• Journal of KIBIM
• /
• v.11 no.4
• /
• pp.53-62
• /
• 2021

The successful implementation of Scan-to-BIM automation depends on the entire process from scanning of buildings, including indoor facilities and furniture, to generating BIM models. However, the conventional Scan-to-BIM process requires a lot of time, manpower, and cost for the manual generation of BIM models including indoor objects. To solve this problem, this study applied a Scan-to-BIM automation process using a deep learning model and parametric algorithm to an existing building, Kangwon Fire Service Academy. To improve the accuracy of the BIM model, after object data was extracted from the scan data, the data was corrected according to actual object-specific conditions. As a result, the accuracy of the BIM model created by the proposed Scan-to-BIM automation process was 91% compared to the actual area of the construction drawings. In addition, it was confirmed that the BIM objects were automatically generated for 10 object classes.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.01a
• /
• pp.223-228
• /
• 2009

We propose an on-line machine learning approach for object recognition, where new images are continuously added and the recognition decision is made without delay. Random forest (RF) classifier has been extensively used as a generative model for classification and regression applications. We extend this technique for the task of building incremental component-based detector. First we employ object descriptor model based on bag of covariance matrices, to represent an object region then run our on-line RF learner to select object descriptors and to learn an object classifier. Experiments of the object recognition are provided to verify the effectiveness of the proposed approach. Results demonstrate that the propose model yields in object recognition performance comparable to the benchmark standard RF, AdaBoost, and SVM classifiers.

• Journal of Korean Society for Geospatial Information Science
• /
• v.19 no.3
• /
• pp.49-56
• /
• 2011

Because urban areas change rapidly, it is necessary to reflect urban changes in a digital map database in a timely manner. To address these issues, LiDAR data was used to detect changes in urban area buildings. The purpose of this study is to detect object-based building change using LiDAR data and existing digital maps, and classify change types. In the study, we classified change type using overlay and shape comparison with building layer of the digital maps and point-based extracted building outline from the LiDAR data. When applying the overlay method, we were able to increase the accuracy and objectivity of the change detection process throughout an adaptive threshold applied to each object. In the experiments, it was demonstrated that classifying and detecting changes in urban areas using the proposed method can provide superior classification accuracy compared with the existing methodology.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.134.1-134
• /
• 2001

The paper presents a sonar based map building method. The environment is a room or area inside a building, which is composed of four types of geometric primitives(corners, edges, cylinders, and walls). We also assume the environment can be modeled into two dimensional map in terms of planes(walls), points(corner and edge), and circle(cylinder). In a real world where most of the object surfaces are specular ones, a sonar sensor suffers from a multipath effect which results in a wrong interpretation of the location of an object. To reduce the effect and uncertainty, the method employs a simple thresholding technique for extracting circular arc features called regions of constant depth(RCD) from scanning sonar data. The usefulness of the approach is illustrated with the results produced by sets of experiments.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.20 no.2
• /
• pp.133-138
• /
• 2011

RFID system can be used to improve object recognition, map building and localization for robot area. A novel method of indoor navigation system for a mobile robot is proposed using RFID technology. The mobile robot With a RFID reader and antenna is able to find what obstacles are located where in circumstance and can build the map similar to indoor circumstance by combining RFID information and distance data obtained from sensors. Using the map obtained, the mobile robot can avoid obstacles and finally reach the desired goal by $A^*$ algorithm. 3D map which has the advantage of robot navigation and manipulation is able to be built using z dimension of products. The proposed robot navigation system is proved to apply for SLAM and path planning in unknown circumstance through numerous experiments.

• Wind and Structures
• /
• v.1 no.3
• /
• pp.271-284
• /
• 1998

This paper uses the numerical simulation to investigate the interference effect of 3-D turbulent flow around two high rise buildings in proximity at the different relative heights, gaps, and wind velocities. The computer program used to carry out the simulation is based on the control volume method and the SIMPLEST algorithm. The ${\kappa}-{\varepsilon}$ model was used to simulate turbulence effects. Since the contracted flow between two adjacent buildings enhances the strength of vortex shedding from the object building, the pressure coefficient on each side wall of the object building is generally increased by the presence of apposed building. The effect is increased as the relative height or the gap between the two buildings decreases. The velocity on the vertical center line between two buildings is about 1.4 to 1.5 times the upstream wind velocity.

• KSCI Review
• /
• v.14 no.1
• /
• pp.241-248
• /
• 2006

The recognition of a relatively big and rarely movable object. such as refrigerator and air conditioner, etc. is necessary because these objects can be crucial global stable features of Simultaneous Localization and Map building(SLAM) in the indoor environment. In this paper. we propose a novel method to recognize these big objects using a sequence of 3D scenes. The particles representing an object to be recognized are scattered to the environment and then the probability of each particles is calculated by the matching test with 3D lines of the environment. Based on the probability and degree of convergence of particles, we can recognize the object in the environment and the pose of object is also estimated. The experimental results show the feasibility of incremental object recognition based on particle filtering and the application to SLAM

• Journal of the Korea Institute of Building Construction
• /
• v.11 no.3
• /
• pp.276-282
• /
• 2011

The construction industry has achieved many benefits through the mechanization of the construction. However, due to accidents for construction machinery, the mortality rate has increased rapidly. Thus, the objective of this study is to develop of tilt modules for object detection system of excavator. The modified laser sensor that increase the efficiency of object detection system was tested. Based on these results, tilt modules were developed through the conceptual design and detailed design. Then, this tilt modules can be applied to the safety management system of excavator and using this tilt modules will be able to improve the efficiency of the system.

• Korean Management Science Review
• /
• v.14 no.2
• /
• pp.113-137
• /
• 1997

Intranet emerges as a key technology for building enterprise information system. This paper proposes a scenario-based object- oriented technique for designing intranet hypermedia information systems. The method consists of six phases such as domain analysis, object modeling, view design, navigational design, implementation design and construction. Users requirements are analyzed in the form of scenarios by the use fo a responsibility-driven object technology. Object-oriented views are generated from the resulting object model and then used for the subsequent navigational and implementation design. Implementation design phase deals integrating enterprise databases with distributed hypermedia systems by employing Java language. To demonstrate its usefulness, a real-life bank case is illustrated.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.21 no.3
• /
• pp.175-195
• /
• 1996

In conventional analysis methods, data and process are loosely coupled for building information systems. Several object oriented approaches have been proposed to integrate data and process. However, object oriented analysis requires a radical paradigm and thus system analysts find difficulties in generating object models direcctly from end users. To alleviate these difficulties, this paper proposes a methodology for deriving an object model by using structured analysis results. Objects are obtianed primarily from entities in Entity-Relationship Diagram. Methods are obtained through the analysis of the relationship between processes and data stores in Data Flow Diagram Methods are assigned to the objects by using object/process matrices. A real-life case is illustrated to demonstrate the usefulness of the methodology.