This document presents a 3-phase approach to automatically generate a pseudo-3D building model from a 2D floor plan image. The 3 phases are:
1. Image processing phase which extracts architectural symbols like walls, windows and doors from the image using techniques like edge detection, line extraction and region growing.
2. Save module phase which stores the extracted symbol information like positions, dimensions in an XML file.
3. Interactive 3D module phase which generates a 3D model of the building from the XML file and allows navigation and viewing of the model.
The approach was implemented using libraries like OpenCV for image processing and Irrlicht engine for 3D generation. Testing on sample floor plans showed
Design and development of DrawBot using image processing IJECEIAES
Extracting text from an image and reproducing them can often be a laborious task. We took it upon ourselves to solve the problem. Our work is aimed at designing a robot which can perceive an image shown to it and reproduce it on any given area as directed. It does so by first taking an input image and performing image processing operations on the image to improve its readability. Then the text in the image is recognized by the program. Points for each letter are taken, then inverse kinematics is done for each point with MATLAB/Simulink and the angles in which the servo motors should be moved are found out and stored in the Arduino. Using these angles, the control algorithm is generated in the Arduino and the letters are drawn.
Characterization of a 2D Geometry Using C++ Interfacedrboon
Problem statement: Usually, we want to represent the final computer aided drawings in a detailed characterization of the design geometry. In some cases, the final design is corrupted with additional geometry parameters, which are not part of the problem, and lack of dimensions. We need an automatic approach to resolve this issue. Approach: Here, we develop a toolkit, which integrates SIEMENS NX8 and C++, to improve the characterization process for 2D geometries. Results: We applied the tool to several different 2D cross-sections and proved that we were able to removal the unwanted parts from the geometry and apply the proper dimensions and constraints to the geometry. Conclusion: This makes the 2D geometry characterization process faster and user-friendly.
Development Prototype Design of Virtual Assembly Application-Based Leap MotionIJAEMSJORNAL
Innovation in design engineering practice is very important in the world of manufacturing in the increasingly competitive global market. Prototyping and evaluation measures are inseparable from the design process in the manufacture of a product. And made one of many physical prototypes require very expensive and time consuming, so the technology of Virtual Reality (VR) is needed, so the industry can quickly and precisely in the decision. VR technology combines a human being with a computer environment visually, touch and hearing, so that the user as if into the virtual world. The goal is that users with hand movements can interact with what is displayed on the computer screen or the user can interact with the environment is unreal to be added into the real world. VR is required for simulations that require a lot of interaction such as prototype assembly methods, or better known as the Virtual Assembly. Virtual Assembly concept which was developed as the ability to assemble a real representation of the physical model, the 3D models in CAD software by simulating the natural movement of the human hand. Leap Motion (accuracy of 0.01mm) was used to replace Microsoft's Kinect (accuracy of 1.5cm) and Motion Glove with flex sensors (accuracy of 1°) in several previous research. Leap mot ion controller is a device that captures every movement of the hand to then be processed and integrated with 3D models in CAD software. And simulation of assembly process virtually in CA D software with hand gestures detected by the leap mot ion, assembly parts can be driven either in translation or rotation, zooming and adding the assembly constraint. It also can perform mouse functions (such as left-click, middle-click, right-click and move the mouse cursor position) to a virtual assembly process simulation on CAD software.
The slide presents introduction to computer aided design and drafting, and introduction to AutoCAD software. 2D and 3D modelling using AutoCAD software is presented.
Design and development of DrawBot using image processing IJECEIAES
Extracting text from an image and reproducing them can often be a laborious task. We took it upon ourselves to solve the problem. Our work is aimed at designing a robot which can perceive an image shown to it and reproduce it on any given area as directed. It does so by first taking an input image and performing image processing operations on the image to improve its readability. Then the text in the image is recognized by the program. Points for each letter are taken, then inverse kinematics is done for each point with MATLAB/Simulink and the angles in which the servo motors should be moved are found out and stored in the Arduino. Using these angles, the control algorithm is generated in the Arduino and the letters are drawn.
Characterization of a 2D Geometry Using C++ Interfacedrboon
Problem statement: Usually, we want to represent the final computer aided drawings in a detailed characterization of the design geometry. In some cases, the final design is corrupted with additional geometry parameters, which are not part of the problem, and lack of dimensions. We need an automatic approach to resolve this issue. Approach: Here, we develop a toolkit, which integrates SIEMENS NX8 and C++, to improve the characterization process for 2D geometries. Results: We applied the tool to several different 2D cross-sections and proved that we were able to removal the unwanted parts from the geometry and apply the proper dimensions and constraints to the geometry. Conclusion: This makes the 2D geometry characterization process faster and user-friendly.
Development Prototype Design of Virtual Assembly Application-Based Leap MotionIJAEMSJORNAL
Innovation in design engineering practice is very important in the world of manufacturing in the increasingly competitive global market. Prototyping and evaluation measures are inseparable from the design process in the manufacture of a product. And made one of many physical prototypes require very expensive and time consuming, so the technology of Virtual Reality (VR) is needed, so the industry can quickly and precisely in the decision. VR technology combines a human being with a computer environment visually, touch and hearing, so that the user as if into the virtual world. The goal is that users with hand movements can interact with what is displayed on the computer screen or the user can interact with the environment is unreal to be added into the real world. VR is required for simulations that require a lot of interaction such as prototype assembly methods, or better known as the Virtual Assembly. Virtual Assembly concept which was developed as the ability to assemble a real representation of the physical model, the 3D models in CAD software by simulating the natural movement of the human hand. Leap Motion (accuracy of 0.01mm) was used to replace Microsoft's Kinect (accuracy of 1.5cm) and Motion Glove with flex sensors (accuracy of 1°) in several previous research. Leap mot ion controller is a device that captures every movement of the hand to then be processed and integrated with 3D models in CAD software. And simulation of assembly process virtually in CA D software with hand gestures detected by the leap mot ion, assembly parts can be driven either in translation or rotation, zooming and adding the assembly constraint. It also can perform mouse functions (such as left-click, middle-click, right-click and move the mouse cursor position) to a virtual assembly process simulation on CAD software.
The slide presents introduction to computer aided design and drafting, and introduction to AutoCAD software. 2D and 3D modelling using AutoCAD software is presented.
Implementation of a Simulation Model Using the Systems Dynamics: Case Study f...IJERA Editor
With computer technology reaching the construction industry, much has changed in the way that these are
idealized and designed. Currently, with the increasing availability of tools for the development of the models and
parametric models intended for digital manufacturing, the insertion of these characteristics as design tool allows
the designer to test solutions on various factors that permeate the design process, especially as regards the
constructive points. The experience of the parameterisation dynamics and the dynamics of digital manufacturing
from the design of the project allows the development of complex geometries, since the control their properties
the analyses of structural variables of environmental comfort and aesthetics. The growth of the complexity of the
projects and the capacity of computing resources, has arisen the need to use a more systemic approach, as well as
have emerged the simulation programs based on dynamic systems, a digital simulation methodology in order to
understand complex forms, which is part of the concept of systemic thought for the resolution of problems. The
objective is in this Article is to identify and analyse the aspects of potential simulation based on dynamic
systems and demonstrate a practical case drawn up in Dynamo software.
