Chapter

An Overview of Methods for Control and Estimation of Capacity in COVID-19 Pandemic from Point Cloud and Imagery Data

Authors:
To read the full-text of this research, you can request a copy directly from the authors.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

ResearchGate has not been able to resolve any citations for this publication.
Article
Full-text available
With the advancement of light detection and ranging (LiDAR) technology, the mobile laser scanner (MLS) has been regarded as an important technology to collect geometric representations of the indoor environment. In particular, methods for detecting indoor objects from indoor point cloud data (PCD) captured through MLS have thus far been developed based on the trajectory of MLS. However, the existing methods have a limitation on applying to an indoor environment where the building components made by concrete impede obtaining the information of trajectory. Thus, this study aims to propose a building component detection algorithm for MLS-based indoor PCD without trajectory using random sample consensus (RANSAC)-based region growth. The proposed algorithm used the RANSAC and region growing to overcome the low accuracy and uniformity of MLS caused by the movement of LiDAR. This study ensures over 90% precision, recall, and proper segmentation rate of building component detection by testing the algorithm using the indoor PCD. The result of the case study shows that the proposed algorithm opens the possibility of accurately detecting interior objects from indoor PCD without trajectory information of MLS.
Article
Full-text available
Due to the rapid aging of the population in recent years, the number of elderly people in hospitals and nursing homes is increasing, which results in a shortage of staff. Therefore, the situation of elderly citizens requires real-time attention, especially when dangerous situations such as falls occur. If staff cannot find and deal with them promptly, it might become a serious problem. For such a situation, many kinds of human motion detection systems have been in development, many of which are based on portable devices attached to a user’s body or external sensing devices such as cameras. However, portable devices can be inconvenient for users, while optical cameras are affected by lighting conditions and face privacy issues. In this study, a human motion detection system using a low-resolution infrared array sensor was developed to protect the safety and privacy of people who need to be cared for in hospitals and nursing homes. The proposed system can overcome the above limitations and have a wide range of application. The system can detect eight kinds of motions, of which falling is the most dangerous, by using a three-dimensional convolutional neural network. As a result of experiments of 16 participants and cross-validations of fall detection, the proposed method could achieve 98.8% and 94.9% of accuracy and F1-measure, respectively. They were 1% and 3.6% higher than those of a long short-term memory network, and show feasibility of real-time practical application.
Conference Paper
Full-text available
Point cloud data of indoor scenes is primarily composed of planar-dominant elements. Automatic shape segmentation is thus valuable to avoid labour intensive labelling. This paper provides a fully unsupervised region growing segmentation approach for efficient clustering of massive 3D point clouds. Our contribution targets a low-level grouping beneficial to object-based classification. We argue that the use of relevant segments for object-based classification has the potential to perform better in terms of recognition accuracy, computing time and lowers the manual labelling time needed. However, fully unsupervised approaches are rare due to a lack of proper generalisation of user-defined parameters. We propose a self-learning heuristic process to define optimal parameters, and we validate our method on a large and richly annotated dataset (S3DIS) yielding 88.1% average F1-score for object-based classification. It permits to automatically segment indoor point clouds with no prior knowledge at commercially viable performance and is the foundation for efficient indoor 3D modelling in cluttered point clouds.
Article
Full-text available
Crowd management and monitoring is crucial for maintaining public safety and is an important research topic. Developing a robust crowd monitoring system (CMS) is a challenging task as it involves addressing many key issues such as density variation, irregular distribution of objects, occlusions, pose estimation, etc. Crowd gathering at various places like hospitals, parks, stadiums, airports, cultural and religious points are usually monitored by Close Circuit Television (CCTV) cameras. The drawbacks of CCTV cameras are: limited area coverage, installation problems, movability, high power consumption and constant monitoring by the operators. Therefore, many researchers have turned towards computer vision and machine learning that have overcome these issues by minimizing the need of human involvement. This review is aimed to categorize, analyze as well as provide the latest development and performance evolution in crowd monitoring using different machine learning techniques and methods that are published in journals and conferences over the past five years.
