• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.3 no.1
• /
• pp.38-46
• /
• 2010

This paper describes a computer vision system based on active IR illumination for real-time gaze discrimination system. Unlike most of the existing gaze discrimination techniques, which often require assuming a static head to work well and require a cumbersome calibration process for each person, our gaze discrimination system can perform robust and accurate gaze estimation without calibration and under rather significant head movement. This is made possible by a new gaze calibration procedure that identifies the mapping from pupil parameters to screen coordinates using generalized regression neural networks (GRNNs). With GRNNs, the mapping does not have to be an analytical function and head movement is explicitly accounted for by the gaze mapping function. Furthermore, the mapping function can generalize to other individuals not used in the training. To further improve the gaze estimation accuracy, we employ a reclassification scheme that deals with the classes that tend to be misclassified. This leads to a 10% improvement in classification error. The angular gaze accuracy is about $5^{\circ}$horizontally and $8^{\circ}$vertically. The effectiveness of our gaze tracker is demonstrated by experiments that involve gaze-contingent interactive graphic display.

• Journal of Broadcast Engineering
• /
• v.15 no.6
• /
• pp.757-767
• /
• 2010

In this paper, we have studied the color compensation method for 3D that enables 3D color presentation similar to 2D. The color compensation method uses the difference of color presentation in 2D and 3D mode. First, the RGB I/O relationship curve was derived in 2D and 3D mode based on the input RGB color bar images. The relationship was modeled in modified power-law forms. Based on the modeling information, we generated color mapping tables, which can be used for compensating the difference of colors. The proposed color mapping block can be added at the output block of a 3DTV system, where the 2D content can be bypassed but the 3D content RGB data can be processed using the color mapping table. The experimental results show that the proposed method improves color presentation of a 3DTV system using a proper color compensation based on 2D presentation.

• Journal of Intelligence and Information Systems
• /
• v.12 no.2
• /
• pp.145-165
• /
• 2006

Information filtering systems, which are designed fur users' needs, do not satisfy user's diverse requests as their filtering accuracy is unstable sometimes. This study suggests an information filtering system based on cognitive brain mapping by simulating the processes of information in human brain. Compared to traditional filtering systems, which use specific words or pattern in their filtering systems, the method suggested in this article uses both key words and relationships among these words. The significance of this study is on simulating information storing processes in human brain by mapping both key words and their relationships among them together. To combine these two methods, this study finds balances in representing two methods by searching optimal weights of each of them.

• The KIPS Transactions:PartA
• /
• v.11A no.5
• /
• pp.303-312
• /
• 2004

In this paper, we propose a methodology solving the problem of the hardware-software partitioning in reconfigurable systems using a Y-chart design space exploration and implement a simulator according to the methodology. The methodology generates a mapping set between tasks and hardware elements using the hardware element model and the application model. We evaluate the throughput by simulating cases in each mapping set. With the throughput evaluation result, we can select the mapping case with the highest throughput. We also propose an heuristic improving the simulation time by reducing the mapping set on the basis of the relationship between workload and parallelism. Simulation results show that we can reduce the size of mapping set which poses difficulties on hardware-software partitioning by up to 80%.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.36 no.5
• /
• pp.423-432
• /
• 2018

Surface water mapping has been widely used in various remote sensing applications. Water indices have been commonly used to distinguish water bodies from land; however, determining the optimal threshold and discriminating water bodies from similar objects such as shadows and snow is difficult. Deep learning algorithms have greatly advanced image segmentation and classification. In particular, FCN (Fully Convolutional Network) is state-of-the-art in per-pixel image segmentation and are used in most benchmarks such as PASCAL VOC2012 and Microsoft COCO (Common Objects in Context). However, these data sets are designed for daily scenarios and a few studies have conducted on applications of FCN using large scale remotely sensed data set. This paper aims to fine-tune the pre-trained FCN network using the CRMS (Coastwide Reference Monitoring System) data set for surface water mapping. The CRMS provides color infrared aerial photos and ground truth maps for the monitoring and restoration of wetlands in Louisiana, USA. To effectively learn the characteristics of surface water, we used pre-trained the DeepWaterMap network, which classifies water, land, snow, ice, clouds, and shadows using Landsat satellite images. Furthermore, the DeepWaterMap network was fine-tuned for the CRMS data set using two classes: water and land. The fine-tuned network finally classifies surface water without any additional learning process. The experimental results show that the proposed method enables high-quality surface mapping from CRMS data set and show the suitability of pre-trained FCN networks using remote sensing data for surface water mapping.