Covenant University location navigation system (CU Naviloc) built on the Android
platform is a mobile based system which can be easily accessed from an android
mobile device across any location, downloadable on Google play store and
deployable anywhere in the world. In this study,Computer-aided design as well as
building Information modelling (CAD/BIM) system applications were introduced to
the simulation of Covenant University, a smart and compact campus whose edifice
elaborates the need for the development of a three dimensional (3D) virtual model of
locations as an advancement over the existing 2D model representation of maps on
the Google map and Google Earth platform. Therefore, this research work outlays the
development of three dimensional models (CU NAViLoc Model) for details of
information at Covenant University as well as the requirements for its development
and implementation for cooperate representation of the institution. A three
dimensional map of CU was packaged into a location navigation system and was
implemented using Unreal Engine, Trimble SketchUp and Revit to effect a user
friendly, smarter and multi-dimensional viewable user interface. CU Naviloc explore
the advantage of the rich benefits of building information models and geographic
information systems to build a free-roam navigator, based on a well-updated model of
Covenant University location map. The benefits of this system is not limited to 3D
capability, real time similarity, portability, no downtime on access, and progressive
scalability with zero or little loss of data. The usage of the CU Naviloc system
requires free download and installation of the app and internet presence.
Wireless network implementation is a viable option for building network infrastructure in rural communities. Rural people lack network infrastructures for information services and socio-economic development. The aim of this study was to develop a wireless network infrastructure architecture for network services to rural dwellers. A user-centered approach was applied in the study and a wireless network infrastructure was designed and deployed to cover five rural locations. Data was collected and analyzed to assess the performance of the network facilities. The results shows that the system had been performing adequately without any downtime with an average of 200 users per month and the quality of service has remained high. The transmit/receive rate of 300Mbps was thrice as fast as the normal Ethernet transmit/receive specification with an average throughput of 1 Mbps. The multiple output/multiple input (MIMO) point-to-multipoint network design increased the network throughput and the quality of service experienced by the users.
3D reconstruction is a technique used in computer vision which has a wide range of applications in areas like object recognition, city modelling, virtual reality, physical simulations, video games and special effects. Previously, to perform a 3D reconstruction, specialized hardwares were required. Such systems were often very expensive and was only available for industrial or research purpose. With the rise of the availability of high-quality low cost 3D sensors, it is now possible to design inexpensive complete 3D scanning systems. The objective of this work was to design an acquisition and processing system that can perform 3D scanning and reconstruction of objects seamlessly. In addition, the goal of this work also included making the 3D scanning process fully automated by building and integrating a turntable alongside the software. This means the user can perform a full 3D scan only by a press of a few buttons from our dedicated graphical user interface. Three main steps were followed to go from acquisition of point clouds to the finished reconstructed 3D model. First, our system acquires point cloud data of a person/object using inexpensive camera sensor. Second, align and convert the acquired point cloud data into a watertight mesh of good quality. Third, export the reconstructed model to a 3D printer to obtain a proper 3D print of the model.
3D reconstruction is a technique used in computer vision which has a wide range of applications in areas like object recognition, city modelling, virtual reality, physical simulations, video games and special effects. Previously, to perform a 3D reconstruction, specialized hardwares were required. Such systems were often very expensive and was only available for industrial or research purpose. With the rise of the availability of high-quality low cost 3D sensors, it is now possible to design inexpensive complete 3D scanning systems. The objective of this work was to design an acquisition and processing system that can perform 3D scanning and reconstruction of objects seamlessly. In addition, the goal of this work also included making the 3D scanning process fully automated by building and integrating a turntable alongside the software. This means the user can perform a full 3D scan only by a press of a few buttons from our dedicated graphical user interface. Three main steps were followed to go from acquisition of point clouds to the finished reconstructed 3D model. First, our system acquires point cloud data of a person/object using inexpensive camera sensor. Second, align and convert the acquired point cloud data into a watertight mesh of good quality. Third, export the reconstructed model to a 3D printer to obtain a proper 3D print of the model.
COMPLETE END-TO-END LOW COST SOLUTION TO A 3D SCANNING SYSTEM WITH INTEGRATED...ijcsit
3D reconstruction is a technique used in computer vision which has a wide range of applications in
areas like object recognition, city modelling, virtual reality, physical simulations, video games and
special effects. Previously, to perform a 3D reconstruction, specialized hardwares were required.
Such systems were often very expensive and was only available for industrial or research purpose.
With the rise of the availability of high-quality low cost 3D sensors, it is now possible to design
inexpensive complete 3D scanning systems. The objective of this work was to design an acquisition and
processing system that can perform 3D scanning and reconstruction of objects seamlessly. In addition,
the goal of this work also included making the 3D scanning process fully automated by building and
integrating a turntable alongside the software. This means the user can perform a full 3D scan only by
a press of a few buttons from our dedicated graphical user interface. Three main steps were followed
to go from acquisition of point clouds to the finished reconstructed 3D model. First, our system
acquires point cloud data of a person/object using inexpensive camera sensor. Second, align and
convert the acquired point cloud data into a watertight mesh of good quality. Third, export the
reconstructed model to a 3D printer to obtain a proper 3D print of the model.
This presentation outlines our vision on the D-Reader, the product our company develops. The product is focused on automated understanding of engineering drawings.
Design patterns refer to reusable program solutions that aim to solve similar design problem during development process. The good practise of using patterns is encouraged and efforts have to be taken to knowledge on design patterns. We use a decorator pattern to improving a web application. Web application is implemented using HTML language, ASP.NET, and C#.
Intelligent Real-time 3D Configuration Platform for Customizing E-commerce Pr...ijcga
In this paper is designed and realized a 3D web configurator platform able to optimize the creation of 3D objects for sale on multi e-commerce platform. The user can select online the preferred object by learning the configurator to optimize the 3D object features. A Support Vector Machine -SVM- algorithm adopting XGBoost approach, improved affect computing, user navigation experience and Visual Merchandising – VM- approach, all elements which support the creation of new object having the best effective features. The intelligent self-learning platform is designed for the real time construction of 3D CAD objects and integrates different user facilities such as augmented reality, web plugin, and WebGL graphical libraries. The proposed work follows the research and development -R&D- guidelines of ‘Frascati’ manual concerning new search engine as research topic and summarize the result of an industry research project.
INTELLIGENT REAL-TIME 3D CONFIGURATION PLATFORM FOR CUSTOMIZING E-COMMERCE PR...ijcga
In this paper is designed and realized a 3D web configurator platform able to optimize the creation of 3D objects for sale on multi e-commerce platform. The user can select online the preferred object by learning the configurator to optimize the 3D object features. A Support Vector Machine -SVM- algorithm adopting XGBoost approach, improved affect computing, user navigation experience and Visual Merchandising – VM- approach, all elements which support the creation of new object having the best effective features. The intelligent self-learning platform is designed for the real time construction of 3D CAD objects and integrates different user facilities such as augmented reality, web plugin, and WebGL graphical libraries. The proposed work follows the research and development -R&D- guidelines of ‘Frascati’ manual concerning new search engine as research topic and summarize the result of an industry research project.
A L -H ARM E XPANSION M OVIE B ASED ON V IRTUAL R EALITYcsandit
Animated movies are excellent virtual environments
for creating models in high quality.