Article
Full-text available
Building Information Modelling (BIM) is a globally adapted methodology by government organisations and builders who conceive the integration of the organisation, planning, development and the digital construction model into a single project. In the case of a heritage building, the Historic Building Information Modelling (HBIM) approach is able to cover the comprehensive restoration of the building. In contrast to BIM applied to new buildings, HBIM can address different models which represent either periods of historical interpretation, restoration phases or records of heritage assets over time. Great efforts are currently being made to automatically reconstitute the geometry of cultural heritage elements from data acquisition techniques such as Terrestrial Laser Scanning (TLS) or Structure From Motion (SfM) into BIM (Scan-to-BIM). Hence, this work advances on the parametric modelling from remote sensing point cloud data, which is carried out under the Rhino+Grasshopper-ArchiCAD combination. This workflow enables the automatic conversion of TLS and SFM point cloud data into textured 3D meshes and thus BIM objects to be included in the HBIM project. The accuracy assessment of this workflow yields a standard deviation value of 68.28 pixels, which is lower than other author’s precision but suffices for the automatic HBIM of the case study in this research.
Article
Full-text available
Improving the resilience of infrastructures is key to reduce their risk vulnerability and mitigate impact from hazards at different levels (e.g., from increasing extreme events, driven by climate change); or from human-made events such as: accidents, vandalism or terrorist actions. One of the most relevant aspects of resilience is preparation. This is directly related to: (i) the risk prediction capability; (ii) the infrastructure monitoring; and (iii) the systems contributing to anticipate, prevent and prepare the infrastructure for potential damage. This work focuses on those methods and technologies that contribute to more efficient and automated infrastructure monitoring. Therefore, a review that summarizes the state of the art of LiDAR (Light Detection And Ranging)-based data processing is presented, giving a special emphasis to road and railway infrastructure. The most relevant applications related to monitoring and inventory transport infrastructures are discussed. Furthermore, different commercial LiDAR-based terrestrial systems are described and compared to offer a broad scope of the available sensors and tools to remote monitoring infrastructures based on terrestrial systems.
Article
Full-text available
Many BIM experts agree that employing BIM for new construction is an easy task. However, applying BIM to existing construction will be difficult but is more suitable for heritage buildings cases. These heritage buildings have unique façades and architectural vocabulary, which are of special interest. Furthermore, studying these architecture heritages require some advanced tools in order to understand and analyse their structure, components, and design. Relying only on traditional methods is not adequate, especially for architectural engineers and experts who need digital representations of architectural heritage in order to draw a complete image of any aspect of the project. Moreover, lots of these heritage architectural elements are not documented or provided in the digital architectural libraries, which in turn requires advanced and easy access methods and tools that can extract basic information professionally and explain the essence of heritage. BIM has emerged as an efficient solution that could possibly help in analysing architectural heritage through effective learning processes. Existing BIM is characterised by their ability to create and operate within a digital database of any existing by 3-D laser through scanning the building and transforming it into point-cloud as digital data, so that engineers and experts can work on existing and buildings via the BIM software. As with many heritage buildings in the world, many of the heritage buildings in the Historic district of Jeddah city, Saudi Arabia, face serious issues in terms of conservation, restoration, documentation, managing, recording, and monitoring of these valuable heritage buildings. Therefore, this paper will examine and evaluate the use of BIM in modelling and for simulation purposes, (e.g. structure and energy simulation) with regard to one of the existing heritage buildings in the Historic district of Jeddah.
