Architectural Floor Plans Research Articles

Multiscale object detection on complex architectural floor plans

Architectural floor plans are essential documents for conveying building information among designers, engineers, and clients. Automated analysis of floor plans enhances user productivity and accuracy, though research on automatic object detection within architectural floor plans has been limited. In this paper, a convolutional neural network (CNN) based architecture, ArchNetv2, is proposed to detect various visual objects, such as stairs, windows, and doors. The proposed ArchNetv2 includes a convolutional block attention module to improve feature learning. It works at multiple detection scales and can efficiently recognize large objects (e.g., stairs) and small objects (e.g., windows) simultaneously. Experimental results show that ArchNetv2 can recognize thirteen types of objects commonly found in architectural floor plans with a mAP of 93.5%, which is superior compared to the state-of-the-art techniques. The proposed architecture can serve as an important module in an automated floor plan analysis system.

Simulation of emergency evacuation with crowd-gathering risk and evaluation methods: a case study of the mass rapid transit system in Taipei

ABSTRACT Mass rapid transit (MRT) systems are the backbone of transportation infrastructure in many cities. During emergencies, indoor subway stations become particularly susceptible to panic and overcrowding, making evacuation difficult. Therefore, simulating the evacuation and sheltering processes of a large number of individuals during emergencies is key in urban disaster response. In this study, a spatial evacuation risk model was used to enhance the safety of evacuation strategies in an MRT system. Data were collected on spatial architectural floor plans, and a database for communication and pedestrian flow was established. The behavioral patterns and distribution of individuals within stations and the characteristics of refuge spaces were also investigated. To simulate the processes of evacuation and sheltering, the Pathfinder evacuation simulation tool was used to construct a three-dimensional spatial model of a subway station. Specific scenarios were then constructed to simulate the procedures of evacuation and sheltering. Subsequently, the factors contributing to overcrowding-related hazards were analyzed to examine the spatial risks associated with crowd aggregation and evacuation and provide recommendations for improvement. Given its numerous advantages, this approach can serve as a valuable reference for conducting risk assessments in other large-scale crowd aggregation spaces.

Deep learning-based text detection and recognition on architectural floor plans

An important aspect of automatic floor plan analysis is the extraction of textual information, as it is essential for a thorough understanding of the drawing. This paper presents a text extraction approach utilizing a deep learning-based object detection model and state-of-the-art Optical Character Recognition (OCR) methods. The paper contributes to the research community in three ways: First, it introduces additional annotations to existing data sets to encompass text elements. Second, it proposes a specialized data synthesis pipeline, allowing for generating training images that mimic important characteristics of real data. Finally, it documents a comparative study of deep learning-based object detection architectures (Tesseract, EAST, CRAFT, Faster R-CNN, YOLOv5, YOLOR, YOLOv7, and YOLOv8) and OCR tools (PARSEq, MATRN, EasyOCR, and Tesseract) for the task. Results indicate that YOLOv7 yields the best text detection performance (up to 97.5% wmAP) and PARSEq excels in character recognition (85.2% CER). The data sets are made available.

Generic materializations for heightening of buildings and their effects on embodied carbon and costs

Targeting urban and sub-urban building archetypes, generic but practice-oriented materializations of basic construction elements (slabs and walls) and their combinations were conceived for heightening of existing buildings by two to four floors. The developed concepts considered numerous construction materials and requirements from architecture, structural engineering, building physics, and fire protection. Evaluations explored the effects on embodied carbon, weight and estimated cost per surface unit of market-oriented element combinations, to identify suitable (and inappropriate) materializations and to detect governing elements and materials. Globally, a heightening by four floors is better than by two, in terms of relative carbon and cost impacts, but some trade-offs in architectural floor plan layout may be required. Seeking cost reductions is generally disadvantageous for embodied carbon while an investment increase does not necessarily provide a reduced carbon footprint. Overall, timber construction results in the lowest embodied carbon (around 5 kg CO2, eq/m2·a) while being up to 10% more expensive than the cheapest and up to 15% heavier than the lightest materializations (which depend on the floor plan layout). Lightweight concrete construction can be the most economic materialization but is also up to 200% heavier than the lightest (which can possibly not be supported by the existing building), and results in up to 45% more embodied carbon than constructing with timber.