Animated movies can include 3D models, sounds and l
ights effects, and detailed maps. In this
paper, a virtual reality movie is applied to Al-Har
am Expansion stages including the future
stage of expansion. 3DMAX program is used to rich t
he maximum benefits of using 3D
modeling. Maps with details are built by using ARCG
IS program in order to understand the real
difference between the three different expansions s
tages clearly and effectively. A novel
technique is presented in order to insert 2D maps a
nd other details in the 3D model built by
3DMAX
IMAGE RECOGNITION USING MATLAB SIMULINK BLOCKSETIJCSEA Journal
The world over, image recognition are essential players in promoting quality object recognition especially in emergency and search-rescue operation. In this paper precise image recognition system using Matlab Simulink Blockset to detect selected object from crowd is presented. The process involves extracting object
features and then recognizes it considering illumination, direction and pose. A Simulink model has been developed to eliminate the tiny elements from the image, then creating segments for precise object recognition. Furthermore, the simulation explores image recognition from the coloured and gray-scale images through image processing techniques in Simulink environment. The tool employed for computation
and simulation is the Matlab image processing blockset. The process comprises morphological operation method which is effective for captured images and video. The results of extensive simulations indicate that this method is suitable for application identifying a person from a crow. The model can be used in emergency and search-rescue operation as well as in medicine, information security, access control, law enforcement, surveillance system, microscopy etc.
Implementation of a Simulation Model Using the Systems Dynamics: Case Study f...IJERA Editor
With computer technology reaching the construction industry, much has changed in the way that these are
idealized and designed. Currently, with the increasing availability of tools for the development of the models and
parametric models intended for digital manufacturing, the insertion of these characteristics as design tool allows
the designer to test solutions on various factors that permeate the design process, especially as regards the
constructive points. The experience of the parameterisation dynamics and the dynamics of digital manufacturing
from the design of the project allows the development of complex geometries, since the control their properties
the analyses of structural variables of environmental comfort and aesthetics. The growth of the complexity of the
projects and the capacity of computing resources, has arisen the need to use a more systemic approach, as well as
have emerged the simulation programs based on dynamic systems, a digital simulation methodology in order to
understand complex forms, which is part of the concept of systemic thought for the resolution of problems. The
objective is in this Article is to identify and analyse the aspects of potential simulation based on dynamic
systems and demonstrate a practical case drawn up in Dynamo software.
Covenant University location navigation system (CU Naviloc) built on the Android
platform is a mobile based system which can be easily accessed from an android
mobile device across any location, downloadable on Google play store and
deployable anywhere in the world. In this study,Computer-aided design as well as
building Information modelling (CAD/BIM) system applications were introduced to
the simulation of Covenant University, a smart and compact campus whose edifice
elaborates the need for the development of a three dimensional (3D) virtual model of
locations as an advancement over the existing 2D model representation of maps on
the Google map and Google Earth platform. Therefore, this research work outlays the
development of three dimensional models (CU NAViLoc Model) for details of
information at Covenant University as well as the requirements for its development
and implementation for cooperate representation of the institution. A three
dimensional map of CU was packaged into a location navigation system and was
implemented using Unreal Engine, Trimble SketchUp and Revit to effect a user
friendly, smarter and multi-dimensional viewable user interface. CU Naviloc explore
the advantage of the rich benefits of building information models and geographic
information systems to build a free-roam navigator, based on a well-updated model of
Covenant University location map. The benefits of this system is not limited to 3D
capability, real time similarity, portability, no downtime on access, and progressive
scalability with zero or little loss of data. The usage of the CU Naviloc system
requires free download and installation of the app and internet presence.
Wireless network implementation is a viable option for building network infrastructure in rural communities. Rural people lack network infrastructures for information services and socio-economic development. The aim of this study was to develop a wireless network infrastructure architecture for network services to rural dwellers. A user-centered approach was applied in the study and a wireless network infrastructure was designed and deployed to cover five rural locations. Data was collected and analyzed to assess the performance of the network facilities. The results shows that the system had been performing adequately without any downtime with an average of 200 users per month and the quality of service has remained high. The transmit/receive rate of 300Mbps was thrice as fast as the normal Ethernet transmit/receive specification with an average throughput of 1 Mbps. The multiple output/multiple input (MIMO) point-to-multipoint network design increased the network throughput and the quality of service experienced by the users.
3D reconstruction is a technique used in computer vision which has a wide range of applications in areas like object recognition, city modelling, virtual reality, physical simulations, video games and special effects. Previously, to perform a 3D reconstruction, specialized hardwares were required. Such systems were often very expensive and was only available for industrial or research purpose. With the rise of the availability of high-quality low cost 3D sensors, it is now possible to design inexpensive complete 3D scanning systems. The objective of this work was to design an acquisition and processing system that can perform 3D scanning and reconstruction of objects seamlessly. In addition, the goal of this work also included making the 3D scanning process fully automated by building and integrating a turntable alongside the software. This means the user can perform a full 3D scan only by a press of a few buttons from our dedicated graphical user interface. Three main steps were followed to go from acquisition of point clouds to the finished reconstructed 3D model. First, our system acquires point cloud data of a person/object using inexpensive camera sensor. Second, align and convert the acquired point cloud data into a watertight mesh of good quality. Third, export the reconstructed model to a 3D printer to obtain a proper 3D print of the model.
3D reconstruction is a technique used in computer vision which has a wide range of applications in areas like object recognition, city modelling, virtual reality, physical simulations, video games and special effects. Previously, to perform a 3D reconstruction, specialized hardwares were required. Such systems were often very expensive and was only available for industrial or research purpose. With the rise of the availability of high-quality low cost 3D sensors, it is now possible to design inexpensive complete 3D scanning systems. The objective of this work was to design an acquisition and processing system that can perform 3D scanning and reconstruction of objects seamlessly. In addition, the goal of this work also included making the 3D scanning process fully automated by building and integrating a turntable alongside the software. This means the user can perform a full 3D scan only by a press of a few buttons from our dedicated graphical user interface. Three main steps were followed to go from acquisition of point clouds to the finished reconstructed 3D model. First, our system acquires point cloud data of a person/object using inexpensive camera sensor. Second, align and convert the acquired point cloud data into a watertight mesh of good quality. Third, export the reconstructed model to a 3D printer to obtain a proper 3D print of the model.
COMPLETE END-TO-END LOW COST SOLUTION TO A 3D SCANNING SYSTEM WITH INTEGRATED...ijcsit
3D reconstruction is a technique used in computer vision which has a wide range of applications in
areas like object recognition, city modelling, virtual reality, physical simulations, video games and
special effects. Previously, to perform a 3D reconstruction, specialized hardwares were required.
Such systems were often very expensive and was only available for industrial or research purpose.
With the rise of the availability of high-quality low cost 3D sensors, it is now possible to design
inexpensive complete 3D scanning systems. The objective of this work was to design an acquisition and
processing system that can perform 3D scanning and reconstruction of objects seamlessly. In addition,
the goal of this work also included making the 3D scanning process fully automated by building and
integrating a turntable alongside the software. This means the user can perform a full 3D scan only by
a press of a few buttons from our dedicated graphical user interface. Three main steps were followed
to go from acquisition of point clouds to the finished reconstructed 3D model. First, our system
acquires point cloud data of a person/object using inexpensive camera sensor. Second, align and
convert the acquired point cloud data into a watertight mesh of good quality. Third, export the
reconstructed model to a 3D printer to obtain a proper 3D print of the model.
This presentation outlines our vision on the D-Reader, the product our company develops. The product is focused on automated understanding of engineering drawings.
