• Journal of Advanced Navigation Technology
• /
• v.27 no.2
• /
• pp.190-196
• /
• 2023

Recently, following the development of LIDAR technology which can detect distance from the object, the interest for LIDAR based 3D object detection network is getting higher. Previous networks generate inaccurate localization results due to spatial information loss during voxelization and downsampling. In this study, we propose an attention-based convergence method and a camera-LIDAR convergence system to acquire high-level features and high positional accuracy. First, by introducing the attention method into the Voxel-RCNN structure, which is a grid-based 3D object detection network, the multi-scale sparse 3D convolution feature is effectively fused to improve the performance of 3D object detection. Additionally, we propose the late-fusion mechanism for fusing outcomes in 3D object detection network and 2D object detection network to delete false positive. Comparative experiments with existing algorithms are performed using the KITTI data set, which is widely used in the field of autonomous driving. The proposed method showed performance improvement in both 2D object detection on BEV and 3D object detection. In particular, the precision was improved by about 0.54% for the car moderate class compared to Voxel-RCNN.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.23 no.12
• /
• pp.1337-1342
• /
• 2012

In this paper, we propose a novel reconfigurable antenna that can change the electrical shape of the matching network using RF switches of PIN diodes. The designed antenna operates at two different modes that are Mode 1 (HSDPA band, 2.1~2.2 GHz) and Mode 2(WiBro WiFi band, 2.3~2.5 GHz). The antenna is built on both sides of a polyarcylate substrate. The measured reflection coefficient shows a matching bandwidth of 547 MHz($S_{11}$ <-3 dB, 2.035~2.582 GHz) for Mode 1 and 600 MHz($S_{11}$ <-3 dB, 2.2~2.8 GHz) for Mode 2, and it shows average vertical gains of -4.4 dBi and -4.5 dBi in x-y plane, respectively.

• Journal of Sensor Science and Technology
• /
• v.16 no.2
• /
• pp.110-114
• /
• 2007

In this paper we present 3-D Navigation View, a three-dimensional visualization of indoor environment which serves as an intuitive and unified user interface for our developed indoor location tracking system via Virtual Reality Modeling Language (VRML) in web environment. The extracted user's spatial information from indoor location tracking system was further processed to facilitate the location indication in virtual 3-D indoor environment based on his location in physical world. External Authoring Interface (EAI) provided by VRML enables the integration of interactive 3-D graphics into web and direct communication with the encapsulated Java applet to update position and viewpoint of user periodically in 3-D indoor environment. As any web browser with VRML viewer plug-in is able to run the platform independent 3-D Navigation View, specialized and expensive hardware or software can be disregarded.

• Proceedings of the KIEE Conference
• /
• 2001.07b
• /
• pp.570-573
• /
• 2001

This paper presents the characteristics of Switched Reluctance Motor (SRM) based on 3D Equivalent Magnetic Circuit Network Method (3D EMCNM). 3D EMCNM supplements magnetic equivalent circuit by numerical technique using distributive magnetic circuit parameters. However in case of SRM the previous 3D EMCNM which uses fan shape element is not proper for analysis. This paper solved this problem by developing the trapezoid element and verified the validity of the suggested element by comparing the results of 3D EMCNM with 2D FEM.

• Journal of the Institute of Convergence Signal Processing
• /
• v.2 no.2
• /
• pp.68-74
• /
• 2001

A design and implementation method for the UHF-band(800MHz) variable attenuator using the 3dB coupler is proposed. The 90 degree, 3-dB Coupler is used for the variable attenuator. The principal theory for the 3-dB coupler is introduced. The 3-dB Coupler is designed by the mathematical analysis and a computer simulation tool. A PIN diode is used for the variable resistor at UHF-band. The variable attenuator using the 3-dB coupler and the PIN diode is designed and implemented. The measured results for the variable attenuator by a network analyzer show that the insertion loss is below -l0dB, and the continuous variable attenuation range is about 10dB.