Article
Full-text available
The tracking of people is an indispensable capacity in almost any robotic application. A relevant case is the @home robotic competitions, where the service robots have to demonstrate that they possess certain skills that allow them to interact with the environment and the people who occupy it; for example, receiving the people who knock at the door and attending them as appropriate. Many of these skills are based on the ability to detect and track a person. It is a challenging problem, particularly when implemented using low-definition sensors, such as Laser Imaging Detection and Ranging (LIDAR) sensors, in environments where there are several people interacting. This work describes a solution based on a single LIDAR sensor to maintain a continuous identification of a person in time and space. The system described is based on the People Tracker package, aka PeTra, which uses a convolutional neural network to identify person legs in complex environments. A new feature has been included within the system to correlate over time the people location estimates by using a Kalman filter. To validate the solution, a set of experiments have been carried out in a test environment certified by the European Robotic League.
Article
Full-text available
This paper presents a system for online learning of human classifiers by mobile service robots using 3D LiDAR sensors, and its experimental evaluation in a large indoor public space. The learning framework requires a minimal set of labelled samples (e.g. one or several samples) to initialise a classifier. The classifier is then retrained iteratively during operation of the robot. New training samples are generated automatically using multi-target tracking and a pair of “experts” to estimate false negatives and false positives. Both classification and tracking utilise an efficient real-time clustering algorithm for segmentation of 3D point cloud data. We also introduce a new feature to improve human classification in sparse, long-range point clouds. We provide an extensive evaluation of our the framework using a 3D LiDAR dataset of people moving in a large indoor public space, which is made available to the research community. The experiments demonstrate the influence of the system components and improved classification of humans compared to the state-of-the-art.
Article
Full-text available
In the field of intelligent transportation systems, pedestrian detection has become a problem that is urgently in need of a solution. Effective pedestrian detection reduces accidents and protects pedestrians from injuries. A pedestrian-detection algorithm, namely, single template matching with kernel density estimation clustering (STM-KDE), is proposed in this paper. First, the KDE-based clustering method is utilized to extract candidate pedestrians in point clouds. Next, the coordinates of the point clouds are transformed into the pedestrians’ local coordinate system and projection images are generated. Locally adaptive regression kernel features are extracted from the projection image and matched with the template features by using cosine similarity, based on which pedestrians are distinguished from other columnar objects. Finally, comparative experiments using KITTI datasets are conducted to verify pedestrian-detection performance. Compared with the STM with radially bounded nearest neighbor (STM-RBNN) algorithm and the KDE-based pedestrian-detection algorithm, the proposed algorithm can segment gathering pedestrians and distinguish them from other columnar objects in real scenarios.
Article
Full-text available
In the event of a flood, being able to build accurate flood level maps is essential for supporting emergency plan operations. In order to build such maps, it is important to collect observations from the disaster area. Social media platforms can be useful sources of information in this case, as people located in the flood area tend to share text and pictures depicting the current situation. Developing an effective and fully automatized method able to retrieve data from social media and extract useful information in real-time is crucial for a quick and proper response to these catastrophic events. In this paper, we propose a method to quantify flood-water from images gathered from social media. If no prior information about the zone where the picture was taken is available, one possible way to estimate the flood level consists of assessing how much the objects appearing in the image are submerged in water. There are various factors that make this task difficult: i) the precise size of the objects appearing in the image might not be known; ii) flood-water appearing in different zones of the image scene might have different height; iii) objects may be only partially visible as they can be submerged in water. In order to solve these problems, we propose a method that first locates selected classes of objects whose sizes are approximately known, then, it leverages this property to estimate the water level. To prove the validity of this approach, we first build a flood-water image dataset, then we use it to train a deep learning model. We finally show the ability of our trained model to recognize objects and at the same time predict correctly flood-water level.
Article
Full-text available
It is important to measure and analyze people behavior to design systems which interact with people. This paper describes a portable people behavior measurement system using a 3D-LIDAR. In this system, an observer carries the system equipped with a 3D LIDAR and follows persons to be measured while keeping them in the sensor view. The system estimates the sensor pose in a 3D environmental map and tracks the target persons. It enables long-term and wide-area people behavior measurements which are hard for existing people tracking systems. As a field test, we recorded the behavior of professional caregivers attending elderly persons with dementia in a hospital. The preliminary analysis of the behavior reveals how the caregivers decide the attending position while checking the surrounding people and environment. Based on the analysis result, empirical rules to design the behavior of attendant robots are proposed.
