• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.5
• /
• pp.1856-1869
• /
• 2015

This paper addresses the issues of 3D human activity detection, tracking and recognition from RGB-D video sequences using a feature structured framework. During human tracking and activity recognition, initially, dense depth images are captured using depth camera. In order to track human silhouettes, we considered spatial/temporal continuity, constraints of human motion information and compute centroids of each activity based on chain coding mechanism and centroids point extraction. In body skin joints features, we estimate human body skin color to identify human body parts (i.e., head, hands, and feet) likely to extract joint points information. These joints points are further processed as feature extraction process including distance position features and centroid distance features. Lastly, self-organized maps are used to recognize different activities. Experimental results demonstrate that the proposed method is reliable and efficient in recognizing human poses at different realistic scenes. The proposed system should be applicable to different consumer application systems such as healthcare system, video surveillance system and indoor monitoring systems which track and recognize different activities of multiple users.

• Journal of Korean Society for Geospatial Information Science
• /
• v.10 no.5 s.23
• /
• pp.75-80
• /
• 2002

Depending upon geographical features and surrounding errors in the survey field, inaccurate positioning is inevitable in a kinematic DGPs survey. Therefore, a data inaccuracy detection algorithm and an interpolation algorithm are essential to meet the requirement of a digital map. In this study, GPS characteristics are taken into account to develop the data inaccuracy detection algorithm. Then, the data interpolation algothim is obtained, based on the feature type of the survey. A digital map for 20km of a rural highway is produced by the kinematic DGPS survey and the features of interests are lines associated with the road. Since the vertical variation of GPS data is relatively higher, the trimmed mean of vertical variation is used as criteria of the inaccuracy detection. Four cases of 0.5%, 1%, 2.5% and 5% trimmings have been experimented. Criteria of four cases are 69cm, 65cm, 61cm and 42cm, respectively. For the feature of a curved line, cublic spine interpolation is used to correct the inaccurate data. When the feature is more or less a straight line, the interpolation has been done by a linear polynomial. Difference between the actual distance and the interpolated distance are few centimeters in RMS.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.49 no.9
• /
• pp.113-127
• /
• 2012

Face detection methods using image processing have been proposed in many different ways. Generally, the most widely used method for face detection is an Adaboost that is proposed by Viola and Jones. This method uses Haar-like feature for image learning, and the detection performance depends on the learned images. It is well performed to detect face images within a certain distance range, but if the image is far away from the camera, face images become so small that may not detect them with the pre-learned Haar-like feature of the face image. In this paper, we propose the far distance face detection method that combine the Aadaboost of Viola-Jones with a saliency map and user's attention information. Saliency Map is used to select the candidate face images in the input image, face images are finally detected among the candidated regions using the Adaboost with Haar-like feature learned in advance. And the user's eye-tracking information is used to select the interest regions. When a subject is so far away from the camera that it is difficult to detect the face image, we expand the small eye gaze spot region using linear interpolation method and reuse that as input image and can increase the face image detection performance. We confirmed the proposed model has better results than the conventional Adaboost in terms of face image detection performance and computational time.

• Journal of Life Science
• /
• v.19 no.9
• /
• pp.1239-1244
• /
• 2009

This study was conducted to clarify the genetic relationship between Calanthe discolor, C. sieboldii and variants, and the cause of flower color variations by using a molecular linkage map and a quantitative trait loci (QTL) analysis for flower and lip color in Calanthe species native to Korea. Twenty plants were included in three C. discolor and three C. sieboldii, and fourteen variants were obtained from their habitat, Jeju-do in Korea. The flowers of C. discolor were brownish red, the values of Commission Internationale de I'Eclairage (CIE) Lab were between 40 and 50. The flowers of C. sieboldii were yellowish, the values of CIE Lab were between 110 and 130. The variants had various mixed colors that were thought to have originated from natural hybridization between C. discolor and C. sieboldii, and the values of CIE Lab were between 50 and 70. The colors of the lips were usually divided into white and yellow. C. discolor had a white lip, C. sieboldii had a yellow one, and the variants had a white to yellow one. The CIE Lab value of each color was 90 in white and 110 to 120 in yellow lips. A molecular linkage mapping was constructed based on the segregation of 154 RAPD markers using a MAPL program. Sixteen linkage groups containing 66 markers were established. It covered a total map distance of 220.4 cM. The distance between adjacent markers ranged from 0 to 6.6 cM, with an average distance of 3.3 cM. These markers are thought to be closely associated with flower and lip color expression. Among the 16 molecular linkage groups, 3 QTLs had flower color trait loci and 1 QTL had lip color trait loci.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.25 no.6_2
• /
• pp.635-643
• /
• 2007