Design patterns refer to reusable program solutions that aim to solve similar design problem during development process. The good practise of using patterns is encouraged and efforts have to be taken to knowledge on design patterns. We use a decorator pattern to improving a web application. Web application is implemented using HTML language, ASP.NET, and C#.
Intelligent Real-time 3D Configuration Platform for Customizing E-commerce Pr...ijcga
In this paper is designed and realized a 3D web configurator platform able to optimize the creation of 3D objects for sale on multi e-commerce platform. The user can select online the preferred object by learning the configurator to optimize the 3D object features. A Support Vector Machine -SVM- algorithm adopting XGBoost approach, improved affect computing, user navigation experience and Visual Merchandising – VM- approach, all elements which support the creation of new object having the best effective features. The intelligent self-learning platform is designed for the real time construction of 3D CAD objects and integrates different user facilities such as augmented reality, web plugin, and WebGL graphical libraries. The proposed work follows the research and development -R&D- guidelines of ‘Frascati’ manual concerning new search engine as research topic and summarize the result of an industry research project.
INTELLIGENT REAL-TIME 3D CONFIGURATION PLATFORM FOR CUSTOMIZING E-COMMERCE PR...ijcga
In this paper is designed and realized a 3D web configurator platform able to optimize the creation of 3D objects for sale on multi e-commerce platform. The user can select online the preferred object by learning the configurator to optimize the 3D object features. A Support Vector Machine -SVM- algorithm adopting XGBoost approach, improved affect computing, user navigation experience and Visual Merchandising – VM- approach, all elements which support the creation of new object having the best effective features. The intelligent self-learning platform is designed for the real time construction of 3D CAD objects and integrates different user facilities such as augmented reality, web plugin, and WebGL graphical libraries. The proposed work follows the research and development -R&D- guidelines of ‘Frascati’ manual concerning new search engine as research topic and summarize the result of an industry research project.
A L -H ARM E XPANSION M OVIE B ASED ON V IRTUAL R EALITYcsandit
Animated movies are excellent virtual environments
for creating models in high quality.
Animated movies can include 3D models, sounds and l
ights effects, and detailed maps. In this
paper, a virtual reality movie is applied to Al-Har
am Expansion stages including the future
stage of expansion. 3DMAX program is used to rich t
he maximum benefits of using 3D
modeling. Maps with details are built by using ARCG
IS program in order to understand the real
difference between the three different expansions s
tages clearly and effectively. A novel
technique is presented in order to insert 2D maps a
nd other details in the 3D model built by
3DMAX
IMAGE RECOGNITION USING MATLAB SIMULINK BLOCKSETIJCSEA Journal
The world over, image recognition are essential players in promoting quality object recognition especially in emergency and search-rescue operation. In this paper precise image recognition system using Matlab Simulink Blockset to detect selected object from crowd is presented. The process involves extracting object
features and then recognizes it considering illumination, direction and pose. A Simulink model has been developed to eliminate the tiny elements from the image, then creating segments for precise object recognition. Furthermore, the simulation explores image recognition from the coloured and gray-scale images through image processing techniques in Simulink environment. The tool employed for computation
and simulation is the Matlab image processing blockset. The process comprises morphological operation method which is effective for captured images and video. The results of extensive simulations indicate that this method is suitable for application identifying a person from a crow. The model can be used in emergency and search-rescue operation as well as in medicine, information security, access control, law enforcement, surveillance system, microscopy etc.
Hybrid optimization of pumped hydro system and solar- Engr. Abdul-Azeez.pdffxintegritypublishin
Advancements in technology unveil a myriad of electrical and electronic breakthroughs geared towards efficiently harnessing limited resources to meet human energy demands. The optimization of hybrid solar PV panels and pumped hydro energy supply systems plays a pivotal role in utilizing natural resources effectively. This initiative not only benefits humanity but also fosters environmental sustainability. The study investigated the design optimization of these hybrid systems, focusing on understanding solar radiation patterns, identifying geographical influences on solar radiation, formulating a mathematical model for system optimization, and determining the optimal configuration of PV panels and pumped hydro storage. Through a comparative analysis approach and eight weeks of data collection, the study addressed key research questions related to solar radiation patterns and optimal system design. The findings highlighted regions with heightened solar radiation levels, showcasing substantial potential for power generation and emphasizing the system's efficiency. Optimizing system design significantly boosted power generation, promoted renewable energy utilization, and enhanced energy storage capacity. The study underscored the benefits of optimizing hybrid solar PV panels and pumped hydro energy supply systems for sustainable energy usage. Optimizing the design of solar PV panels and pumped hydro energy supply systems as examined across diverse climatic conditions in a developing country, not only enhances power generation but also improves the integration of renewable energy sources and boosts energy storage capacities, particularly beneficial for less economically prosperous regions. Additionally, the study provides valuable insights for advancing energy research in economically viable areas. Recommendations included conducting site-specific assessments, utilizing advanced modeling tools, implementing regular maintenance protocols, and enhancing communication among system components.
Sachpazis:Terzaghi Bearing Capacity Estimation in simple terms with Calculati...Dr.Costas Sachpazis
Terzaghi's soil bearing capacity theory, developed by Karl Terzaghi, is a fundamental principle in geotechnical engineering used to determine the bearing capacity of shallow foundations. This theory provides a method to calculate the ultimate bearing capacity of soil, which is the maximum load per unit area that the soil can support without undergoing shear failure. The Calculation HTML Code included.
Cosmetic shop management system project report.pdfKamal Acharya
Buying new cosmetic products is difficult. It can even be scary for those who have sensitive skin and are prone to skin trouble. The information needed to alleviate this problem is on the back of each product, but it's thought to interpret those ingredient lists unless you have a background in chemistry.
Instead of buying and hoping for the best, we can use data science to help us predict which products may be good fits for us. It includes various function programs to do the above mentioned tasks.
Data file handling has been effectively used in the program.
The automated cosmetic shop management system should deal with the automation of general workflow and administration process of the shop. The main processes of the system focus on customer's request where the system is able to search the most appropriate products and deliver it to the customers. It should help the employees to quickly identify the list of cosmetic product that have reached the minimum quantity and also keep a track of expired date for each cosmetic product. It should help the employees to find the rack number in which the product is placed.It is also Faster and more efficient way.
Planning Of Procurement o different goods and services
3-Phase Recognition Approach to Pseudo 3D Building Generation from 2D Floor Plan
1. International Journal of Computer Graphics & Animation (IJCGA) Vol.1, No.2, June 2011
DOI : 10.5121/ijcga.2011.1202 13
3-PHASE RECOGNITION APPROACH TO PSEUDO 3D
BUILDING GENERATION FROM 2D FLOOR PLAN
Raj Kishen Moloo
Computer Science and Engineering Department,
University of Mauritius, Reduit, Mauritius
r.moloo@uom.ac.mu
Muhammad Ajmal Sheik Dawood
MSc Student, University of Essex, Colchester, Essex, UK
ajmalsd@yahoo.com
Abu Salmaan Auleear
Senior Web-Designer, Chesteroc ltd (Mundocom Mauritius), Ebène, Mauritius
prince_05@live.com
ABSTRACT
Nowadays three dimension (3D) architectural visualisation has become a powerful tool in the
conceptualisation, design and presentation of architectural products in the construction industry, providing
realistic interaction and walkthrough on engineering products. Traditional ways of implementing 3D
models involves the use of specialised 3D authoring tools along with skilled 3D designers with blueprints of
the model and this is a slow and laborious process. The aim of this paper is to automate this process by
simply analyzing the blueprint document and generating the 3D scene automatically. For this purpose we
have devised a 3-Phase recognition approach to pseudo 3D building generation from 2D floor plan and
developed a software accordingly.