3D crowdsourced parametric cadastral mapping: Pathways integrating BIM/IFC, crowdsourced data and LADM

This paper is part of a doctoral dissertation (PhD) research aligned with global trends aiming to develop practical technical tools for the collection, management and visualization of three-dimensional (3D) property rights in urban areas. Recently, the Building Information Models (BIMs) claim a prominent position in the field of 3D cadastres and the digital twins of the cities. In addition to all other building information, the BIM can also provide data about the exact boundaries of all kind of legal spaces (e.g., property ownership rights as well as land use restrictions), under the support of Industry Foundation Classes (IFCs). However, the utilization of BIMs for 3D cadastral surveys is still accompanied with two main drawbacks. The first refers to BIMs limited availability, as till now they mainly encounter in new large constructions. The second refers to the complexity in defining the exact geometric location of the legal boundaries regarding the exterior/interior partitions of the construction and of the common spaces. The latter parameter is formed on the basis of the current legal legislation in each country. Through in-depth investigation of the current legal framework of each country it may be possible to visualize the various 3D legal spaces within complex constructions and thus to facilitate the integration of existing BIMs in the development of 3D cadasters that will enable a better understanding and communication of all involved parties in the operation of cross-boundary real estate markets. In parallel, crowdsourcing has already been proved to be a powerful data collection method for the initial participatory implementation of fast, reliable and affordable 3D cadastral surveys, utilizing all capabilities provided by the latest low-cost devices, mobile services (m-services), open-source software (OSS) and the international standard of Land Administration Domain Model (LADM ISO 19152). If no precise 3D building models are already available, the currently available 2D architectural plans combined with the additional geometric and descriptive cadastral information may be utilized for a participatory crowdsourced cadastral survey of the 3D property units. In this paper a ‘two-route’ crowdsourced approach is described. This approach suggests both the use of existing BIMs – those available – to proceed with 3D crowdsourced cadastral surveys of those constructions, as well as the use of 2D georeferenced basemaps (e.g., orthophotos for the compilation of 2D crowdsourced cadastral surveys, and all existing architectural floor plans of the constructions) to proceed with 3D crowdsourced cadastral surveys of all other constructions. A database schema describing the linkage between LADM standard, BIM/IFC and 3D crowdsourced geometric and descriptive cadastral information is developed and presented. A hybrid mobile application enabling the manipulation of BIM/IFC descriptive data – if existing; the collection of 3D crowdsourced geometric and descriptive information by property owners/users/non-professionals; the registration of the cadastral data and their relationships within a LADM-based cadastral geodatabase; the automated generation of 3D property unit models as block models (LoD1), using Model-driven approach; and the objects visualization in real-time, are developed. An investigation regarding the legally correct representation of the location of property unit boundaries, focusing mainly in the Greek territory is conducted. A practical experiment for each one of the cases of the ‘two-route’ crowdsourced approach is implemented, for two multi-storey buildings in the city of Athens, Greece. The potentials of the proposed crowdsourced solution as well as the achieved geometric accuracy – in the absence of BIM – are discussed and assessed. The results show that integrating BIM data with cadastral information derived from crowdsourcing, may significantly contribute to the implementation of 3D Cadastres, providing also a better visual understanding of 3D property rights. Nonetheless, even in the absence of a BIM the achieved accuracy seems to satisfy the cadastral specifications of the Greek cadaster enhancing the potential of exploiting crowdsourced data in the initial phases of the cadastral formal procedures.

