• Journal of KIISE:Software and Applications
• /
• v.31 no.7
• /
• pp.929-938
• /
• 2004

This paper proposes a pose-invariant face recognition method using cylindrical model and stereo camera. We divided this paper into two parts. One is single input image case, the other is stereo input image case. In single input image case, we normalized a face's yaw pose using cylindrical model, and in stereo input image case, we normalized a face's pitch pose using cylindrical model with previously estimated pitch pose angle by the stereo geometry. Also, since we have an advantage that we can utilize two images acquired at the same time, we can increase overall recognition performance by decision-level fusion. Through representative experiments, we achieved an increased recognition rate from 61.43% to 94.76% by the yaw pose transform, and the recognition rate with the proposed method achieves as good as that of the more complicated 3D face model. Also, by using stereo camera system we achieved an increased recognition rate 5.24% more for the case of upper face pose, and 3.34% more by decision-level fusion.

• Korean Journal of Remote Sensing
• /
• v.34 no.6_1
• /
• pp.1131-1142
• /
• 2018

The autonomous vehicles are being developed and operated widely because of the advantages of reducing the traffic accident and saving time and cost for driving. The vehicle localization is an essential component for autonomous vehicle operation. In this paper, localization algorithm based on sensor fusion is developed for cost-effective localization using in-vehicle sensors, GNSS, an image sensor and reference images that made in advance. Information of the reference images can overcome the limitation of the low positioning accuracy that occurs when only the sensor information is used. And it also can acquire estimated result of stable position even if the car is located in the satellite signal blockage area. The particle filter is used for sensor fusion that can reflect various probability density distributions of individual sensors. For evaluating the performance of the algorithm, a data acquisition system was built and the driving data and the reference image data were acquired. Finally, we can verify that the vehicle positioning can be performed with an accuracy of about 0.7 m when the route image and the reference image information are integrated with the route path having a relatively large error by the satellite sensor.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.67 no.7
• /
• pp.871-877
• /
• 2018

In this paper, we propose a triple nested PID control scheme for stable hovering of a quadrotor and propose a complementary filter based sensor fusion technique to improve the performance of attitude, altitude and velocity measurement. The triple nested controller has a structure in which a double nested attitude controller that has the angular velocity PD controller in inner loop and the angular PI controller in outer loop, is nested in a velocity control loop to enable stable hovering even in the case of disturbance. We also propose a sensor fusion technique by applying a complementary filter in order to reduce the noise and drift error included in the acceleration and gyro sensor and to measure the velocity by fusing image, gyro, and acceleration sensor. In order to verity the performance, we applied the proposed control and measurement scheme to hovering control of quadrotor.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.6 no.4
• /
• pp.314-320
• /
• 2006

Navigation in environments that are densely cluttered with obstacles is still a challenge for Autonomous Ground Vehicles (AGVs), especially when the configuration of obstacles is not known a priori. Reactive local navigation schemes that tightly couple the robot actions to the sensor information have proved to be effective in these environments, and because of the environmental uncertainties, STSF(Space and Time Sensor Fusion)-based fuzzy behavior systems have been proposed. Realization of autonomous behavior in mobile robots, using STSF control based on spatial data fusion, requires formulation of rules which are collectively responsible for necessary levels of intelligence. This collection of rules can be conveniently decomposed and efficiently implemented as a hierarchy of fuzzy-behaviors. This paper describes how this can be done using a behavior-based architecture. The approach is motivated by ethological models which suggest hierarchical organizations of behavior. Experimental results show that the proposed method can smoothly and effectively guide a robot through cluttered environments such as dense forests.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.1042-1044
• /
• 2003

A methodology of the integrating the similar color circle search of the spectral data and segmentation of the height data is developed. The method is then applied to study areas, and the results by IKONOS, LIDAR and data fusion are verified with the ground truth, and examined in terms of the accuracy. Results show that with the data fusion the accuracy are improved by about 15% in most of the study areas. The methodology for the detection of individual tree stands by data fusion is explored, and the utility of combinatorial use of the spectral and the height information is demonstrated.

