• Korean Journal of Remote Sensing
• /
• v.13 no.1
• /
• pp.57-73
• /
• 1997

This paper is related to the correction of radiometric distortions induced by topographic relief. RADARSAT SAR image data were obtained over the mountainous area near southern part of Seoul. Initially, the SAR data was geometrically corrected and registered to plane rectangular coordinates so that each pixel of the SAR image has known topographic parameters. The topographic parameters (slope and aspect) at each pixel position were calculated from the digital elevation model (DEM) data having a comparable spatial resolution with the SAR data. Local incidence angle between the incoming microwave and the surface normal to terrain slope was selected as a primary geometric factor to analyze and to correct the radiometric distortions. Using digital maps of forest stands, several fields of rather homogeneous forest stands were delineated over the SAR image. Once the effects of local incidence angle on the radar backscatter were defined, the radiometric correction was performed by an empirical fuction that was derived from the relationship between the geometric parameters and mean radar backscatter. The correction effects were examined by ground truth data.

• Journal of the Korea Computer Graphics Society
• /
• v.25 no.3
• /
• pp.123-131
• /
• 2019

In this study, we modified CT images of femoral head in consideration of anatomically meaningful structure, proposing the method to augment the training data of convolution Neural network for segmentation of femur mesh model. First, the femur mesh model is obtained from the CT image. Then divide the mesh model into meaningful parts by using cluster analysis on geometric characteristic of mesh surface. Finally, transform the segments by using an appropriate mesh deformation algorithm, then create new CT images by warping CT images accordingly. Deep learning models using the data enhancement methods of this study show better image division performance compared to data augmentation methods which have been commonly used, such as geometric conversion or color conversion.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.4
• /
• pp.157-163
• /
• 2022

Recently, with the development of automation in the industrial field, research on anomaly detection is being actively conducted. An application for anomaly detection used in factory automation is camera-based defect inspection. Vision camera inspection shows high performance and efficiency in factory automation, but it is difficult to overcome the instability of lighting and environmental conditions. Although camera inspection using deep learning can solve the problem of vision camera inspection with much higher performance, it is difficult to apply to actual industrial fields because it requires a huge amount of normal and abnormal data for learning. Therefore, in this study, we propose a network that overcomes the problem of collecting abnormal data with 72 geometric transformation deep learning methods using only normal data and adds an outlier exposure method for performance improvement. By applying and verifying this to the MVTec data set, which is a database for auto-mobile parts data and outlier detection, it is shown that it can be applied in actual industrial sites.

• Journal of Korean Society for Geospatial Information Science
• /
• v.24 no.4
• /
• pp.115-123
• /
• 2016

The road name address system has been in effect in Korea since 2012. However, the existing address system is still being used in many fields because of the difference between the spatial awareness of people and the road name address system. For the spatial awareness based on the road name address system, various spatial datasets in daily life should be referenced by the road names. The goal of this paper is to link the road name address system with the wayfinding process, which is closely related to the spatial awareness. To achieve our goal, we designed and implemented a geometric matching method for spatial data sets. This method generates network neighborhoods from road objects in the 'road name address map' and the 'pedestrian network data'. Then it computes the geometric similarities between the neighborhoods to identify corresponding road name for each object in the network data. The performance by F0.5 was assessed at 0.936 and it was improved to 0.978 by the manual check for 10% of the test data selected by the similarity. By help of our method, the road name address system can be utilized in the wayfinding services, and further in the spatial awareness of people.

• The KIPS Transactions:PartB
• /
• v.11B no.4
• /
• pp.403-410
• /
• 2004

Recently, studies of image analysis, as the preprocessing stage for medical image analysis or image retrieval, are actively carried out. This paper intends to propose a way of utilizing color components for image retrieval. For image retrieval, it is based on color components, and for analysis of color, CLCM (Color Level Co-occurrence Matrix) and statistical techniques are used. CLCM proposed in this paper is to project color components on 3D space through geometric rotate transform and then, to interpret distribution that is made from the spatial relationship. CLCM is 2D histogram that is made in color model, which is created through geometric rotate transform of a color model. In order to analyze it, a statistical technique is used. Like CLCM, GLCM (Gray Level Co-occurrence Matrix)[1] and Invariant Moment [2,3] use 2D distribution chart, which use basic statistical techniques in order to interpret 2D data. However, even though GLCM and Invariant Moment are optimized in each domain, it is impossible to perfectly interpret irregular data available on the spatial coordinates. That is, GLCM and Invariant Moment use only the basic statistical techniques so reliability of the extracted features is low. In order to interpret the spatial relationship and weight of data, this study has used Principal Component Analysis [4,5] that is used in multivariate statistics. In order to increase accuracy of data, it has proposed a way to project color components on 3D space, to rotate it and then, to extract features of data from all angles.

