• Journal of Korea Multimedia Society
• /
• v.20 no.8
• /
• pp.1271-1281
• /
• 2017

It is essential to collect and analyze target information around the vehicle for autonomous driving of the vehicle. Based on the analysis, environmental information such as location and direction should be analyzed in real time to control the vehicle. In particular, obstruction or cutting of objects in the image must be handled to provide accurate information about the vehicle environment and to facilitate safe operation. In this paper, we propose a method to simultaneously generate 2D and 3D bounding box proposals using LiDAR Edge generated by filtering LiDAR sensor information. We classify the classes of each proposal by connecting them with Region-based Fully-Covolutional Networks (R-FCN), which is an object classifier based on Deep Learning, which uses two-dimensional images as inputs. Each 3D box is rearranged by using the class label and the subcategory information of each class to finally complete the 3D bounding box corresponding to the object. Because 3D bounding boxes are created in 3D space, object information such as space coordinates and object size can be obtained at once, and 2D bounding boxes associated with 3D boxes do not have problems such as occlusion.

• Journal of Korea Society of Industrial Information Systems
• /
• v.17 no.7
• /
• pp.95-100
• /
• 2012

Currently, crimes are becoming increasingly dense, the technique is also very tricky. Pickpocket, the theft of crime, is occurred at the most crowded or congested places. However, nowadays occurs more commonly at the rare and secluded place in the human. In this paper, we investigate the techniques and types of pickpockets. And realistically to submit classified video, to standing, sitting, and lying to the classification. Target for pickpockets, we classify it to submit a video forensic evidence. This paper will be needed in order to prevent and cope with pickpockets crimes.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.07b
• /
• pp.673-675
• /
• 2005

The art psychotherapy expert system is a computer system which helps to analyse one's psychology through pictures. However we need a standard criterion because the psychology, the target of the art psychotherapy, does not only have a ambiguous criterion but also a vast range. We're going to suggest a criterion in the field of the art psychotherapy by constructing systematic database through knowledge acquirement of the art psychotherapy expert system. In this study we introduce a system which enables systematic classification and confirmation of symptoms according to mental analyses. The suggested system enables confirmation of a classical structure and systematic classification of knowledges through conversation by extracting nouns through sentence analysis from the knowledge of descriptive form based on the clinical purpose of sentence analysis.

• The KIPS Transactions:PartB
• /
• v.13B no.1 s.104
• /
• pp.1-6
• /
• 2006

The prediction of protein function basically make use of a protein-protein interaction map based on the concept of guilt-by-association. The method however cannot determine the functions of proteins in case that the target protein does not interact with proteins with known functions directly. This paper studies protein function prediction considering the given problem as a K-class classification problem and proposes a predictive approach utilizing a modular neural network. The proposed method uses interaction data and protein related attributes as well. The experimental results demonstrate that the proposed approach can predict the functional roles of Yeast proteins whose interaction knowledge is not known and shows better performance than the graph-based models that use protein interaction data.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.16 no.3 s.94
• /
• pp.335-343
• /
• 2005

In this paper, we propose a method to classify inverse synthetic aperture radar(ISAR) image from different target. The approach can provide efficient features for classification by the combined use of a polar mapping procedure and a well-designed classifier The resulting feature vectors are able to meet requirements that efficient features should have : invariance with respect to rotation and scale, small dimensionality, as well as highly discriminative information. Typical experimental examples of the proposed method are provided and discussed.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.48 no.2
• /
• pp.207-220
• /
• 2015

This study identified the species-specific, frequency-dependent characteristics of broadband acoustic scattering that facilitate classifying fish species using the pulse compression (PC) technique. Controlled acoustic scattering laboratory experiments were conducted with nine commercially important fish species using linear chirp signals (95-220 kHz) over an orientation angle range of ${\pm}45^{\circ}$ in the dorsal plane at approximately $1^{\circ}$ increments. The results suggest that the angular-dependent characteristics of the broadband echoes and the frequency-dependent variability in target strength (TS) were useful for inferring the fish species of interest. The scattering patterns in the compressed pulse output were extremely complex due to morphological differences among fish species, but the x-ray images strongly suggested that spatial separation correlated well with scattering for the head, skeleton, bone, otoliths, and swim bladder within each specimen.

