city models. Furthermore, we train a model on a dataset 50 times larger than existing datasets and achieve superior performance while applying it to a wider variety of buildings. We evaluate the approach by comparing networks trained on four dataset configurations, including an existing dataset and our novel large-scale dataset. The results show that the network performance measured as intersection over union can be increased from 0.60 for the existing dataset to 0.70 when the large-scale model is applied on the same region. The large-scale model performs superiorly even when applied to more diverse test samples, achieving 0.635. The novel approach contributes to solving the dataset bottleneck and consequently to improving semantic segmentation of roof segments. The resulting remotely sensed information is crucial for applications such as solar potential analysis or urban planning.

Space-to-speed architecture supporting acceleration on VHR image processing

One of the major focuses in the remote sensing community is the rapid processing of deep neural networks (DNNs) on very high resolution (VHR) aerial images. Few studies have investigated the acceleration of training and prediction by optimizing the architecture of the DNN system rather than designing a lightweight DNN. Parallel processing using multiple graphics processing units (GPUs) increases VHR image processing performance. It drives extremely large and frequent data transfers (input/output(I/O)) from random access memory (RAM) to GPU memory. As a result, the system bus congestion causes the system to hang, resulting in long latency in training/predicting. In this paper, we evaluate the causes of long latency and propose a space-to-speed (S2S) DNN system to overcome the aforementioned challenges. A three-level memory system aiming to reduce data transfer during system operation is presented. Distribution optimization with parallel processing was used to accelerate the training. Training optimizations on VHR images (such as hot-zone searching and image/ground truth queues for data saving) were used to train the VHR images efficiently. Inference optimization was performed to speed up prediction in the release mode. To verify the efficiency of the proposed system, we used aerial image labeling from the Institut National de Recherche en Informatique et en Automatique (INRIA) and benchmarks from the Massachusetts Institute of Technology Aerial Imagery for Roof Segmentation (MITAIRS) to test the system performance and accuracy. Without the loss of accuracy, the S2S system improved prediction speed on the testing dataset by eight GPUs in a normal setting in both the INRIA dataset (from 534 to 72 s) and the MITAIRS dataset (818 to 120 s). With the prediction in half-float (using float-16 data), an 8-GPU parallel processing increased the speed to 38 s in the INRIA dataset and 83 s in the MITAIRS dataset. In a pressure test, our proposed system operated on 18,000 images with a size of 5000 × 5000 from 18.2 to 1.8 h with the prediction in full-float (using float-32 data) and 43 min with the prediction in half-float, increasing the speed by a factor of 9.78 and 25.3, respectively, when compared to system runs without optimization.

Roof Segmentation From Airborne LiDAR Using Octree-Based Hybrid Region Growing and Boundary Neighborhood Verification Voting

Building roof segmentation is a key step in the process of three-dimensional (3D) building reconstruction using airborne Light Detection and Ranging (LiDAR) point cloud data. Voxel-based region growing is one of the most widely used methods to segment planes because of its high efficiency and easy implementation, but it is easy to omit roof planes due to the unreasonable voxel size and the complex roof structures. In addition, boundaries between adjacent roof planes are inaccurate. To solve the issues, a roof segmentation method using octree-based hybrid region growing and boundary neighborhood verification voting is proposed. First, an octree-based voxelization is performed on the raw building points to generate two basic units: planar voxels and individual points (i.e., points that are not in the planar voxels). Then, the hybrid region growing is conducted on these two basic units to segment coarse roof planes. A parameter-free boundary neighborhood verification voting strategy is used to assign the boundary points to the correct roof planes by verifying the neighborhoods of the boundary points and using reliable neighborhood information. Experimental results of four datasets, including two datasets provided by the International Society for Photogrammetry and Remote Sensing (ISPRS), and two high-density datasets provided by OpenTopography, verify that roof planes can be successfully segmented by the proposed method with over 96.8% completeness and a minimum 93.2% correctness. In addition, boundary points are assigned to the correct roof planes by the neighborhood verification voting strategy. Thus, the segmented roof planes can be used in various applications.

Prevalence, Predictors and Mechanisms of Steam Pops in Ablation Index-Guided High-Power Pulmonary Vein Isolation.