The Archetype and Shifting Morphology of Sri Lankan Mosque Architecture

Sri Lanka is home to a number of historic mosques that date back over a thousand years and were built during various eras. In an attempt to fill the knowledge gap on such topic, particularly in regard to the built context of historic mosques in Sri Lanka, this paper illustrates the plan typology of mosques and compares them with the known Muslims architectural typologies. Is the design of the historical mosques in Sri Lanka related to a particular Islamic architectural style? Was the primary inquiry driving the study? The research's goal is to examine mosque plan arrangement typologies and determine how closely these typologies resemble well-known Muslims design elements. The approach includes providing examples of well-known Islamic architectural styles from around the globe and choosing historical mosques to graphically analyze and compare with Muslims architectural styles. A typological trend in Sri Lanka's historic mosques was identified through research. The discussion in this article emphasizes the similarity between the known Muslims architectural mosque typologies and the typology of historical mosques in Sri Lanka. As a consequence, the architectural floor plan typologies of Sri Lanka's historic mosques were identified. Numerous kinds of plans, including Ottoman, Umayyad, and Mughal were found to be effective. The results of the recent study showed that Sri Lanka's ancient mosques have different types of plans. This paper will discuss the history of the development of structures and the origins of their various components.

Semi-automatic Residential Floor Plan Detection: Developing a Tool for Humanities Research

Architectural floor plans are tangible cultural history artifacts, valuable for documenting how people live, work, and recreate, both in the present and going back hundreds or even thousands of years. Today, floor plans are created using digital technologies, but prior to the 1990s, the majority were drafted by hand. For archives, the storage, conservation, and handling of hand-drawn floor plans, which are often oversized and fragile, can be challenging. The digitization of floor plans, therefore, can reduce the possibility of damage while providing greater accessibility. However, even with digitization, researchers and archives continue to encounter barriers utilizing and cataloging these collections. Manually processing, collecting, and analyzing information contained in floor plans is time-consuming and prone to human error, particularly when working with large floor plan corpora. A tool to facilitate the broader use of archival floor plans is needed. Here, we propose the Building Database and Analytics System (BuDAS), a tool for the detection, storage, and analysis of floor plan information to assist scholars' research on collections and support metadata capture. In this article, we explore the challenges of detecting floor plan information and provide an overview of BuDAS, which we developed to address these problems. In addition, we test BuDAS's room detection system on two groups of floor plans sourced from archives, representing different levels of detection difficulty. The limitations and future implications of BuDAS and floor plan detection will also be discussed.

Deep learning-based text detection on architectural floor plan images

Architectural drawings are an important source of information for many construction-related tasks, as they contain geometric and semantic information about building parts. However, the manual extraction of room stamps and the insertion of such gleaned information into facility management systems is quite laborious and, thus, its automation is anticipated. In this paper, a method is proposed to detect and classify obscure or illegible text elements on legacy 2D architectural drawings of possibly poor quality. In contrast to existing approaches, a deep learning model is specifically trained for the task at hand rather than making use of transfer learning approaches. The resulting text snippets can be further processed with natural language processing tools to be fed into a facility management system automatically. Other conceivable applications include the extraction of drawing header information, material type or any additional text given on the drawing, to facilitate the enrichment of digital twins of existing structures with semantic data. To provide training data, two floor plan datasets are annotated in a consistent manner. The influence of different data augmentation techniques is investigated systematically. With regard to performance and efficiency, the presented method is compared to alternative tools for the task at hand.

MoTiS Parameters for Expressive Multi-Robot Systems: Relative Motion, Timing, and Spacing

Multi-robot systems are moving into human spaces, such as working with people in factories (Bacula et al., in: Companion of the 2020 ACM/IEEE international conference on human–robot interaction, pp 119–121, 2020) or in emergency support (Wagner in Front Robot AI 8, 2021; Baxter et al., in: Autonomous robots and agents, Springer, pp 9–16, 2007) and it is crucial to consider how robots can communicate with the humans in the space. Our work evaluates a parameter framework to allow multi-robot groups of x, y, theta robots to effectively communicate using expressive motion. While expressive motion has been extensively studied in single robots (Knight et al., in: 2016 IEEE international conference on intelligent robots and systems (IROS), IEEE, 2016; Bacula and LaViers in Int J Soc Robot, 1–16, 2020; Dragan et al., in: 2013 8th ACM/IEEE international conference on human–robot interaction (HRI), IEEE, pp 301–308, 2013; Kirby et al., in: The 18th IEEE international symposium on robot and human interactive communication, 2009, RO-MAN 2009, IEEE, pp 607–612, 2009), moving to multi-robots creates new challenges as the state space expands and becomes more complex. We evaluate a hierarchical framework of six parameters to generate multi-robot expressive motion consisting of: (1) relative direction, (2) coherence, (3) relative speed, (4) relative start time, (5) proximity, and (6) geometry. We conducted six independent online studies to explore each parameter, finding that four out of six of the parameters had significant impact on people’s perception of the multi-robot group. Additional takeaways of our studies clarify what humans interpret as a robot group, when the group is perceived positively versus negatively, and the critical role of architectural floor plan in interpreting robot intent.