• Journal of Broadcast Engineering
• /
• v.26 no.5
• /
• pp.553-565
• /
• 2021

Point cloud is a set of points for representing a 3D object, and consists of geometric information, which is 3D coordinate information, and attribute information, which is information representing color, reflectance, and the like. In this way of expressing, it has a vast amount of data compared to 2D images. Therefore, a process of compressing the point cloud data in order to transmit the point cloud data or use it in various fields is required. Unlike color information corresponding to all 2D geometric information constituting a 2D image, a point cloud represents a point cloud including attribute information such as color in only a part of the 3D space. Therefore, separate processing of geometric information is also required. Based on these characteristics of point clouds, MPEG under ISO/IEC standardizes V-PCC, which imitates point cloud images and compresses them into 2D DCT-based 2D image compression codecs, as a compression method for high-density point cloud data. This has limitations in accurately representing 3D spatial information to proceed with compression by converting 3D point clouds to 2D, and difficulty in processing non-existent points when utilizing 3D DCT. Therefore, in this paper, we present 3D Discrete Cosine Transform-based Point Cloud Compression (3DCT PCC), a method to compress point cloud data, which is a 3D image by utilizing 3D DCT, and confirm the efficiency of 3D DCT compared to V-PCC based on 2D DCT.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.67 no.5
• /
• pp.669-674
• /
• 2018

Data augmentation is an efficient way to reduce overfitting on models and to improve a performance supplementing extra data for training. It is more important in deep learning based industrial machine vision. Because deep learning requires huge scale of learning data to learn a model, but acquisition of data can be limited in most of industrial applications. A very generic method for augmenting image data is to perform geometric transformations, such as cropping, rotating, translating and adjusting brightness of the image. The effectiveness of data augmentation in image classification has been reported, but it is rare in defect inspections. We explore and compare various basic augmenting operations for the metal surface defects. The experiments were executed for various types of defects and different CNN networks and analysed for performance improvements by the data augmentations.

• Proceedings of the KSRS Conference
• /
• 2008.10a
• /
• pp.406-409
• /
• 2008

Clustering data streams has an importance over many applications like sensor networks. Existing hierarchical methods follow a semi fuzzy clustering that yields duplicate clusters. In order to solve the problems, we propose an extended online divisive agglomerative clustering on data streams. It builds a tree-like top-down hierarchy of clusters that evolves with data streams using geometric time frame for snapshots. It is an enhancement of the Online Divisive Agglomerative Clustering (ODAC) with a pruning strategy to avoid duplicate clusters. Our main features are providing update time and memory space which is independent of the number of examples on data streams. It can be utilized for clustering sensor data and network monitoring as well as web click streams.

• Journal of Fisheries and Marine Sciences Education
• /
• v.17 no.3
• /
• pp.427-433
• /
• 2005

The study of Web GUI(Graphic User Information) system for Virtual Reality System is mainly performed on effective methodology which transform real world data to computing world data. MGIS(Marine Geographic Information System) has its own target on reliable data service by acquisition of geometric information using accurate measurement and graphical visualization. This type of raw data visualization can be built without software tools, yet is incredibly useful for interpreting and communicating data. Even simple visualizations can aid in the interpretation of complex 3D relationships that are frequently encountered in the geosciences. The Virtual Reality Modeling Language provides an easy way for geoscientists to construct complex visualizations that can be viewed with free software. This study propose a three dimensional Web GUI system using MGIS-based three dimensional data models and virtual imaging system. Finally, we design a Web GUI system integrating above data models.

• IE interfaces
• /
• v.13 no.4
• /
• pp.646-657
• /
• 2000

We developed an efficient small sized weather condition forecasting system (WFS). A cheap NT-server was utilized for handling a large amount of data, while traditional WFS has conventionally relied on Unix based workstation server. The proposed WFS contains automatic weather observing system (AWS). AWS was designed for collecting weather conditions automatically, and it was linked to WFS in order to provide various weather condition information. The existing two phase scheme and chain code algorithm were used for transforming AWS's data into WFS's data. The WFS's data were mapped into geometric information system using various display techniques. Finally the transformed WFS's data was also converted into JPG (Joint Photographic Group) data type, and the final JPG data could be accessible by others though Internet. The developed system was implemented using WWW environment and has provided weather condition forecasting information. Real case is given to show the presented integrated WFS with detail information.