Despite the good cooling effect of the contact-force porous catheter, the risk of steam pops (SP) remains one of the major concerns in high-power circumferential pulmonary vein isolation (CPVI). This study aimed to investigate the prevalence, predictors and possible mechanisms of SPs in CPVI. Patients experiencing SPs in de novo high-power CPVI were 1:3 matched by non-SP patients with gender, age (±5 years) and left atrial diameter (LAD) (±5 mm) to compare the ablation parameters of SP and non-SP lesions. Catheter tip displacement (Tipdisp) was compared between "edge-of-ridge" and "PV-side-of-ridge" placement at anterior and roof segments of the left pulmonary vein (PV). SPs occurred in 11 (1.57%) of 701 patients, including 6 at the antero-superior left PV, 2 at the roof, 1 at the postero-superior left PV, 1 at the bottom left PV and 1 at the antero-superior aspect of the right PV. There was significantly shorter RF delivery duration (13.9 ± 6.3 vs. 23.3 ± 6.0 s), greater Δimpedance (17.6 ± 6.7 vs. 6.7 ± 4.1 Ω) and lower ablation index (357.7 ± 68.8 vs. 430.2 ± 30.7) in SP patients than those in non-SP patients. Δimpedance >12 Ω during ablation could predict SP occurrence. Tipdisp was greater in "PV-side-of-ridge" than that in "edge-of -ridge" placement (3.2 ± 1.6 mm vs. 2.0 ± 0.8 mm) at antero-superior and roof segments of the left PV. The prevalence of SP was 1.57% in high-power CPVI procedures, with the most common site at the antero-superior segment of the left PV. Δimpedance was a significant predictor of SP occurrence. "PV-side-of-ridge" ablation at antero-superior and roof segments of left PV might predispose to SP occurrence due to excessive tissue coverage.

VOXEL-BASED POINT CLOUD SEGMENTATION AND BUILDING DETECTION

Abstract. In the recent years, point cloud technologies, such as Unmanned Aerial Vehicles (UAV), Terrestrial Laser Scanners (TLS), Aerial Laser Scanners (ALS), let alone Mobile Mapping Systems (MMS), have come into the focus of attention and have been a subject of considerable public concern in mapping. Thanks to these new techniques, experts can survey large areas with sufficient and homogenous accuracy with high resolution. It comes from this that there are several areas where the point clouds can be used. One of the possible applications of point clouds is updating land registry maps. Many countries worldwide face the issue that a significant part of their large-scale land registry maps are outdated and inaccurate. One of these countries is Hungary, where more than eighty percent of digital cadastre maps were digitised using analogous maps in a scale range of 1:1000 – 1:4000. In this paper, a novel processing queue is presented to find the footprints of the building. Our solution is based on primarily well-known algorithms (RANSAC, DBSCAN) implemented in open-source Python packages. An automated flow was developed, composed of simple processing steps, to cut the point cloud into wall and roof segments and vectorise the wall points under roofs into building footprints. The algorithms and Python programs were tested in villages where detached houses are typical. Tests were made on three study areas in Hungary and we achieved well-promising results.

RID—Roof Information Dataset for Computer Vision-Based Photovoltaic Potential Assessment

Computer vision has great potential to accelerate the global scale of photovoltaic potential analysis by extracting detailed roof information from high-resolution aerial images, but the lack of existing deep learning datasets is a major barrier. Therefore, we present the Roof Information Dataset for semantic segmentation of roof segments and roof superstructures. We assessed the label quality of initial roof superstructure annotations by conducting an annotation experiment and identified annotator agreements of 0.15–0.70 mean intersection over union, depending on the class. We discuss associated the implications on the training and evaluation of two convolutional neural networks and found that the quality of the prediction behaved similarly to the annotator agreement for most classes. The class photovoltaic module was predicted to be best with a class-specific mean intersection over union of 0.69. By providing the datasets in initial and reviewed versions, we promote a data-centric approach for the semantic segmentation of roof information. Finally, we conducted a photovoltaic potential analysis case study and demonstrated the high impact of roof superstructures as well as the viability of the computer vision approach to increase accuracy. While this paper’s primary use case was roof information extraction for photovoltaic potential analysis, its implications can be transferred to other computer vision applications in remote sensing and beyond.

A Comprehensive Workflow for High Resolution 3D Solar Photovoltaic Potential Mapping in Dense Urban Environment: A Case Study on Campus of Delft University of Technology

Solar Photovoltaic Potential In article number 2100478, Hesan Ziar and co-workers developed a semi-automatized workflow to estimate residential solar photovoltaic (PV) potential, which only requires LiDAR data and building footprints as inputs, and delivers accurate geo-referenced 3D building models, annual solar irradiation map, annual DC/AC yield maps, and prioritizes roof segments based on their PV potential.