An Assessment of the Relation between Architectural and Structural Systems in the Design of Tall Buildings in Turkey

Many professionals from several disciplines need to cooperate in designing and constructing tall buildings since their design and construction require more complex systems and technologies in terms of structure, installation, facade, vertical circulation and fire systems compared to low-rise buildings. The architects who design tall buildings have to know the architectural and structural design considerations of tall buildings and their interrelations well. This study is expected to reveal the status of tall buildings completed in Turkey and help designers understand architectural forms, floor plans, core planning and structural systems of tall buildings. For this purpose, the factors affecting architectural and structural design were examined, and the relation between the architectural–structural system was revealed for tall buildings completed in Turkey. In order to study architectural and structural considerations, 230 constructed tall buildings in Turkey were selected and analyzed in detail. According to the results of the study, it is seen that the prismatic form is widely used as the building tower form in Turkey, the shear-frame system as the structural system, the residential function as the building function, the single symmetrical central core as the building core, and the rectangular plan as the building floor plan alike.

FloorplanGAN: Vector residential floorplan adversarial generation

An architectural floorplan is a class of drawings that reflects the layout of rooms. The difference between a floorplan and a natural image and its dual features as both a vector graphic and a raster image makes it difficult to be generated by conventional deep neural generative models. We propose an adversarial generative framework that combines vector generation and raster discrimination for residential floorplan generation tasks. The floorplan is first generated in vector format with room areas as constraints and then discriminated in raster format visually using convolutional layers. A Differentiable Renderer connects the gap between the Vector Generator and Raster Discriminator. A self-attention mechanism is utilized to capture the interrelations of rooms in each floorplan. Experiments were conducted to demonstrate the feasibility of the proposed FloorplanGAN. In addition, we evaluated the effectiveness of generation based on diverse objective metrics and a user study. The code is available here: https://github.com/luozn15/FloorplanGAN.

Graph-based SLAM using architectural floor plans without loop closure

In simultaneous localization and mapping (SLAM), map distortions appear due to the accumulation of pose estimation errors. Loop closure of graph-based SLAM is known as one of the major solutions to this problem. However, there was still a problem that the errors remain in typical loop closure because it only uses constraints of the relative poses of nodes for error correction. In response, the authors focused on the fact that the environments in architectural floor plans are described accurately. Therefore we propose a method using constraints of the absolute poses based on floor plans. Specifically, the robot pose estimated on the floor plan is used as the absolute pose constraint. By optimizing the pose graph that includes absolute pose constraints, a map can be built more accurately to align with the floor plan. Although SLAM using floor plan has been studied before, we propose a new method to reduce computational cost while maintaining map accuracy without loop closure. In this study, we evaluated the effectiveness of our method by comparison with two other methods, one using loop closure and the other using both loop closure and absolute pose constraints. The results proved that the proposed method can maintain the map accuracy and reduce the computational cost.

A High-Precision Method for Segmentation and Recognition of Shopping Mall Plans.

Most studies on map segmentation and recognition are focused on architectural floor plans, while there are very few analyses of shopping mall plans. The objective of the work is to accurately segment and recognize the shopping mall plan, obtaining location and semantic information for each room via segmentation and recognition. This work can be used in other applications such as indoor robot navigation, building area and location analysis, and three-dimensional reconstruction. First, we identify and match the catalog of a mall floor plan to obtain matching text, and then we use the two-stage region growth method we proposed to segment the preprocessed floor plan. The room number is then obtained by sending each segmented room section to an OCR (optical character recognition) system for identification. Finally, the system retrieves the matching text to match the room number in order to obtain the room name, and outputs the needed room location and semantic information. It is considered a successful detection when a room region can be successfully segmented and identified. The proposed method is evaluated on a dataset including 1340 rooms. Experimental results show that the accuracy of room segmentation is 92.54%, and the accuracy of room recognition is 90.56%. The total detection accuracy is 83.81%.