Our 3-phased 3D building system has been implemented using C, C++ and OpenCV library [24] for the
Image Processing module; The Save Module generated an XML file for storing the processed floor plan
objects attributes; while the Irrlitch [14] game engine was used to implement the Interactive 3D module.
Though still at its infancy, our proposed system gave commendable results. We tested our system on 6 floor
plans with complexities ranging from low to high and the results seems to be very promising with an
average processing time of around 3s and a 3D generation in 4s. In addition the system provides an
interactive walk-though and allows users to modify components.
KEYWORDS
Image processing, analysis, recognition, 3D, Architectural plan, Virtual Reality, Object Extraction.
2. International Journal of Computer Graphics & Animation (IJCGA) Vol.1, No.2, June 2011
14
1. INTRODUCTION
Enhanced graphic processing capabilities of today’s computers have enabled the widespread of
three dimension (3D) visualisation in the construction industry, providing realistic interaction and
walkthrough on engineering products. 3D architectural visualization has become a powerful tool
in the conceptualisation, design and presentation of architectural products [8]. Long gone are the
days of mere 2D plans. Several 3D authoring packages like Google Sketchup [10], 3DS Max
[11], Maya [12], Blender [13] can be used to model 3D buildings based on the 2D plan.
Nevertheless, creating 3D buildings requires skilled and trained people, besides being a lengthy,
cumbersome and tedious task [8].
Nowadays, popular architectural package like AutoCad [9] allows saving of 2D plans digitally, in
particular formats, e.g., DWG/ DWF format, out of which enhanced design and visualization
workflows on 3D buildings can be generated. However, not every building plan has been
implemented using these advanced engineering software. There are still plans which are hand-
drawn or are not saved in particular structured format allowing 3D generation. The aim of our
paper is to tackle this issue whereby no digital 2D plan exists (e.g. DWG format), accepting a 2D
floor plan, scan it as an image, pre-process the plan and reconstruct a 3D plan out of the 2D plan.
For this purpose, image processing techniques have been applied to extract semantic and spatial
information, recognising architectural symbols such as walls, windows and doors only and locate
their positions in the 2D sketch so that these details are used to generate a pseudo-3D model of
the building. Information extracted is stored in an XML format which can be easily ported to any
architectural package format. Based on our XML format we used a game engine, Irrlicht [14], to
generate the 3D plan allowing navigation and walkthrough.
Several work on this theme have been done previously as elaborated in section 2 (Previous work),
with research spanning from effective symbol recognition algorithm to implementation of whole
system/package to generate 3D models out of 2D floor plans. The aim of our paper is to
implement the same step, but using existing image processing library like OpenCV [24] and 3D
engines. Functions provided by image processing library OpenCV 2.2 [24] have been used
throughout this project to evaluate and find the most appropriate image processing techniques that
can be used for the detection and extraction of the architectural symbols. Only walls, windows
and doors symbols which are oriented horizontally and vertically have been considered while
building the system. Also, once pre-processed, spatial and semantic information are gathered and
stored in XML format to be loaded in an existing 3D engine. Though our system is at its infancy
and does not cater for all possibilities, interesting results have been noticed in terms of symbol
detection rate and 3D building generation.
The rest of the paper is organised as follows: Section 2 describes similar existing work on
architectural symbol recognition and 3D building generation. Section 3 provides a description of
our proposed system while in Section 4, we elaborate on our implementation process. Section 5
analyses the results achieved through our system. Finally, we conclude in terms of our
achievements and future directions in Section 6.
2. PREVIOUS WORK
Lewis and Sequin in [1] developed a robust, semi-automatic system to create 3D polyhedral
building models from computer drawn floor plans, requiring minimal user interaction. A
prototype system, called the Building Model Generator (BMG) which accepts 2D floor plans in
the common AutoCAD DXF geometry description format. The system starts by cleaning up the
2D floor plans, then converting them into suitable internal data structure permitting efficient
geometric manipulations and analysis. After correction of small local geometrical inconsistencies
3. International Journal of Computer Graphics & Animation (IJCGA) Vol.1, No.2, June 2011
15
and necessary adjustments are made, a consistent layout topology is obtained. Semantic
information such as room identities and connecting portals are extracted, walls are extruded to a
specified height, and door and window geometries are inserted where appropriate. Hence a 3D
representation of the building is generated permitting visualisation in interactive walkthrough.
The models generated are directly compatible with the Berkeley WALKTHROUGH system and
with the NIST CFAST fire simulator. Such work demonstrated that the construction of 3D
building models from existing floor plans have become manageable and affordable reducing
built-in time from months to days.
Lu & al in [2] proposed new methods to analyse architectural working drawings (AWD), usually
used to describe design intents of architects, recognize typical structural objects and architectural
symbols. They categorise AWD into three types of entities namely structural, functional and
decorative. Structural entities deal with structures supporting and resisting load of the whole
building such as steel and concrete. Functional entities such as doors, windows, lifts etc provide
convenience to the users. Examples of decorative entities are partition walls, hung ceilings etc.
Processing of these AWDs are done in three steps namely recognition of structural entities using
shape based method, removal of graphical primitives of recognized structural objects and
recognition of architectural entities from simplified drawings using feature-based symbol
recognition. They test their model on 257 plane architectural/structural drawings, 40 being
synthetic and 217 being real data. This model shows interesting detection rate with above 80%
correct detection rate on synthetic and real data set.
S.-H. Or et al in [3] proposes a highly automated approach to generate 3D model from a 2D floor
plan. The system parses a floor plan into a number of connected segments and analyse their
relationship to generate the 3D model accordingly. The plan is pre-processed converting the raster
image into vector image. Symbol recognition is used for identification of doors and windows. The
system assumes that any arc symbols represent a door while thin boxes represent windows. Once
these symbols are identified, the 3D models are generated and loaded on Genesis 3D, a 3D game
engine. However, the system needs some improvements in terms of recognition of building
entities, creation of multi-storey building.
Dosch et al in [4] proposed a complete system to reconstruct scanned 2D architectural drawings
in 3D. They describe their robust image processing and feature extraction algorithm by dividing
the scanned 2D images into tiles, processing each part separately and then merging them again
after vectorisation. Using the skeleton based approach they extract lines and represent them into
segments using the polygon approximation technique. Their system allows for dashed line, arcs
and staircase detection. A user interface is provided for human interaction and correction of errors
and interactively manipulating the resulting data. They advocate the stability of their image
processing and feature and dashed line extraction algorithm. Nevertheless, there are still some
improvements need to be done and tests need to be made on how scalable this system is.
[5] proposed a Self-Incremental Axis-Net-based Hierarchical Recognition (SINEHIR) model for
automatic recognition and interpretation of real-life complex electronic construction structural
drawings. They designed and implemented shape-independent algorithms based on internal
semantic constraints rather than visual graphical constraints. Their methods identifies
characteristic features of structural components from the more regular constituents, and then
tracks the graphic objects as far as possible under the guidance and constraints of recognised
objects and the domain knowledge. Their approach was tested on more than 200 real-life
drawings and the results showed a 90% average recognition rate.