Automatic extraction of 3d building meshes from LiDAR data

Terrestrial laser scanning and drone-based scanning are often combined to gain a complete picture of the external envelopes of buildings. While the resulting data sets are easy to understand for humans, extracting any semantical information from the data is a challenging task. Creating a BIM model for large cities is desirable, especially for planning and inventory purposes. Nowadays, engineers manually separate buildings and draw corresponding floor plans or any other required entities, making creating such a model nearly impossible for the whole city.In this contribution, we present the design and concept of a C++ library that provides fundamental algorithmic tools for automated detection of the ground points, separation of buildings into individual point clouds, and segmentation of walls and roofs. The buildings’ 3D model (both volumetric and surface mesh) can be constructed using the data. Furthermore, the building area and volume calculation can be performed on such a model, which can then be compared with the existing city plan.

Two years after pulmonary vein isolation guided by ablation index-a multicenter study.

BackgroundThe use of the Ablation Index (AI) software for paroxysmal atrial fibrillation (AF) has been associated with higher acute effectiveness and higher 1‐year arrhythmia freedom. There is, however, a lack of data concerning longer follow‐up. We aim to evaluate the 2‐year outcomes after a standardized AI‐guided pulmonary vein isolation (PVI).MethodsProspective, multicenter study of consecutive patients referred for paroxysmal AF ablation from January 2018 to July 2019. PVI was guided by a tailored AI value (≥500 for anterior segment, ≥450 for the roof segments and inferior segments, and 400 for the posterior wall) and an ILD ≤6 mm. The primary endpoints were acute and long‐term effectiveness.ResultsThe study included 218 (842 PV) patients (61% males, median age of 60 [IQR 49–68] years) with paroxysmal AF. First‐pass isolation was obtained in 93% of the patients, with an acute reconnection occurring in 10.6% of the patients (3.2% of the PV) following adenosine trial. After a median follow‐up of 26 (IQR 20–30) months, freedom from any documented atrial arrhythmia was 83.4%, off‐AAD. The rate of adverse events was 1.4%. Although procedural parameters differ across centers (p < 0.001), the acute (p = 0.56) and long‐term effectiveness (p = 0.83) were consistent between centers.ConclusionsPatients with paroxysmal AF submitted to an AI‐guided PVI workflow presented high arrhythmia freedom at 2‐years of follow‐up.

Effect of Insulin Resistance and Role of Plasma Adiponectin Level in Susceptibility to Colorectal Cancer

Background: Catheter ablation (CA) is a potentially curative method for treatment of severe symptomatic and drug-refractory atrial fibrillation (AF). Objectives: This study aimed to evaluate the impact of steerable sheaths on catheter stability in paroxysmal AF based on contact force (CF). Methods: Fifty-two patients were included in this study and they were randomly enrolled to two groups: Pulmonary vein isolation using steerable (Group 1, n=26) or fixed-curve (Group 2, n=26) sheaths employing a force-sensing ablation catheter. We analyzed the operator-blinded and unblinded CFs when the operators were satisfied with the catheter position. Results: The average CF was 23.56±9.43 g (Group 1) vs. 22.03±10.56 g (Group 2) for the blind condition (P&lt;0.05) and 24.61±10.46 g (Group 1) vs. 22.18± 9.84 g (Group 2) for the unblinded condition (P&lt;0.05). There was significant heterogeneity of CFs between the segments: the CFs of the anterior-middle, anteroinferior, posterior-middle, and inferior posteroinferior segments of the right pulmonary vein (RPV), as well as of the roof, superior anterosuperior, anterior-middle, inferior anteroinferior, and inferior posteroinferior segments of the left pulmonary vein (LPV), showed statistical differences in the blinded condition (P&lt;0.05). The CFs of the roof, anterosuperior, anterior-middle, and inferior anteroinferior, and posteroinferior segments of the RPV, and LPV showed statistical differences in the unblinded condition (P&lt;0.05). Posterior and roof segments showed enhanced CFs in both groups. Group one had a lower proportion of acute reconnection rate and less tendency for recurrence. Conclusion: In conclusion, catheter stability improved with steerable sheaths owing to the potential of attaining higher CFs. This advantage was independent of CF guidance and exhibited a specific distribution pattern.