• Journal of IKEEE
• /
• v.26 no.3
• /
• pp.408-415
• /
• 2022

In this paper, we propose a 3D point cloud reconstruction technique from 2D images using efficient feature map extraction network. The originality of the method proposed in this paper is as follows. First, we use a new feature map extraction network that is about 27% efficient than existing techniques in terms of memory. The proposed network does not reduce the size to the middle of the deep learning network, so important information required for 3D point cloud reconstruction is not lost. We solved the memory increase problem caused by the non-reduced image size by reducing the number of channels and by efficiently configuring the deep learning network to be shallow. Second, by preserving the high-resolution features of the 2D image, the accuracy can be further improved than that of the conventional technique. The feature map extracted from the non-reduced image contains more detailed information than the existing method, which can further improve the reconstruction accuracy of the 3D point cloud. Third, we use a divergence loss that does not require shooting information. The fact that not only the 2D image but also the shooting angle is required for learning, the dataset must contain detailed information and it is a disadvantage that makes it difficult to construct the dataset. In this paper, the accuracy of the reconstruction of the 3D point cloud can be increased by increasing the diversity of information through randomness without additional shooting information. In order to objectively evaluate the performance of the proposed method, using the ShapeNet dataset and using the same method as in the comparative papers, the CD value of the method proposed in this paper is 5.87, the EMD value is 5.81, and the FLOPs value is 2.9G. It was calculated. On the other hand, the lower the CD and EMD values, the better the accuracy of the reconstructed 3D point cloud approaches the original. In addition, the lower the number of FLOPs, the less memory is required for the deep learning network. Therefore, the CD, EMD, and FLOPs performance evaluation results of the proposed method showed about 27% improvement in memory and 6.3% in terms of accuracy compared to the methods in other papers, demonstrating objective performance.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2003.05a
• /
• pp.151-154
• /
• 2003

In this paper, we propose a method to recognize a face image of the 3D using the LVQ network. LVQ network of the proposed method, We used the front view of a face image to get to a coded light to a training data, can group a face image including the side of various angle. For an usefulness authentication of this algorithm, Various experiment which classifies a face image of the angle was the low.

• Journal of Internet Computing and Services
• /
• v.21 no.2
• /
• pp.121-130
• /
• 2020

This paper suggests evaluating the military closed network data as an image which is generated by Generative Adversarial Network (GAN), applying an image evaluation method such as the InceptionV3 model-based Inception Score (IS) and Frechet Inception Distance (FID). We employed the famous image classification models instead of the InceptionV3, added layers to those models, and converted the network data to an image in diverse ways. Experimental results show that the Densenet121 model with one added Dense Layer achieves the best performance in data converted using the arctangent algorithm and 8 * 8 size of the image.

• Journal of Korea Multimedia Society
• /
• v.13 no.11
• /
• pp.1657-1666
• /
• 2010

Underwater wireless sensor networks (UWSN) need stable efficient data transmission methods because of environmental characteristics such as limited energy resource, limited communication bandwidth, variable propagation delay and so on. In this paper, we explain an enhanced hybrid transmission method that uses a hexagon tessellation with an ideal cell size in a two-dimensional underwater wireless sensor network model (2D) that consists of fixed position sensors on the bottom of the ocean. We also propose an energy efficient sensing and communication coverage method for effective data transmission in a three-dimensional underwater wireless sensor network model (3D) that equips anchored sensors on the bottom of the ocean. Our simulation results show that proposed methods are more energy efficient than the existing methods for each model.

• Proceedings of the Korean Institute Of Construction Engineering and Management
• /
• 2004.11a
• /
• pp.349-352
• /
• 2004

4D system has applied to construction project as a management tool after the late 1990's. Various 4D systems have been developed, however they have some problems that should be improved. Especially, the 4D system for civil engineering project needs synthesized 3D model between natural ground condition and physical facility type. It is a different problem comparing with the system for building project. 1'his study suggests an automatically synthesizing methodology between ground triangulate network and design triangulate network. Furthermore the study develops an application methodology of multi-texturing technique defined in virtual reality modeling language (VRML) for skipping the 3D model synthesizing process from the 4D model development processes. The suggested methodology is applied to the prototype of real 4D system. The proposed technique for 3D modeling may be used as an essential methodology for developing 4D system for civil engineering projects.