In [6], based on Messmer & Bunke ideas on symbol recognition, Ah-Soon & Tombre proposed a
method for recognising architectural symbols based on the description of the model through a set
4. International Journal of Computer Graphics & Animation (IJCGA) Vol.1, No.2, June 2011
16
of constraints on geometrical features, and on propagating the features extracted from a drawing
through the network of constraints. They advocate the flexibility of such approach to
accommodate new symbols.
[7] proposes a reconstruction method based on four main phases namely (i) 2D edges processing
for removing geometrical inconsistencies, (ii) topological reconstruction with semantic
information, (iii) 3D building extrusion (iv) superimposing of floors. The model expresses
incidence and adjacency relations between all the elements with semantic information associated
with all volumes such as openings, portals, stairs etc. Conversion from 2D to 3D is done through
extrusion using specific rules guided by the semantics.
3. PROPOSED SYSTEM
Our proposed system is a 3-phase system namely (i) Image Processing (ii) Save Module (iii)
Interactive 3D module.
Figure 1. 3-Phase proposed system
3.1 Image Processing
The system loads a scanned image of a floor plan consisting of wall, door and window symbols
oriented horizontally and vertically. The scanned image is pre-processed by removing noise and
text, converting it to a vectorised image. The image is preprocessed through thresholding and
edge detection. Horizontal and vertical lines are extracted from the resulting image using
probabilistic Hough transform. A region growing approach is then applied to cluster the lines to
identify walls and derive their positions. Then, the windows are identified from the detected walls
using the characteristics of the wall symbols and using a similar region growing approach to
walls’ identification. The detected walls and windows are subtracted from the image to obtain a
difference image with only door symbols. The difference image is then segmented using a
contour finding approach to locate the position of each object in the image. A histogram matching
technique is applied to identify which of the remaining objects is a door.
Interactive 3D Module
6.3 Doors and Windows from
Template
6.2 Wall
6.1 Floor &
Ceiling
8. Interactive
7. Navigation
6. 3D Model
Save Module
4. Architectural Symbol Localisation 5. XML Generation
Image Processing Module
Scanned
Image
3. Symbol Recognition
Wall Identification
Door Identification
Text Removal
Noise Removal
Raster to Vector
2. Image
Window Identification
5. International Journal of Computer Graphics & Animation (IJCGA) Vol.1, No.2, June 2011
17
3.2 Save Module
For each recognised door, their relative position in the image is calculated. Finally the dimension
of a pseudo roof for the building is calculated from the sketch. All of the identified walls,
windows, doors and approximate roof positions are saved in an XML file.
3.3 Interactive 3D module
The Interactive 3D generation module calculates the position, length, height and width of the
wall, door and windows models from the generated XML file. The 3D model of the building is
then generated, enabling the user to visualise the building in different perspective and navigate
through it. The system provides a user friendly menu driven interface. Additional facilities that
the system provides are (i) full screen navigation, (ii) interactive modification of 3D models, e.g.,
applying different textures to either interior or exterior walls separately, (iii) loading different
window, door models, roof and tiles, (iv) taking screenshots while navigating, (v) saving XML
containing the modified 3D building description and (vi) loading of existing XML file.
4. IMPLEMENTATION
The object detection and localisation module of the system has been implemented in C/C++ using
data structures and functions provided by OpenCV image processing library. TinyXml was used
as an XML parser with Irrlicht Engine to create the GUI, 3D generation and navigation system.
4.1 Image Preprocessing
4.1.1 Wall Identification
To identify walls, lines are extracted from the preprocessed image using probabilistic Hough
transform [20] resulting in a sequence of lines. The lines are then separated as horizontal and
vertical lines sequences only. For each sequence of lines, a clustering technique is applied to
group lines into bounding boxes, each representing a potential wall. The bounding box is
dynamically adjusted to accommodate inserted elements. The number of horizontal bounding
boxes is reduced by any grouping overlapping boxes as one single bounding box. Similar
technique is used to group vertical lines in bounding boxes to identify potential vertical walls.
Intersecting endings of vertical and horizontal bounding boxes are then aligned as shown in the
Figure 2 below. It should be noted that the identified walls may contain windows. The clustering
algorithm assumes that for a line forming part of wall in the image and to be included in a
bounding box, it should be at most 50 pixels from the edge of the bounding box. This value is
used as the stopping condition for the clustering algorithm. Considering horizontal sequence of
lines only, the following algorithm identifies potential horizontal walls:
6. International Journal of Computer Graphics & Animation (IJCGA) Vol.1, No.2, June 2011
18
Figure 2. (top left) Partial floor plan. Figure 3. (top middle) Walls symbols grouped in bounding
boxes. Figure 2. (top right) Bounding box adjusted.
WHILE (horizontal sequence, h, contains lines)
Put first line from h in bounding box
FOR each line in h
IF (line is in (bounding box enlarged by 50 pixels on each side)) THEN
Put line in bounding box
END IF
END FOR
FOR each line in bounding box
Remove line from h
END FOR
END WHILE
Figure 3. Algorithm for bounding box.
4.1.2 Window Identification
The same procedure as wall identification, before
alignment of horizontal and vertical boxes, has
been used for window detection. The algorithm
in Figure 4 describes the extractions of vertical
window symbols. Here we assumed that the
length of line_v should be at least three-quarter
(3/4) height of bounding box to be accepted.
Height of bounding box must be at two (2) times
longer than width of Region of Interest (ROI) and
less than half height of ROI to be considered a
window. Windows cannot be found at corners of
ROI i.e., wall symbols.
Figure 4. Wall Identification
7. International Journal of Computer Graphics & Animation (IJCGA) Vol.1, No.2, June 2011
19
FOR each vertical bounding box positions
Set positions as ROI in preprocessed image
Get width and height of ROI
Extract lines L, from in ROI using probabilistic Hough transform function
Get horizontal lines h, from L using getHorLines method
Get vertical lines v, from L using getVerLines method
Sort h using sortTopDown method
Sort v using sortLeftRight method
FOR each line in h, line_h
Put line_h and next line_h in bounding box
FOR each line in v, line_v
IF (length line_v ≥ 0.75 × height of bounding box) THEN
Put line_v in bounding box
END IF
END FOR
IF (height of bounding box ≥ (2 ×width) AND ≤ height÷2) THEN
Increment number of windows (bounding boxes)
ELSE
Clear bounding box
END IF
END FOR
Figure 5. Window symbol detection
4.1.3 Door Identification
After walls and windows were
identified from a floor plan image,
these symbols were subtracted from
the edge image leaving only the
remaining objects which were then
segmented and put in bounding boxes.
Figure 6 shows the isolation of
remaining objects before door
identification.
Figure 6. Isolation of remaining objects
Each identified object then undergoes histogram matching which used a combination of Chi-
square and Bhattacharyya distance metrics, and template matching which uses normalised
correlation coefficient matching method, against a small set of sample door images. After this
operation, only bounding boxes containing door symbols are left. It can be said that door symbols
detection has a very high success rate.
During histogram matching, an assumption was made that for a ROI to be detected as a door
symbol, the results by both Chi-square and Bhattacharyya distance metrics must be below than
0.2, i.e., a good match. The assumption that was made for template matching template is that the
result obtained by using correlation coefficient method should be above 0.9 to be a good match. If
either histogram or template matching is successful, then the ROI is classed as a door symbol.
The implementation details to detect doors is describes in Figure 7.