Roof Segmentation Towards Digital Twin Generation in LoD2+ Using Deep Learning

There is an increasing need for digital twins of cities and their base maps, 3D city models. Creating and updating these twins is not an easy task, so automating and streamlining the process is a field of active research. A significant part of the urban geometry is residential buildings and their roofs. Modeling of roofs for urban buildings can be divided into three main areas - building detection, roof recognition and building reconstruction. The building and roofs are segmented with the help of machine learning and image processing. Afterwards the extracted information is used to generate parametric models for the roofs using methods from computational geometry. The goal is to create correct virtual models of roofs belonging to many different types of buildings. In this study, a supervised deep learning approach is proposed for the segmentation of roof edges from a single orthophoto. The predicted features include the linear elements of roofs. The experiments show that, despite the small amount of training data, even in the presence of noise, the proposed method performs well on semantic segmentation of roofs with different shapes and complexities. The quality of the extracted roof elements for the test area is about 56% and 71% for mean intersection over union (IOU) and Dice metric scores, respectively.

Building Instance Mapping From ALS Point Clouds Aided by Polygonal Maps

Building region extraction from ALS point clouds has been widely studied, whereas instance-level building mapping has been overlooked and remains unsolved. In this study, we present a method to extract individual buildings from ALS point clouds with the help of widely accessible polygonal footprints. The key idea is to merge roof segments to a set of building candidates, from which correct instances are selected by finding optimal matches between polygonal footprints and building candidates. The method has three steps: roof segmentation, building candidate generation, and instance-polygon matching. The method is tested on two large-scale scenes of different building types and can generally achieve high instance-level building mapping accuracy (around 90%) when there are large positioning errors (6.0 m) among polygons. Future work will focus on classification errors in preprocessing, shape inconsistency between point clouds and polygons, and building footprint delineation and updating in postprocessing.

IRU-Net: An Efficient End-to-End Network for Automatic Building Extraction From Remote Sensing Images

Automatic extraction of buildings from High-Resolution Remote Sensing (RS) Imagery is of great practical interest for numerous applications; including urban planning, change detection, disaster management, estimation of human population, and many other geospatial related applications. This paper proposes a novel efficient Improved ResU-Net architecture called IRU-Net, integrating spatial pyramid pooling module with an encoder-decoder structure, in combination with Atrous convolutions, modified residual connections, and a new skip connection between the encoder-decoder features for automatic extraction of buildings from RS images. Moreover, a new dual loss function called binary cross-entropy-dice-loss (BCEDL) is opted that make cross-entropy (CE) and dice loss (DL) and consider both local information and global information to decrease the class imbalance influence and improve the building extraction results. The proposed model is examined to demonstrate its generalization on two publicly available datasets; the Aerial Images for Roof Segmentation (AIRS) Dataset and the Massachusetts buildings dataset. The proposed IRU-Net achieved an average F-1 accuracy of 92.34% for the Massachusetts dataset and 95.65% for the AIRS dataset. When compared to other state-of-the-art deep learning-based models such as SegNet, U-Net, E-Net, ERFNet and SRI-Net, the overall accuracy improvements of our IRU-Net model is 9.0% (0.9725 vs. 0.8842), 5.2% (0.9725 vs. 0.9218), 3.0% (0.9725 vs. 0.9428), 1.4% (0.9725 vs. 0.9588) and 0.93% (0.9725 vs. 0.9635), for AIRS dataset and 11.6%, 5.9%, 3.1%, 2.7% and 1.4%, for Massachusetts building dataset. These results prove the superiority of the proposed model for building extraction from high-resolution RS images.

POI Detection of High-Rise Buildings Using Remote Sensing Images: A Semantic Segmentation Method Based on Multitask Attention Res-U-Net