Chapter
Full-text available
Object detection and classification in 3D is a key task in Automated Driving (AD). LiDAR sensors are employed to provide the 3D point cloud reconstruction of the surrounding environment, while the task of 3D object bounding box detection in real time remains a strong algorithmic challenge. In this paper, we build on the success of the one-shot regression meta-architecture in the 2D perspective image space and extend it to generate oriented 3D object bounding boxes from LiDAR point cloud. Our main contribution is in extending the loss function of YOLO v2 to include the yaw angle, the 3D box center in Cartesian coordinates and the height of the box as a direct regression problem. This formulation enables real-time performance, which is essential for automated driving. Our results are showing promising figures on KITTI benchmark, achieving real-time performance (40 fps) on Titan X GPU.
Article
Full-text available
Tracking people has many applications, such as security or safe use of robots. Many onboard systems are based on Laser Imaging Detection and Ranging (LIDAR) sensors. Tracking peoples' legs using only information from a 2D LIDAR scanner in a mobile robot is a challenging problem because many legs can be present in an indoor environment, there are frequent occlusions and self-occlusions, many items in the environment such as table legs or columns could resemble legs as a result of the limited information provided by two-dimensional LIDAR usually mounted at knee height in mobile robots, etc. On the other hand, LIDAR sensors are affordable in terms of the acquisition price and processing requirements. In this article, we describe a tool named PeTra based on an off-line trained full Convolutional Neural Network capable of tracking pairs of legs in a cluttered environment. We describe the characteristics of the system proposed and evaluate its accuracy using a dataset from a public repository. Results show that PeTra provides better accuracy than Leg Detector (LD), the standard solution for Robot Operating System (ROS)-based robots.
Article
Full-text available
Aerial laser scanning or photogrammetric point clouds are often noisy at building boundaries. In order to produce regularized polygons from such noisy point clouds, this study proposes a hierarchical regularization method for the boundary points. Beginning with detected planar structures from raw point clouds, two stages of regularization are employed. In the first stage, the boundary points of an individual plane are consolidated locally by shifting them along their refined normal vector to resist noise, and then grouped into piecewise smooth segments. In the second stage, global regularities among different segments from different planes are softly enforced through a labeling process, in which the same label represents parallel or orthogonal segments. This is formulated as a Markov random field and solved efficiently via graph cut. The performance of the proposed method is evaluated for extracting 2D footprints and 3D polygons of buildings in metropolitan area. The results reveal that the proposed method is superior to the state-of-art methods both qualitatively and quantitatively in compactness. The simplified polygons could fit the original boundary points with an average residuals of 0.2 m, and in the meantime reduce up to 90% complexities of the edges. The satisfactory performances of the proposed method show a promising potential for 3D reconstruction of polygonal models from noisy point clouds.
Article
Full-text available
In the last years, point clouds have become the main source of information for building modelling. Although a considerable amount of methodologies addressing the automated generation of 3D models from point clouds have been developed, indoor modelling is still a challenging task due to complex building layouts and the high presence of severe clutters and occlusions. Most of methodologies are highly dependent on data quality, often producing irregular and non-consistent models. Although manmade environments generally exhibit some regularities, they are not commonly considered. This paper presents an optimization-based approach for detecting regularities (i.e., same shape, same alignment and same spacing) in building indoor features. The methodology starts from the detection of openings based on a voxel-based visibility analysis to distinguish ‘occluded’ from ‘empty’ regions in wall surfaces. The extraction of regular patterns in windows is addressed from studying the point cloud from an outdoor perspective. The layout is regularized by minimizing deformations while respecting the detected constraints. The methodology applies for elements placed in the same plane.