8. International Journal of Computer Graphics & Animation (IJCGA) Vol.1, No.2, June 2011
20
Number of samples for templates, TS.
Number of samples for histogram matching, HS.
FOR each vertical wall positions
Subtract vertical wall image part from image_edge
END FOR
FOR each horizontal wall positions
Subtract horizontal wall image part from image_edge
END FOR
Find contours in image_edge using cvfindContours function
Redraw on image_edge using redraw method
Find contours in image_edge
FOR each set of contours
Put contour in bounding box, b
Increment number of bounding box
END FOR
Set isDoorHistogram to false
Set isDoorTemplate to false
FOR each ROI represented by array b
FOR i:=0 to HS
Load sample door image i
Do histogram matching of ROI and sample using compareImgHistogram method
IF (match is true) THEN
Set isDoorHistogram to true
END IF
END FOR
FOR i:=0 to TS
Load template door image i
Do template matching of ROI and template using compareImgTemplate method
IF (match is true) THEN
Set isDoorTemplate to true
END IF
END FOR
IF ((isDoorHistogram OR is DoorTemplate) is true) THEN
Get door bounding box positions
Increment number of doors
END IF
END F
Figure 7. Algorithm for door identification
4.2 Getting door and window positions
After the doors have been identified, they need to
be aligned to their adjoining walls. This is
achieved by incrementing each side of the
bounding boxes representing doors by the
average thinness of walls in the image and
calculating which horizontal or vertical walls,
previously detected, ends in the bounding box.
Thus the start and end points for the doors, being
represented as lines, are obtained.
Figure 8. Aligning window to adjoining wall
9. International Journal of Computer Graphics & Animation (IJCGA) Vol.1, No.2, June 2011
21
4.3 Saving detected object
After symbols identification, the detected walls, windows, doors [and roof positions] are saved in
the XML file. Figure 9 shows a partial xml contain a floor plan attributes.
Figure 9. Partial XML file generated
4.4 User Interaction and Generating the 3D model
The graphical user interface (GUI) and the 3D generation module have been implemented in
C/C++ using the Irrlicht engine. The GUI allows the user to load a floor plan image and trigger
the object detection module which produces an XML file with the detected objects’ attributes.
The latter is then read by the 3D generation module which calculates the coordinates at which to
place the objects such as the walls, doors and windows in the 3D model. It then loads the
corresponding objects, map their textures draws the 3D model of the building. The GUI presents
the 3D model of the building in several different views as shown in Figure 10-11 below.
10. International Journal of Computer Graphics & Animation (IJCGA) Vol.1, No.2, June 2011
22
Figure 10. Generated 3D house Figure 11. Top view without the
roof
5. RESULTS AND ANALYSIS
Our 3-phased 3D building system has been implemented using C, C++ and OpenCV library for
the Image Processing module; The Save Module generated an XML file for storing the processed
floor plan objects attribute; while the Irrlitch [14] game engine was used to implement the
Interactive 3D module. Though still at its infancy, our proposed system gave commendable
results.
5.1 Benchmark Testing
In this section, testing is made to know the performance of the software on different hardware
systems. The frame rate of the software has been considered for testing the performance.
Table 1. Capabilities of system tested
System A:
32MB graphics card
256Mb of RAM
750 MHz Processor
Frame rate: ~16
System B:
256MB graphics card
1GB of RAM
2.2 MHz Processor
Frame rate: ~250-300
5.2 Portability Test
The software was compiled and run using Codeblocks IDE on Fedora 10, a Linux operating
system and it could successfully load an image and generate the 3D model, though the frame rate
was very low (frame rate was 4 as the proper graphic card driver was not installed on the OS).
Top view
Navigate view
Animated view
Front view
11. International Journal of Computer Graphics & Animation (IJCGA) Vol.1, No.2, June 2011
23
5.3 Performance Test
Table 2. Performance Testing results
Floor
plan 1
Floor
plan 2
Floor
plan 3
Floor
plan 4
Floor
plan 5
Floor
plan 6
Complexity Low Medium Medium Low High Medium
Wall detection 25 /25 -
100%
24/30 –
80%
34/54 –
62.9%
17/17 -
100%
38/49 –
77.6%
26/30-
86.7%
Window detection 8/9 -
88.9%
3/9 –
33.3%
2/8 -
25%
3/6 -
50%
2/12 –
16.7%
2/4 -
50%
Door Detection 6/7 –
85.7%
6/14-
42.8%
4/17 -
23.5%
8/8 -
100%
9/15 -
60.0%
5/9 –
55.6%
Image Processing time 2 s 3s 3s 2s 5s 4s
3D model
Faces 246 355 535 203 511 427
Triangle mesh 884 1210 1590 758 1528 1409
Average generation time 4s 4s 4s 4s 4s 4s
Based on Table 2 it can be said that the system produced commendable results in recognition of
walls, windows and doors symbols from floor plans images. Walls positions were extracted with
a very high success rate. Windows identification using probabilistic Hough transform and
clustering technique produced fairly good results. The combination of histogram and template
matching to identify door symbols was very reliable as it allows all doors to be detected in floor
plan images. However, better algorithms could not be derived to properly align all the detected
doors in their final positions.
5.4 Processing Time
Image processing for identification of objects depended on the complexity of the floor plan. The
more complex the plan in terms of walls, doors and windows, the more time it took for
processing. Computing times have been obtained with a 2.2 MHz processor, 1GB RAM and 256
MB RAM graphic card. However, once the XML was generated, then the 3D model generation of
the Building was performed on average in 3 seconds irrespective of the model.
6. CONCLUSION
In this paper we demonstrated some conclusive results on our 3-phase recognition approach to
pseudo 3D building generation from 2D floor plan. Our system based itself on the idea of using
existing concepts, algorithm, libraries and technologies with minimal modifications and
integrating them into a fully functional working application. Hence, we used C, C++ and OpenCV
for image processing and recognition, XML for storage and Irrlitch [14] engine for 3D model
generation.
We used existing image processing techniques like probabilistic Hough transform, Histogram
matching, Clustering and bounding box techniques , Chi-square and Bhattacharyya distance
metrics to recognise architectural symbols like walls, doors and windows. XML technology was
used as the interface and the bonding technology between image processing and 3D generation of
building. It was also used as the storage format for saving models and processed image. The 3D
Generation module provides a sense of aesthetic and a realistic scene in the 3D environment for
the user’s experience with the system. It provides interactive and navigational capabilities with
different views.
12. International Journal of Computer Graphics & Animation (IJCGA) Vol.1, No.2, June 2011
24
However, the system developed is not 100% efficient in identifying and localising the walls,
windows and doors within the floor plans used during the tests carried out. Nevertheless it has
been able to detect most wall symbols with some tiny false positive walls forming part of a door
symbols being recognised as part of a wall; most window symbols are detected and door symbols
have been recognised. However, some doors are not properly aligned with their adjacent walls
after being detected and this need some improvement. Also, our 3D module is still at an embryo
stage. It needs to be enhanced to produce more realistic model, enabling users to change color,
textures, light and the ability to interactively insert objects like tables, chairs etc in the model
generated.
6.1 Future Works
Though the object detection and localisation module is fully functional some feasible
improvements can done to further enhance the system. Some identified future works are as
follows: (1) recognition of walls other than at horizontal and vertical positions. This would
involve replacing the rectangular bounding box used in the module by a bounding box which can
be inclined at different angles; (2) detection of different types of doors and windows such as
sliding, interior or exterior doors by modifying the template matching method; (3) detection of
other architectural symbols such as supporting columns and stairs; (4) extending the system to
allow for multiple storey building.