A Point-Of-Interest (POI) represents a specific point location that may be useful or interesting for people, and therefore each and every building footprint in a topographic map can be recognized as a POI. Automatic extraction of building footprints using remote sensing images has become a challenging and important research topic which is in demand for urban planning and development. Extensive studies have explored a variety of semantic segmentation methods using deep learning algorithms to achieve better performance in building footprint extraction, however the existing algorithms were shown to have some limitations which lead to poor segmentation results. Building roofs were recognized as building footprints in the previous studies. This is prone to error especially for high-rise buildings due to different sensor view angles. In this paper, we propose a multi-task Res-U-Net model with attention mechanism for the extraction of the building roofs and the whole building shapes from remote sensing images, then use an offset vector method to detect the footprints of the high-rise buildings based on the boundaries of the corresponding building roofs and shapes. We also apply the online food delivery (OFD) data to parse the POI name of every building footprint. Several strategies are also developed in combination with the proposed model, including data augmentation and post-processing. We conduct numerical experiments using real data of remote sensing images and OFD historical order data. Results demonstrate that our proposed model achieves a total F1-score of 77.05% and intersection over union (IoU) of 63.55% in terms of the building roof segmentation, and an overall F1-score of 79.02% and IoU of 66.05% for the whole building shape segmentation, which both achieve the best performance among all baseline models.

Extracting Rectified Building Footprints from Traditional Orthophotos: A New Workflow

Deep learning techniques such as convolutional neural networks have largely improved the performance of building segmentation from remote sensing images. However, the images for building segmentation are often in the form of traditional orthophotos, where the relief displacement would cause non-negligible misalignment between the roof outline and the footprint of a building; such misalignment poses considerable challenges for extracting accurate building footprints, especially for high-rise buildings. Aiming at alleviating this problem, a new workflow is proposed for generating rectified building footprints from traditional orthophotos. We first use the facade labels, which are prepared efficiently at low cost, along with the roof labels to train a semantic segmentation network. Then, the well-trained network, which employs the state-of-the-art version of EfficientNet as backbone, extracts the roof segments and the facade segments of buildings from the input image. Finally, after clustering the classified pixels into instance-level building objects and tracing out the roof outlines, an energy function is proposed to drive the roof outline to maximally align with the building footprint; thus, the rectified footprints can be generated. The experiments on the aerial orthophotos covering a high-density residential area in Shanghai demonstrate that the proposed workflow can generate obviously more accurate building footprints than the baseline methods, especially for high-rise buildings.

Spatiotemporal gait parameters while cross-slope residential roof walking

Falls from residential roofs account for 80% of roofing industry fatalities. Furthermore, roofing work represents 44.7% of work in residual construction specialty trades and residential roofers count for 2.1% of overall workers in construction, with an anticipated growth in roofers of 14.9% by 2024. The purpose of the study was to evaluate the alterations in spatiotemporal gait parameters while traversing along a 6/12 pitched residential roof segment. Eighteen of the nineteen calculated spatiotemporal variables were statistically, significantly changed by walking across a 6/12 pitched simulated residential roof. The study clearly demonstrates that spatiotemporal gait variables increase and decrease while traversing across a residential roof. The changes in spatiotemporal parameters might suggest alterations to a person's balance system resulting in an increased risk of falling. The knowledge generated in the current study will be relevant to the residential roofing industry when it can be used in educational materials to increase awareness of how a roofer's altered gait while working on a pitched roof may increase their falling risk.

Coal Wall and Roof Segmentation in the Coal Mine Working Face Based on Dynamic Graph Convolution Neural Networks.

The intersection line information of the point cloud between the coal wall and the roof can not only accurately reflect the direction information of the scraper conveyor but also provide a preliminary basis for realizing the intelligent coal mine. However, the indirect method of using deep learning to segment the point cloud of coal mine working face cannot make full use of the rich information provided by the point cloud data. The direct method of using deep learning to segment the point cloud ignores the local feature relationship between points. Therefore, we propose to use dynamic graph convolution neural networks (DGCNNs) to segment the point cloud of the coal wall and roof so as to obtain the intersection line between them. First, in view of the characteristics of heavy dust and strong electromagnetic interference in the environment of the coal mine working face, we have installed an underground inspection robot so that we use light detection and ranging to obtain the point cloud of the coal mine working face. At the same time, we put forward a fast labeling method of the point cloud of the coal mine working face and an efficient training method of the depth neural network. Second, on the basis of edge convolution, being the greatest innovation of DGCNNs, we analyze the influence of the number of layers, K value, and output feature dimension of edge convolution on the effect of DGCNNs segmenting the point cloud of the coal mine working face and obtaining the intersection line of the coal wall and roof. Finally, we compare DGCNNs with PointNet and PointNet++. The results show that the DGCNN exhibits the best performance. What is more, the results provide a research foundation for the application of DGCNNs in the field of energy. Last but not least, the research results provide a direct and key basis for the adjustment of the scraper conveyor, which is of great significance for an intelligent coal mine working face and accurate construction of a geological information model.