Article
Full-text available
3D models derived from point clouds are useful in various shapes to optimize the trade-off between precision and geometric complexity. They are defined at different granularity levels according to each indoor situation. In this article, we present an integrated 3D semantic reconstruction framework that leverages segmented point cloud data and domain ontologies. Our approach follows a part-to-whole conception which models a point cloud in parametric elements usable per instance and aggregated to obtain a global 3D model. We first extract analytic features, object relationships and contextual information to permit better object characterization. Then, we propose a multi-representation modelling mechanism augmented by automatic recognition and fitting from the 3D library ModelNet10 to provide the best candidates for several 3D scans of furniture. Finally, we combine every element to obtain a consistent indoor hybrid 3D model. The method allows a wide range of applications from interior navigation to virtual stores.
Article
Full-text available
The techniques of measuring and 3D modelling based on images, as is typical in photogrammetry, grew in interest again in recent years, since a new generation of software tools has spread. These ones implement in different measure the algorithms developed by computer vision, increasing the automation of the standard photogrammetric process. This made the use of image-based approaches for 3D models reconstruction enormously increase, which is an essential part of the Cultural Heritage documentation and analysis processes. Starting from these assumptions, the aim of the paper is to evaluate what and where it is possible nowadays to find the main differences between photogrammetry and computer vision approaches and how these have to be considered in the choice of the processing technique. The analysis has been performed starting from a theoretical point of view in order to trace the main characteristics of the two methods. Moreover, in order to complete the investigation, an experimental part is reported on two particular cases study, considered as representative of two types of usually surveyed objects. The results allow to enlighten some differences between the two image processing approaches, in terms of accuracy and achieved products.
Article
3D pedestrian detection is a challenging task in automated driving because pedestrians are relatively small, frequently occluded and easily confused with narrow vertical objects. LiDAR and camera are two commonly used sensor modalities for this task, which should provide complementary information. Unexpectedly, LiDAR-only detection methods tend to outperform multisensor fusion methods in public benchmarks. Recently, PointPainting has been presented to eliminate this performance drop by effectively fusing the output of a semantic segmentation network instead of the raw image information. In this paper, we propose a generalization of PointPainting to be able to apply fusion at different levels. After the semantic augmentation of the point cloud, we encode raw point data in pillars to get geometric features and semantic point data in voxels to get semantic features and fuse them in an effective way. Experimental results on the KITTI test set show that SemanticVoxels achieves state-of-the-art performance in both 3D and bird's eye view pedestrian detection benchmarks. In particular, our approach demonstrates its strength in detecting challenging pedestrian cases and outperforms current state-of-the-art approaches.
Article
Last developments in terrestrial surveying technologies, regarding both hardware and software, provide the possibility for their application to new interesting uses. Among them, this article focuses on indoor mapping and surveying with mobile devices. The use of these devices implies an improvement in the flexibility of the data acquisition in comparison to the static equivalent. This article analyses and compares the different options for mobile indoor mapping currently in the market, focusing on interesting characteristics for construction and architectural studies, such as physical configuration, weight, sensor type, colouring possibilities and operating time. The aim of this paper is to show the reader the state of the art of the technology, helping to fill the gap of information which exists in this field of technology and to help on the selection of the best option depending on the requirements.
Chapter
As three-dimensional point clouds become an increasingly popular for representing spatial data, with it comes a need to classify objects represented in this form. This paper proposes an image feature representation technique which classifies 3D point clouds by using several 2D perspectives, and then using YOLO (v3) as object classification tool to classify the rasters. Existing works have had limited success using this technique with only one perspective, likely due to the considerable information loss during the projection. We hypothesize that multiple projections mitigate these effects by capturing lost data in other perspective projections. Our method is effective in classifying pedestrians and vehicles in the Sydney Urban Objects data set, where our technique has achieved a classification accuracy of \(98.42\%\) and an f1 score of 0.9843, which are considerably higher than existing methods.