As a concluding note it can be said that using our approach based on architectural symbol
recognition, significant reduction in time is to be noted in 3D building generation as compared to
traditional 3D building implementation. This has opened new avenues with commercial potential
in creating virtual environments especially for games, virtual property advertising and also for
modelling historical building having only paper plans.
Figure 12. Overview of generated 3D model
13. International Journal of Computer Graphics & Animation (IJCGA) Vol.1, No.2, June 2011
25
REFERENCES
[1] R.Lewist, & C.Sequin. (1998). Generation of 3D building models from 2D architectural plans.
Computer Aided Design.30(10), pp. 765-779. Elsevier Science.
[2] Lu, T., Yang, H., Yang, R., & Cai, S. (2006). Automatic analysis and integration of architectural
drawings. Internation Journal on Document Analysis and Recognition IJDAR, 9 (1)pp. 31-47.
[3] S H Or, et al. (2005). Highly Automatic Approach to Architectural Floorplan Image Understanding and
Model Genration. Proceedings of Vision, Modeling, and Visualization,Pattern Recognition (pp. 25-
32). Erlangen,Germany: IOS Press.
[4] Dosch, P., Tombre, K., Ah-Soon, C., & Masini, G. (2000). A complete system for the analysis of
architectural drawings. International Journal on Document Analysis and Recognition , 3(2) pp.102-
116.
[5] Lu, T., Taib, C.-L., Sua, F., & Cai, S. (2005). A new recognition model for electronic architectural
drawings. Computer Aided Design , 37(10) pp. 1053-1069.
[6] Ah-Soon, C. et al., (2001). Architectural symbol recognition using a network of constraints. Pattern
Recognition Letters, 22 (2) p.231-248.
[7] Horna, S. et al., (2007). Building 3D indoor scenes topology from 2D architectural plans. Computer
Graphics Theory and Applications (GRAPP), pp. 37-44. Available at: http://citeseerx.ist-
.psu.edu/viewdoc/download?doi=10.1.1.100.6752 &rep=rep1& type=pdf.
[8] AutoDesk, AutoCad Products.[online] Available at:
http://usa.autodesk.com/adsk/servlet/pc/index?siteID=123112&id=15408302 [Accessed 15 Dec 2010]
[9] Pilkinton, B. (2010, 12 1). Article Squeeze .The Importance Of 3D Architectural Visualisation.
[online] Available at: http://www.articlesqueeze.com/business-articles/the-importance-of-3d-
architectural-visualisation/ [Accessed 15 Dec 2010]
[10]Google Sketchup. [online] Available at: http://sketchup.google.com/intl/en/index.html[Accessed27
Dec, 2010]
[11]Autodesk. 3DS Max Products.[online] Available at:
http://usa.autodesk.com/adsk/servlet/pc/index?siteID=123112&id=15683479 [ Accessed 27 Dec 2010]
[12]Autodesk .Autodesk Maya. [online] Available at:
http://usa.autodesk.com/adsk/servlet/pc/index?id=13577897&siteID=123112 [ Accessed 27 Dec 2010]
[13]Blender .Blender. [online] Available at: http://www.blender.org/ [ Accessed 27 Dec 2010]
[14]IrrlitchIrrlitch . [online] Available at:http://irrlicht.sourceforge.net/ [ Accessed 10 Jan 2011]
[15]Gibson S, Howard T. (2000). Interactive reconstruction of virtual environments from photographs,
with application to scene-of-crime analysis. In Proceedings of ACM Symposium in Virtual Reality
Software and Technology 2000, October 2000. Seoul, Korea
[16]Fisher R et al. (2004). Explore with Java -Adaptive Thresholding Chapter.[online] Available
at:http://homepages.inf.ed.ac.uk/rbf/HIPR2/adpthrsh.htm[Accessed 24 Dec 2008]
[17]Green B (2002). Canny edge Detection Tutorial[online] Available at:
http://www.pages.drexel.edu/~weg22/can_tut.html [Accessed 14 March 2009]
[18]Raphael C.. Gonzalez, Richard E. Woods(2002). Digital image Processing, 2nd Edition, Prentice Hall
International.
[19]Fisher R et al. Intensity Histogram[online] Available at: http://homepages.inf.ed.ac.uk/rbf-
/HIPR2/histgram.htm[Accessed 25 Dec 08]
[20]Gary Bradsky , Adrian Kaehler. (2008). Learning OpenCV. O’Reilly.
14. International Journal of Computer Graphics & Animation (IJCGA) Vol.1, No.2, June 2011
26
[21]G.Chaudri.. Bhattacharyya distance. Available at: http://eom.springer.de/B/b110490.htm [Accessed 15
March 09]
[22] Longin Jan Latecki. Template matching, Available at:
http://www.cis.temple.edu/~latecki/Courses/CIS601-03/Lectures/TemplateMatching03.ppt[Accessed
14 March 2009]
[23] Binary Image processing.[Online].Available at http://homepages.inf.ed.ac.uk/rbf/CVonline/-
LOCAL_COPIES/MARBLE/medium/binary/ [Accessed 18 March 2009]
[24] OpenCV, Open Computer Vision Library. [Online] Available at:
http://sourceforge.net/projects/opencvlibrary/ [Accessed 18 June 2011]
[25]Xuetao Yin, Wonka, P., Razdan, A. Generating 3D Building Models from Architectural Drawings: A
Survey, IEEE Computer Graphics and Applications - CGA , vol. 29, no. 1, pp. 20-30, 2009. DOI:
10.1109/MCG.2009.9
[26]Su, F., Song, J. & Cai, S., 2001. Dimension recognition and geometry reconstruction in vectorization
of engineering drawings. Proceedings of the 2001 IEEE Computer Society Conference on Computer
Vision and Pattern Recognition CVPR 2001, p.I-710-I-716
[27]Jung, I., Mckinley, L. & Weg, Z., 2008. Three-dimensional building reconstruction : a process for the
creation of 3D buildings from airborne LiDAR and 2D building footprints for use in urban planning
and environmental scenario modelling. Available at: http://www.virtualcitysystems.de/uploads/-
media/BuildingReconstruction_Paper_RealCorp_2008.pdf
[28]Horna, S. et al., 2009. Consistency constraints and 3D building reconstruction.Computer-Aided Design,
41(1), p.13-27
15. International Journal of Computer Graphics & Animation (IJCGA) Vol.1, No.2, June 2011
27
Authors
Mr. Raj Kishen Moloo is a lecturer at the University of Mauritius. He has a
Masters degree in Advanced Computer Science at the University of Manchester
(2005) and BEng (Hons) Computer Science and Engineering (2003) from the
University of Mauritius. He is also an ACCA affiliate. His main research
interests are Computer Graphics, Games, Virtual Reality and Mobile
Application Development.
Mr. Muhammad Ajmal Sheik Dawood is enrolled for an MSc Financial
Software Engineering at the University of Essex. He has a BSc (Hons) Computer
Science and Engineering from the University of Mauritius.
Mr. Abu Salmaan Auleear is working as a Senior Web-Designer at Chesteroc
Ltd (Mundocom Mauritius). He has a BSc (Hons) Computer Science and
Engineering from the University of Mauritius.