• Journal of the Korea Computer Graphics Society
• /
• v.8 no.4
• /
• pp.9-16
• /
• 2002

In this paper we propose an extended billboarding method for rendering trees. It's difficult circumstance that PC renders forest by general polygon rendering in real-time, because tree object consists of tremendous leaves and limbs. Our method re-creates an appropriate texture image by 3D image warping equation each frame, and we use it for a billboarding method. For speed up, we use warping method with a image pyramid and image caching.

• ETRI Journal
• /
• v.27 no.6
• /
• pp.708-712
• /
• 2005

We propose a computational reconstruction technique in large-depth integral imaging where the elemental images have information of three-dimensional objects through real and virtual image fields. In the proposed technique, we reconstruct full volume information from the elemental images through both real and virtual image fields. Here, we use uniform mappings of elemental images with the size of the lenslet regardless of the distance between the lenslet array and reconstruction image plane. To show the feasibility of the proposed reconstruction technique, we perform preliminary experiments and present experimental results.

• Proceedings of the KIEE Conference
• /
• 1995.07b
• /
• pp.974-976
• /
• 1995

Segmentation of image data is an important problem in computer vision, remote sensing, and image analysis. Most objects in the real world have textured surfaces. Segmentation based on texture information is possible even if there are no apparent intensity edges between the different regions. There are many existing methods for texture segmentation and classification, based on different types of statistics that can be obtained from the gray-level images. In this paper, we use a neural network model --- ART-2 (Adaptive Resonance Theory) for textures in an image, proposed by Carpenter and Grossberg. In our experiments, we use Walsh matrix as feature value for textured image.

• Proceedings of the IEEK Conference
• /
• 2000.07b
• /
• pp.1088-1091
• /
• 2000

Superquadrics can represent various and complex 3D objects with only some parameters(size, position, deformation etc.). So if we use both superquadrics and deformed superquadrics, we can also represent more realistic 3D objects which are existed in real world. In this paper we use the z-buffer algorithm and stencil buffer together because this is very useful when the superquadric primitives are combined. The fundamental ideas are illustrated with a number of tables and figures.

• Korean Journal of Remote Sensing
• /
• v.23 no.1
• /
• pp.15-20
• /
• 2007

This study is to apply QuickBird satellite image for the simulation of storm runoff in a small rural watershed. For a $1.05km^2$ watershed located in Goesan-Gun of Chungbuk Province, the land use from the QuickBird image was produced by on-screening digitising after ortho-rectifying using 2 m DEM. For 3 cases of land use, soil and elevation scale (1:5,000, 1:25,000 and 1:50,000), SCS-CN and the watershed physical parameters were prepared for the storm runoff model, HEC-HMS (Hydrologic Modelling System). The model was evaluated for each case and compared the simulated results with couple of selected storm events.

• Journal of IKEEE
• /
• v.23 no.2
• /
• pp.395-402
• /
• 2019

In this paper, we implemented an Ultra-Wideband radar system using Stripmap Synthetic Apertrure Radar algorithm to recognize objects inside a box. Different window functions such as Hanning, Hamming, Kaiser, and Taylor functions to improve image recognition performance are applied and implemented to radar system. The Ultra-Wideband radar system with 3.1~4.8 GHz broadband and UWB antenna were implemented to recognize the conductor plate located inside 1m3 box. To obtain the image, we use the propagation data in the time domain according to the 1m movement distance and use the Range Doppler algorithm. The effect of different window functions to improve the recognition performance of the image are analyzed. From the compared results, we confirmed that the Kaiser window function can obtain a relatively good image.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.2
• /
• pp.1064-1079
• /
• 2019

Information security has become increasingly important with the rapid development of mobile devices and internet. An efficient encryption system is a key to this end. In this paper, we propose an image encryption method based on the cross diffusion of two chaotic maps. We use two chaotic sequences, namely the Logistic map and the Chebyshev map, for key generation which has larger security key space than single one. Moreover, we use these two sequences for further image encryption diffusion which decreases the correlation of neighboring pixels significantly. We conduct extensive experiments on several well-known images like Lena, Baboon, Koala, etc. Experimental results show that our algorithm has the characteristics of large key space, fast, robust to statistic attack, etc.

• Journal of Information Processing Systems
• /
• v.17 no.3
• /
• pp.453-461
• /
• 2021

Many deep learning approaches have been studied for image classification in computer vision. However, there are not enough data to generate accurate models in medical fields, and many datasets are not annotated. This study presents a new method that can use both unlabeled and labeled data. The proposed method is applied to classify cervix images into normal versus cancerous, and we demonstrate the results. First, we use a patch self-supervised learning for training the global context of the image using an unlabeled image dataset. Second, we generate a classifier model by using the transferred knowledge from self-supervised learning. We also apply attention learning to capture the local features of the image. The combined method provides better performance than state-of-the-art approaches in accuracy and sensitivity.

• Journal of Korea Multimedia Society
• /
• v.19 no.1
• /
• pp.10-21
• /
• 2016

Images taken from outdoor are degraded quality by fog or haze, etc. In this paper, we propose a method that provides the visibility improved images through fog or haze removal. We proposed haze removal method that uses brightness transform based on atmospheric scatter coefficient rate with local histogram equalization. To calculate the transmission rate that indicate fog rate in original image, we use atmospheric scatter coefficient rate based on quadratic equations about haze model. And primary brightness transformed image can be obtained by using the obtained transmission rate. Also we use local histogram equalization with proposed brightness transform for effectively image visibility enhancement. Unlike existing methods, our method can process real-time with stable and effect image visibility enhancement. Proposed method use only the luminance images processed by good performance surveillance systems because it represents the real-time processing is required, black-box, digital camera and multimedia equipment is applicable. Also because it shows good performance only with the luminance images processed, Surveillance systems, black boxes, digital cameras, and multimedia devices etc, that require real-time processing can be applied.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2003.10a
• /
• pp.999-1002
• /
• 2003

In this study, embodied supervisory system that apply motion detection technique to small web camera and detects watch picture. Motion detection technique that use pixel value of car image that use in existing need memory to store background image. Also, there is sensitive shortcoming at increase of execution time by data process of pixel unit and noise. Suggested technique that compare extracting motion information by block unit to do to have complexion that solve this shortcoming and is strong at noise. Because motion information by block compares block characteristic value of image without need frame memory, store characteristic cost by block of image. Also, can get effect that reduce influence about noise and is less sensitive to flicker etc.. of camera more than motion detection that use pixel value in process that find characteristic value by block unit.