An automated system for electrical power symbol placement in electrical plan drawing

An electrical plan drawing–sometimes called a wiring diagram or electrical drawing–consists of lines and symbols. Electrical plan drawings are prepared on 2D architectural floor plans using Computer-Aided Design and/or Drafting (CAD) programs. The placement/drawing of electrical power symbols–such as sockets, lights, and switches–is the first step of an electrical plan drawing. For this purpose, a smart system has been developed in this study to automatically draw/place electrical power symbols in appropriate locations. The system is based on the detection and classification/recognition of furnishing (decorative) symbols in the floor plans. We have created a furnishing symbol dataset drawing on dozens of architectural plan drawings that contain symbols of the most commonly used tools in floor plans, such as furniture, appliances, plumbing, doors, and windows. We used a Deep Convolutional Neural Network (D-CNN) with transfer learning–Inception-v3 model– to classify furnishing symbols. We tested the model on 20 real floor plans and achieved a very satisfactory accuracy of 97.05% in furnishing symbol classification. The symbol drawing step, which is the first step of drawing the electrical plan, was automated using the work developed, thus achieving the aim of saving time and labour. Experimental studies show the effectiveness of the proposed automated system.

Towards Robust Object Detection in Floor Plan Images: A Data Augmentation Approach

Object detection is one of the most critical tasks in the field of Computer vision. This task comprises identifying and localizing an object in the image. Architectural floor plans represent the layout of buildings and apartments. The floor plans consist of walls, windows, stairs, and other furniture objects. While recognizing floor plan objects is straightforward for humans, automatically processing floor plans and recognizing objects is challenging. In this work, we investigate the performance of the recently introduced Cascade Mask R-CNN network to solve object detection in floor plan images. Furthermore, we experimentally establish that deformable convolution works better than conventional convolutions in the proposed framework. Prior datasets for object detection in floor plan images are either publicly unavailable or contain few samples. We introduce SFPI, a novel synthetic floor plan dataset consisting of 10,000 images to address this issue. Our proposed method conveniently exceeds the previous state-of-the-art results on the SESYD dataset with an mAP of 98.1%. Moreover, it sets impressive baseline results on our novel SFPI dataset with an mAP of 99.8%. We believe that introducing the modern dataset enables the researcher to enhance the research in this domain.

Geometry-based symbol spotting in born-digital architectural floor plans

We describe a method for symbol spotting in born-digital architectural floor plans, which are a particularly challenging application domain of graphics recognition. We propose a hybrid method that capitalizes on strengths of both vector-based and pixel-based symbol spotting techniques. In the description phase, the salient geometric constituents of a symbol are extracted by a variety of vectorization techniques, including a proposed voting-based algorithm for finding partial ellipses. This enables us to better handle local shape irregularities and boundary discontinuities, as well as partial occlusion and overlap. In the matching phase, the spatial relationship between the geometric primitives is encoded via a primitive-aware proximity graph. A statistical approach is then used to rapidly yield a coarse localization of symbols within the plan. Localization is further refined with a pixel-based step implementing a modified cross-correlation function. Experimental results on the public SESYD synthetic dataset demonstrate that our approach clearly outperforms other popular symbol spotting approaches. Moreover, our approach yields promising results on real-world images, which are significantly more challenging in terms of overlap, occlusion, and presence of non-symbolic graphical data.

How Many Bedrooms Do You Need? A Real-Estate Recommender System from Architectural Floor Plan Images

This paper introduces an automated image processing method to analyze an architectural floor plan database. The floor plan information, such as the measurement of the rooms, dimension lines, and even the location of each room, can be automatically produced. This assists the real-estate agents to maximise the chances of the closure of deals by providing explicit insights to the prospective purchasers. With a clear idea about the layout of the place, customers can quickly make an analytical decision. Besides, it reduces the specialized training cost and increases the efficiency in business actions by understanding the property types with the greatest demand. Succinctly, this paper utilizes both the traditional image processing and convolutional neural networks (CNNs) to detect the bedrooms by undergoing the segmentation and classification processes. A thorough experiment, analysis, and evaluation had been performed to verify the effectiveness of the proposed framework. As a result, a three-class bedroom classification accuracy of ∼ 90% was achieved when validating on more than 500 image samples that consist of the different room numbers. In addition, qualitative findings were presented to manifest visually the feasibility of the algorithm developed.