Preprint
Automated human action recognition is one of the most attractive and practical research fields in computer vision, in spite of its high computational costs. In such systems, the human action labelling is based on the appearance and patterns of the motions in the video sequences; however, the conventional methodologies and classic neural networks cannot use temporal information for action recognition prediction in the upcoming frames in a video sequence. On the other hand, the computational cost of the preprocessing stage is high. In this paper, we address challenges of the preprocessing phase, by an automated selection of representative frames among the input sequences. Furthermore, we extract the key features of the representative frame rather than the entire features. We propose a hybrid technique using background subtraction and HOG, followed by application of a deep neural network and skeletal modelling method. The combination of a CNN and the LSTM recursive network is considered for feature selection and maintaining the previous information, and finally, a Softmax-KNN classifier is used for labelling human activities. We name our model as "Feature Reduction \& Deep Learning" based action recognition method, or FR-DL in short. To evaluate the proposed method, we use the UCF dataset for the benchmarking which is widely-used among researchers in action recognition research. The dataset includes 101 complicated activities in the wild. Experimental results show a significant improvement in terms of accuracy and speed in comparison with six state-of-the-art articles.
Article
The application of Deep Learning techniques to point clouds for urban object classification is limited by the large number of samples needed. Acquiring and tagging point clouds is more expensive and tedious labour than its image equivalent process. Point cloud online datasets contain few samples for Deep Learning or not always the desired classes This work focuses on minimizing the use of point cloud samples for neural network training in urban object classification. The method proposed is based on the conversion of point clouds to images (pc-images) because it enables: the use of Convolutional Neural Networks, the generation of several samples (images) per object (point clouds) by means of multi-view, and the combination of pc-images with images from online datasets (ImageNet and Google Images). The study is conducted with ten classes of objects extracted from two street point clouds from two different cities. The network selected for the job is the InceptionV3. The training set consists of 5000 online images with a variable percentage (0% to 10%) of pc-images. The validation and testing sets are composed exclusively of pc-images. Although the network trained only with online images reached 47% accuracy, the inclusion of a small percentage of pc-images in the training set improves the classification to 99.5% accuracy with 6% pc-images. The network is also applied at IQmulus & TerraMobilita Contest dataset and it allows the correct classification of elements with few samples.
Article
This paper proposes a human gait tracking method using a low scanning rate (10Hz) 360 degree 2D LiDAR. The sensor is placed at a height of 0.3m above the ground. Since the human gait tracking is important not only for normal people but also for patients who need to use walking-aid such as walker, we propose a tracking method for both normal people and walker users. The left and right legs are separated and tracked even with nonlinear walking path. To overcome the occlusions between legs when one leg might be hidden from the sensor, a spline-based smooth and interpolation method is applied to estimate the legs trajectories. An algorithm is also proposed to remove the walker legs data in case of walker users. The performance of proposed tracking method is investigated in different walking speed levels and two positions of LiDAR (parallel and horizontal to walking direction). The experimental results verify the robustness of our proposed method. The horizontal position of LiDAR gives better estimation than the parallel position. The RMSE is less than 2.0cm for walking step length estimation and is less than 1.4cm for walking distance estimation. The tracking range of our system is up to 10m with horizontal placement of LiDAR.
Article
This paper intends to analyze the Light Detection and Ranging (Lidar) sensor performance on detecting pedestrians under different weather conditions. Lidar sensor is the key sensor in autonomous vehicles, which can provide high-resolution object information. Thus, it is important to analyze the performance of Lidar. This paper involves an autonomous bus operating several pedestrian detection tests in a parking lot at the University at Buffalo. By comparing the pedestrian detection results on rainy days with the results on sunny days, the evidence shows that the rain can cause unstable performance and even failures of Lidar sensors to detect pedestrians in time. After analyzing the test data, three logit models are built to estimate the probability of Lidar detection failure. The rainy weather still plays an important role in affecting Lidar detection performance. Moreover, the distance between a vehicle and a pedestrian, as well as the autonomous vehicle velocity, are also important. This paper can provide a way to improve the Lidar detection performance in autonomous vehicles.