As the availability of images from Airborne Digital Camera with high resolution is expanded, a lot of concern are in the production and update of digital map. This study presents the method of updating the digital map at the scale of 1/1,000 using images from Aerial Digital Camera. Geometric correction was completed using GPS surveying data. For digital mapping, digital photogrammetric system was utilized to digitize buildings and roads. The absolute positional accuracy was evaluated using GPS surveying data and the relative positional accuracy was evaluated using the digital map produced by analytical mapping. The absolute positional accuracy was as follows: RMSE in X and Y were ${\pm}0.172m\;and\;{\pm}0.127m$, and average distance error was 0.208m. The relative positional accuracy was as follows: RMSE in X and Y were ${\pm}0.238m\;and\;{\pm}0.281m$, and average distance error was 0.337m. Accuracies of updating digital map using images from airborne Digital Camera were within allowable error established by NGII. Consequently, images from airborne Digital Camera can be used in various fields including the production of the national basic map and the GIS of local government.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.15 no.6 s.99
• /
• pp.635-644
• /
• 2005

This paper is. a study on relation between road traffic noise(RTN) and sound power level(PWL). At present, many experimental formulae and prediction formulae are used for prediction of RTN. But these formulae are difficult to appiy to the metropolitan area because these formulae are inaccurate in the different condition from reference condition. This paper calculate RTN and PWL of each prediction formula, choose the best one and make a noise map of the subject area. Procedure is as follows. First, calculate $L_{eq}$ of RTN using experimental formulae and prediction formulae. Second, calculate PWL using $L_{eq}$ of RTN and distance attenuation for point source at semi-free field. Third, choose the most accurate formula. And finally, make a noise map of the subject area at present and future. The result using noise map will be able to apply to application field. Noise mapping tool used on this paper is Raynoise program using Ray Tracing Method(RTM), Mirror Image Source Method(MISM) and Hybrid Method(HM).

• Proceedings of the KSRS Conference
• /
• v.2
• /
• pp.717-720
• /
• 2006

In this study, we predicted locations vulnerable to ground subsidence hazard using fuzzy logic and geographic information system (GIS). Test was carried out at an abandoned underground coal mine in Samcheok City, Korea. Estimation of relative ratings of eight major factors influencing subsidence and determination of effective fuzzy operators are presented. Eight major factors causing ground subsidence were extracted and constructed as a spatial database using the spatial analysis and the probability analysis functions. The eight factors include geology, slope, landuse, depth of mined tunnel, distance from mined tunnel, RMR, permeability, and depth of ground water. A frequency ratio model was applied to calculate relative rating of each factor, and the ratings were integrated using fuzzy membership function and five different fuzzy operators to produce a ground subsidence susceptibility map. The ground subsidence susceptibility map was verified by comparing it with the existing ground subsidences. The obtained susceptibility map well agreed with the actual ground subsidence areas. Especially, ${\gamma}-operator$ and algebraic product operator were the most effective among the tested fuzzy operators.

• Electronics and Telecommunications Trends
• /
• v.35 no.5
• /
• pp.112-122
• /
• 2020

In 3D computer graphics, a depth map is an image that provides information related to the distance from the viewpoint to the subject's surface. Stereo sensors, depth cameras, and imaging systems using an active illumination system and a time-resolved detector can perform accurate depth measurements with their own light sources. The 3D image information obtained through the depth map is useful in 3D modeling, autonomous vehicle navigation, object recognition and remote gesture detection, resolution-enhanced medical images, aviation and defense technology, and robotics. In addition, the depth map information is important data used for extracting and restoring multi-view images, and extracting phase information required for digital hologram synthesis. This study is oriented toward a recent research trend in deep learning-based 3D data analysis methods and depth map information extraction technology using a convolutional neural network. Further, the study focuses on 3D image processing technology related to digital hologram and multi-view image extraction/reconstruction, which are becoming more popular as the computing power of hardware rapidly increases.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.6 no.1
• /
• pp.148-156
• /
• 2011

This paper presents an implementation of autonomous navigation of a mobile robot indoors. It explains methods for map building, localization, obstacle avoidance and path planning. Geometric map is used for localization and path planning. The localization method calculates sensor data based on the map for comparison with the real sensor data. Monte Carlo Localization(MCL) method is adopted for estimation of the robot position. For obstacle avoidance, an artificial potential field generates repulsive and attractive force to the robot. Dijkstra algorithm plans the shortest distance path from a start position to a goal point. The methods integrate into autonomous navigation method and implemented for indoor navigation. The experiments show that the proposed method works well for safe autonomous navigation.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.30 no.3
• /
• pp.323-332
• /
• 2012

An airborne LiDAR system performs several observations on flight routes to collect data of targeted regions accompanying with discrepancies between the collected data strips of adjacent routes. This paper aims to present an automatic error correction technique using modified ICP as a way to remove relative errors from the observed data of strip data between flight routes and to make absolute correction to the control data. A control point data from the existing digital topographic map were created and the modified ICP algorithm was applied to perform the absolute automated correction on the relatively adjusted airborne LiDAR data. Through such process we were able to improve the absolute accuracy between strips within the average point distance of airborne LiDAR data and verified the possibility of automation in the geometric corrections using a large scale digital map.