• Journal of Korea Multimedia Society
• /
• v.9 no.7
• /
• pp.782-800
• /
• 2006

Image metadata is used to describe the contents, structures, and features of a given image data. There exist lots of image metadata-related standards. Such image metadata need to be shared to efficiently search the correct images in Web Service-enabled distributed environments. In this thesis, we propose an interoperable system architecture for utilizing image metadata and query extension methods using metadata mapping for correct and efficient image searching. In the proposed methods, the provider schema using non-standard image metadata are mapped into the schema based on the VRA standard, which is used as a basic and intermediary standard in MDR(MetaData Registry). We support mapping operations between the VRA standard and other standards, to reduce the number of unnecessary mapping operations and to enable easier addition of schema based on new standards. Using the proposed query processing methods, which utilize image metadata and overcome their heterogeneity by metadata mapping, it is possible to severely improve the correctness of image retrieval. To show the usefulness of our schemes, we implemented a prototype system and performed retrieval performance evaluation on extended query processing.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.1215-1220
• /
• 2003

People's Republic of China is one of the most rapidly developing countries in the world today. There is a great demand on highly actual and accurate spatial information of the whole country, especially of West China which becomes the focus of development of the Chinese government right now and in the next years, but where still not enough topographic maps are available. This raises great challenges to the surveying and mapping community in China. Facing the new challenges the Chinese Academy of Surveying and Mapping (CASM) started its pioneer work early 2002 to explore new techniques and technologies available today toward increasing the map productivity. With import of a CCNS/AEROcontrol system in November 2002 the first DGPS/IMU-based photogrammetric project in China was successfully accomplished jointly by CASM and the Germany-based companies IGI and Techedge. Two photogrammetric blocks of 1:4,000 and 1:20,000 photo scale, respectively, were flown in Anyang, China. Direct georeferencing and integrated sensor orientation were conducted. Results achieved were proven by using ground check points and compared with those of aerial triangulation. Orthophotos generated based on direct georeferencing shows the high efficiency and quality, and thus proved the promise of the new technology. Furthermore several DGPS/IMU-based photogammetric projects was accomplished one by one and a big project of more than 100,000 km2 in the Inner Mongolia will be started in August 2003. The paper presents experiences with DGPS/IMU-based photogrammetry in China. Results achieved in concrete projects are shown and evaluated. Politic and technical specialties in China are discussed. Conclusions outline the potential of DGPS/IMU-based photogrammetric production in China.

• Journal of the Korean Society of Industry Convergence
• /
• v.23 no.6_2
• /
• pp.1043-1050
• /
• 2020

In this study, a study was presented on the monitoring technique of landslide area using UAV. In the case of disaster investigation using drone mapping, it can be used at various disaster sites. The mission can be carried out at various disaster sites, including surveys of damage to mountainous areas caused by landslides, building collapses surveys of flood damage, typhoons, earthquakes. The damage investigation plan using drone mapping is expected to be highly utilized at disaster sites where investigators cannot access it like in mountainous areas and where it is difficult to conduct direct damage investigations at the site. Drone mapping technology has many advantages in terms of disaster follow-up, such as recovery. Compared to the existing survey system, which was mainly carried out manually, the investigation time can be drastically reduced, and it can also respond to disaster sites that are difficult to carry out or are difficult to access directly. In addition, it is possible to establish and guide spatial data at the disaster site based on accurate mapping data from the time of the disaster, which has considerable strength in managing the situation of the disaster site, selecting priority areas for recovery, and establishing recovery plans. As such, drone mapping is a technology that can be used in a wide range of sites along with natural disasters and social disasters. If a damage investigation system is established through this, it is believed that it will contribute significantly to the rapid establishment of recovery plans along with the investigation of disaster response time and extent of damage recovery.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.17 no.4
• /
• pp.373-381
• /
• 1999

Mobile Mapping System can be defined as vehicle mapping system which collects rapidly spatial data by integrated Gps/digital imaging system. Kinematic positioning by GPS is essential technology of Mobile Mapping System. This paper aims at analysing the accuracy and efficiency of kinematic positioning by GPS platformed on moving vehicle. For the purpose, roads were surveyed by vehicle/kinematic GPS. The results show that vehicle/kinematic GPS can measure spatial position faster, and still maintain a reasonable accuracy. But inertial navigation system and GPS should be integrated to compute continuous vehicle track and overcome gaps by blocked satellite signals for the more accurate positioning.

• Spatial Information Research
• /
• v.14 no.2 s.37
• /
• pp.235-244
• /
• 2006

The MMS(Mobile Mapping System) using the vehicle equipped GPS, IMU and CCD Cameras is the effective system for the management of the road facilities, update of the digital map, and etc. The image, vehicle's 3 dimensional position and attitude information provided MMS is a important source for positioning objects included the image. In this research we applied the tracking technique to the specific object in image. The extraction of important object from immense MMS data makes more effectiveness in this system.