Article
The capillary spread of personal devices equipped with wireless communication capabilities has enabled a series of high level services which build on capturing and processing the data packets emitted by such devices. In this paper we tackle the problem of exploiting such a methodology to perform occupancy estimation, i.e., understanding how many people are present in a specific place, an information which is valuable in many scenarios (HVAC and lighting system control, building energy optimisation, allocation and reservation of spaces, etc.). Traditional systems for occupancy estimation rely either on environmental sensors (CO2, temperature, humidity) or video cameras, which both have drawbacks: the former have generally low accuracy, while the latter require high setup and maintenance costs. In this paper we depart from such traditional approaches and propose a cheap and accurate occupancy estimation system based on the capture of both Wi-Fi and Bluetooth or Bluetooth Low Energy management frames transmitted from users’ devices. The system, implemented on low-cost hardware, leverages a supervised learning model to adapt to different spaces and transmits estimated occupancy information to a web-based dashboard. Experimental results in both indoor and outdoor uncontrolled scenarios demonstrate the validity of the proposed solution.
Chapter
In this paper, we address the problem of real-time human pose-based robust person tracking for a person following shopping support robot. We achieve this by cropping the target person’s body from the image and then apply a color histogram matching algorithm for tracking a unique person. After tracking the person, we used an omnidirectional camera and ultrasonic sensor to find the target person’s location and distance from the robot. When the target person is fixed in front of shopping shelves the robot finds the fixed distance between the robot and target person. In this situation our robot finds the target person’s body movement orientation using our proposed methodology. According to the body orientation our robot assumes a suitable position so that the target person can easily put his shopping product in the basket. Our proposed system was verified in real time environments and it shows that our robot system is highly effective at following a given target person and provides proper support while shopping the target person.
Article
Recent developments in laser scanning systems have inspired substantial interest in indoor modelling. Considering the cellular nature (i.e., rooms) of indoor spaces, recent studies have often reconstructed 3D indoor models from point clouds in heavily occluded environments by room space decomposition. However, recent space decomposition methods have excluded connection spaces (e.g., doors) and neglected the fact that room spaces are connected by connection spaces. These methods have thus failed to determine the correct room spaces in the case of unclear wall constraints from point clouds. Difficulties also arise from modelling stair connection spaces when the method is extended to a multi-storey environment. In this study, the semantic structural constraints of architectural components are defined. Three-dimensional indoor modelling from point clouds is cast as indoor space decomposition and recognition with structural constraints. The room space decomposition method is improved using wall structural constraints to effectively decompose the room space. The connection spaces, including wall connection spaces and stair connection spaces, are recognized from point clouds to generate a unified 3D indoor model. The proposed method is verified with two synthetic and four real-world datasets. The results indicate that the proposed method is reliable and accurate, especially for complex and large-scale indoor scenes.
Article
This letter presents a resampling algorithm for the refinement of the aerial laser scanning of urban buildings. We aim at resolving the problems of noise, variable density, and vacancies in the 3-D laser scanning. The assumption is that the points are locally sampled from piecewise building surfaces, so we can generate a new point cloud to present a plausible approximation of the surface based on the local smooth trait. First, we introduce an oversegmentation approach to generate super voxels by analyzing local geometric similarities. Then, each voxel is treated as a node composing a graph. Based on the combination method of the graph-constrained filtering and the interpolation, the resampling approach is designed to reproject the points to their optimized positions. This approach could ensure that the resampling points are faithful to the initial data and maintain a global even density. Furthermore, we evaluated the effectiveness and efficiency of the proposed method on the applications of geometry primitive extraction and surface modeling. The experiments demonstrate that the approach is suitable for improving the quality of aerial scanning data in different aspects.