• Ecology and Resilient Infrastructure
• /
• v.7 no.3
• /
• pp.199-207
• /
• 2020

Against the backdrop of frequent weather disasters such as floods, droughts, and heat waves worldwide, urban parks should provide functions for the safety of urban residents as well as rest, culture, and ecological functions. In this study, a classification system for urban disaster prevention parks is proposed for the safety of the urbanites with the aim of securing a complex function in a green space in response to climate changes in the city. Analytical indicators were extracted through literature research, and the classification system was verified through on-site surveys of the target sites and interviews with those involved. The large class for evaluation was divided into three types: location, spatial composition, and disaster prevention complex facilities of urban parks; the direction of improvement was proposed for problems identified through empirical analysis.

• Journal of the Korean Institute of Educational Facilities
• /
• v.8 no.2
• /
• pp.27-40
• /
• 2001

The opinion on merits and defects caused the unified-closed of a school facilities have collected by a people and government since the unified-closed system was enforced. There is a problem that the data occurring in the middle of the unified-closed is not analized. Therefore, in this study the target is elementary schools in kyung-gi province which did the unified-closed. First, this study is for analysing which condition influenced to make them. Second, it is for understanding the problems occurred in the process of the unified-closed and presenting the direction of improvement. Third, it is for setting up the planned guide by classification of school facility for the unified-closed which considers the previous problems sufficiently. The range of the study treats of general trend and analysis about the unified-closed of elementary school. The core subject is to analyse the standard and classification of the unified-closed of the object school by the policy of the unified-closed and to compare the carried process of the case school.

• Journal of Korean Society of Rural Planning
• /
• v.23 no.3
• /
• pp.21-36
• /
• 2017

Climate change is intensifying storms and floods around the world. Where nature has been destroyed by development, communities are at risk from these intensified climate patterns. This study was to suggest a methodology for estimating flood vulnerability using Potential Flood Damage(PFD) concept and classify city/county about Potential Flood Damage(PFD) using various typology techniques. To evaluate the PFD at a spatial resolutions of city/county units, the 20 representative evaluation indexing factors were carefully selected for the three categories such as damage target(FDT), damage potential(FDP) and prevention ability(FPA). The three flood vulnerability indices of FDT, FDP and FPA were applied for the 167 cities and counties in Korea for the pattern classification of potential flood damage. Potential Flood Damage(PFD) was classified by using grouping analysis, decision tree analysis, and cluster analysis, and characteristics of each type were analyzed. It is expected that the suggested PFD can be utilized as the useful flood vulnerability index for more rational and practical risk management plans against flood damage.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2003.05b
• /
• pp.62-65
• /
• 2003

When a mobile robot moves around autonomously without man-made corrupted bye landmarks, it is essential to recognize the placement of surrounding objects especially for self localization, obstacle avoidance, and target classification and localization. To recognize the environment we use many Kinds of sensors, such as ultrasonic sensors, laser range finder, CCD camera, and so on. Among the sensors, ultra sonic sensors(sonar)are unexpensive and easy to use. In this paper, we analyze the sonar data and propose a method to recognize features of indoor environment. It is supposed that the environments are consisted of features of planes, edges, and corners, For the analysis, sonar data of plane, edge, and corner are accumulated for several given ranges. The data are filtered to eliminate some noise using the Kalman filter algorithm. Then, the data for each feature are compared each other to extract the character is ties of each feature. We demonstrate the applicability of the proposed method using the sonar data obtained form a sonar transducer rotating and scanning the range information around a indoor environment.