An RGB-D Camera Based Visual Positioning System for Assistive Navigation by a Robotic Navigation Aid

There are about 253 million people with visual impairment worldwide. Many of them use a white cane and/or a guide dog as the mobility tool for daily travel. Despite decades of efforts, electronic navigation aid that can replace white cane is still research in progress. In this paper, we propose an RGB-D camera based visual positioning system (VPS) for real-time localization of a robotic navigation aid (RNA) in an architectural floor plan for assistive navigation. The core of the system is the combination of a new 6-DOF depth-enhanced visual-inertial odometry (DVIO) method and a particle filter localization (PFL) method. DVIO estimates RNA's pose by using the data from an RGB-D camera and an inertial measurement unit (IMU). It extracts the floor plane from the camera's depth data and tightly couples the floor plane, the visual features (with and without depth data), and the IMU's inertial data in a graph optimization framework to estimate the device's 6-DOF pose. Due to the use of the floor plane and depth data from the RGB-D camera, DVIO has a better pose estimation accuracy than the conventional VIO method. To reduce the accumulated pose error of DVIO for navigation in a large indoor space, we developed the PFL method to locate RNA in the floor plan. PFL leverages geometric information of the architectural CAD drawing of an indoor space to further reduce the error of the DVIO-estimated pose. Based on VPS, an assistive navigation system is developed for the RNA prototype to assist a visually impaired person in navigating a large indoor space. Experimental results demonstrate that: 1) DVIO method achieves better pose estimation accuracy than the state-of-the-art VIO method and performs real-time pose estimation (18 Hz pose update rate) on a UP Board computer; 2) PFL reduces the DVIO-accrued pose error by 82.5% on average and allows for accurate wayfinding (endpoint position error ≤ 45 cm) in large indoor spaces.

IMPROVING PUBLIC AWARENESS OF EARTHQUAKE-RESISTANT HOUSE USING FACTOR ANALYSIS

Public awareness of earthquake-resistant houses can increase the region's resilience to earthquake disasters. This study aims to conduct a comparison of two different groups of people on their opinion of earthquake-resistant housing variables and actions to increase the quality of their house. For supporting public awareness of earthquake-resistant houses, the purpose also determines the latent variables of earthquake-resistant houses and actions from the public for improving the quality of the residential building. This quantitative research begins by collecting the earthquake-resistant house attributes through journal reviews. This study used a closed-ended questionnaire to collect data on public experiences of earthquake disasters. For data analysis, it used correlation analysis to determine the level of relationship between two variables and utilizes factor analysis to reduce the variable to a small number of factors. The two groups that have different experiences of earthquakes tend to have similar opinions about the characteristics of earthquake-resistant houses and the actions for improving the quality of residential buildings. The five factors of the earthquake-resistant housing are safe structural construction, fixed architectural elements, open accessibility, lightweight-ductile material, and simple floor plan. Then the three factors represent the actions for improving residential quality, namely, to improve material and structure, build an earthquake-resistant house, and make accessibility easy and safe.

A transformation algorithm to construct a rectangular floorplan

A rectangular floorplan (RFP) is a partition of a rectangle R into rectangles R1,R2,…,Rn such that no four of them meet at a point. Rectangular floorplans find their applications in VLSI circuits floorplanning and architectural floorplanning. Let G represent the graph of a rectangular floorplan RFP1(n) and H be a subgraph of G, where n denotes the number of rectangles in RFP1(n). It is well known that every subgraph of G may not admit an RFP, i.e., G does not have a hereditary property. Hence, in this paper, we first derive a necessary and sufficient condition for H to admit a rectangular floorplan RFP2(n). Further, if H admits an RFP, we present a linear time transformation algorithm for deriving RFP2(n) from RFP1(n).