Article
Pathfinding applications for the citizen in urban environments are usually designed from the perspective of a driver, not being effective for pedestrians. In addition, urban scenes have multiple elements that interfere with pedestrian routes and navigable space. In this paper, a methodology for the direct use of point clouds for pathfinding in urban environments is presented, solving the main limitations for this purpose: (a) the excessive number of points is reduced for transformation into nodes on the final graph, (b) urban static elements acting as permanent obstacles, such as furniture and trees, are delimited and differentiated from dynamic elements such as pedestrians, (c) occlusions on ground elements are corrected to enable a complete graph modelling, and (d) navigable space is delimited from free unobstructed space according to two motor skills (pedestrians without reduced mobility and wheelchairs). The methodology is tested into three different streets sampled as point clouds by mobile laser scanning (MLS) systems: an intersection of several streets with ground composed of sidewalks at different heights; an avenue with wide sidewalks, trees and cars parked on one side; and a street with a single-lane road and narrow sidewalks. By applying Dijkstra pathfinding algorithm to the resulting graphs, the correct viability of the generated routes has been verified based on a visual analysis of the generated routes on the point cloud and on the knowledge of the urban study area. The methodology enables the automatic generation of graphs representing the navigable urban space, on which safe and real routes for different motor skills can be calculated.
Article
Light Detection and Ranging (LiDAR) is a remote sensing technology widely used in many areas ranging from making precise medical equipment to creating accurate elevation maps of farmlands. In transportation, although it has been used to assist some design and planning works, the application has been predominantly focused on autonomous vehicles, regardless of its great potential in precise detection and tracking of all road users if implemented in the field. This paper explores fundamental concepts, solution algorithms, and application guidance associated with using infrastructure-based LiDAR sensors to accurately detect and track pedestrians and vehicles at intersections. Based on LiDAR data collected in the field, investigations were conducted in the order of background filtering, object clustering, pedestrian and vehicle classification, and tracking. The results of the analysis include accurate and real-time information of the presence, position, velocity, and direction of pedestrians and vehicles. By studying the data from infrastructure-mounted LiDAR sensors at intersections, this paper offers insights into some critical techniques that are valuable to both researchers and practitioners toward field implementation of LiDAR sensors.
Article
In the recent past, deep learning methods have demonstrated remarkable success for supervised learning tasks in multiple domains including computer vision, natural language processing, and speech processing. In this article, we investigate the impact of deep learning in the field of biometrics, given its success in other domains. Since biometrics deals with identifying people by using their characteristics, it primarily involves supervised learning and can leverage the success of deep learning in other related domains. In this article, we survey 100 different approaches that explore deep learning for recognizing individuals using various biometric modalities. We find that most deep learning research in biometrics has been focused on face and speaker recognition. Based on inferences from these approaches, we discuss how deep learning methods can benefit the field of biometrics and the potential gaps that deep learning approaches need to address for real-world biometric applications.
Conference Paper
The progress of IoT technologies, which connect, control and operate everything in the physical world, is expected to realize secure and convenient societies and communities, and embodies a variety of smart city applications. However, not a few smart city applications are also "human-centric", which require individuals' and human crowds' locations and behavior. In this paper, we present our project that realizes a human tracking and sensing platform called Hitonavi for enabling location sharing and trajectory identification of people. The core technology is LIDAR-based highly-accurate tracking of people as well as human crowd detection, and smartphone-assisted trajectory identification for enabling "pinpoint" indoor location services in retail shops, exhibition halls, office buildings, commercial complex, train stations and so on. Leveraging these functions, we may build a variety of smart city applications on top of the Hitonavi platform, which would bring new values for smart communities, smart buildings and many other smart city applications. We introduce the concept and key technologies of the Hitonavi platform and discuss potential applications.