• Journal of Institute of Control, Robotics and Systems
• /
• v.21 no.11
• /
• pp.1017-1026
• /
• 2015

Infrared (IR) target detection is one of the key technologies in Automatic Target Detection/Recognition (ATD/R) for military applications. However, IR sensors have limitations due to the weather sensitivity and atmospheric effects. In recent years, sensor information fusion study is an active research topic to overcome these limitations. SAR sensor is adopted to sensor fusion, because SAR is robust to various weather conditions. In this paper, a Boolean Map Visual Theory-Morphology (BMVT-M) method is proposed to detect targets in SAR and IR images. Moreover, we suggest the IR and SAR image registration and decision level fusion algorithm. The experimental results using OKTAL-SE synthetic images validate the feasibility of sensor fusion-based target detection.

• Journal of the Korean Society of Costume
• /
• v.61 no.1
• /
• pp.20-33
• /
• 2011

This study aims to analyze the expressive aspect of men's shirts through examining the characteristics of men's shirts with the trend of men's clothing well represented and considering the image based on high fashion. As a method of study, I have conducted the theoretical study through the literature data, and the content analysis through photographs from S/S collection in 2005 to F/W collection in 2010. Especially, in the recent men's shirts, the materials of glossy silk, wavering chiffon and jersey tend to appear and various images are represented with the tasteful decoration such as ruffle, pin tuck and embroidery together with pastel color. Thus, the result of this study is as follows: Firstly, the masculine and romantic images are researched as the image emphasizing the beautiful and soft masculine beauty which doesn't lose the masculinity even though they are seasonedwith the feminine elements. Soft masculine beauty is emphasized by using romantic image decoration of frills, beads and pin tuck or silk, jersey and openwork materials etc. Secondly, practical fusion images have been studied as the images emphasizing the freshness with practicality through grafting more than two items onto one shirt. Many designs have appeared representing new shapes by grafting more than two items onto one shirt or obscuring the border line between the items. Thirdly, the avant-garde chic images have been researched as the image emphasizing the chick image through representing unique and odd designs by excessively expressing the silhouette in general shapes of shirts. I come. to find that the mainly the designs, emphasizing the uniqueness and the distinction with transformed shapes of sleeves, cuffs and the length of shirts in exaggerated manners, are appearing. Thus, I intend to provide a reference material which will be useful for the newly created design and researches of figurative characteristics in the future fashion.

• Proceedings of the KSRS Conference
• /
• 2005.10a
• /
• pp.497-500
• /
• 2005

Research and development of remote sensing technique is necessary so that more accurate and extensive information may be obtained. To achieve this goal, the synthesized technique which integrates the high resolution optic and SAR image, and topographical information was examined to investigate the quantitative/qualitative characteristics of the Earth's surface environment. For this purpose, high-precision DEMs of Jeju-Island was generated and data fusion algorithm was developed in order to integrate the multi-spectral optic and polarimetric SAR image. Three dimensional land-cover and two dimensional soil moisture maps were generated conclusively so as to investigate the Earth's surface environments and extract the geophysical parameters.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.44 no.5
• /
• pp.407-415
• /
• 2016

In this paper, an aviation image fusion method is proposed for creating an informative fused image through gray scale images within noise. The proposed method is based on an adaptive guided filter which adjusts regulation parameter of the filter based on peak signal noise ratio (PSNR) in order to behave as an edge-preserving filtering property. Simulation results demonstrate that the proposed method preserves the edge information of the input image and reduces the noise effect while maintaining designed PSNR.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.7
• /
• pp.2328-2344
• /
• 2022

2D human pose estimation still faces difficulty in low-resolution images. Most existing top-down approaches scale up the target human bonding box images to the large size and insert the scaled image into the network. Due to up-sampling, artifacts occur in the low-resolution target images, and the degraded images adversely affect the accurate estimation of the joint positions. To address this issue, we propose a multi-resolution input feature fusion network for human pose estimation. Specifically, the bounding box image of the target human is rescaled to multiple input images of various sizes, and the features extracted from the multiple images are fused in the network. Moreover, we introduce a guiding channel which induces the multi-resolution input features to alternatively affect the network according to the resolution of the target image. We conduct experiments on MS COCO dataset which is a representative dataset for 2D human pose estimation, where our method achieves superior performance compared to the strong baseline HRNet and the previous state-of-the-art methods.