• Journal of Broadcast Engineering
• /
• v.23 no.2
• /
• pp.316-319
• /
• 2018

When a terrestrial UHD broadcasting service and a mobile HD broadcasting service are provided using the PLP function provided by ATSC 3.0 and domestic UHD broadcasting standard, a small amount of data may be additionally transmitted to further provide high quality UHD-3D broadcasting service. The left and right images of the stereoscopic image are input, one view image is encoded by the SHVC method, and the other view images are encoded by the SHVC method of the two-view cross-referencing method. However, since the base layers (BL) of the two encoders are mutually common, the two encoders correspond to encoders that generate one BL stream and two enhancement layer (EL) streams. The average encoding efficiency is 16% more efficient compared to the third independent HEVC encoding for the UHD-3D broadcast service. The proposed scheme reduces the fluctuation of PSNR per image frame and increases the image quality of minimum PSNR frame by 0.6dB.

• Journal of the Institute of Convergence Signal Processing
• /
• v.14 no.1
• /
• pp.27-32
• /
• 2013

It plays an important role to detect the shape of an ellipse in many application areas of image processing. But it is very difficult to detect the ellipse in the real image because the noise was involved in the image, other objects obscured the ellipse or the ellipses were overlap with each other. In this paper, we extract the boundary (edge) to detect ellipse in the image and perform the grouping process in order to reduce amount of information. As a result, the speed of the ellipse detection was improved. Also in order to the ellipse detection, we selected the five ellipse parameters at random And then to select the optimal parameters of the ellipse, the linear least-squares approximation is applied. To verify the ellipse detection, RANSAC algorithm is applied. After the algorithm proposed in this study was implemented, the results applied to the real images showed an aocuracy of 75% and speed was very fast to compared with other researches. It mean that the proposed algorithm was valuable to detect the ellipses in the image.

• Journal of the Institute of Convergence Signal Processing
• /
• v.11 no.3
• /
• pp.183-188
• /
• 2010

In this paper we propose a detection method of the medial axis of the continuous stripes on the LoG scale-space. Our method detects the medial axis of continuous stripes iteratively by varying the scale of LoG operator. Small-scale LoG operator detects two +/- pole pairs centered on the edge positions of stripe by the zero-crossing detection. The more increase the scale of LoG scale-space, the more close two poles to the medial axis of stripe. The medial axis of continuous stripe is the position where two poles is overlapped. The proposed method detected robustly the medial axis of continuous stripes stronger than the thinning methods used to binary image.

• Korean Journal of Remote Sensing
• /
• v.36 no.6_2
• /
• pp.1523-1535
• /
• 2020

In order to produce crop information using remote sensing, we use classification and growth monitoring based on crop phenology. Therefore, time-series satellite images with a short period are required. However, there are limitations to acquiring time-series satellite data, so it is necessary to use fusion with other earth observation satellites. Before fusion of various satellite image data, it is necessary to overcome the inherent difference in radiometric characteristics of satellites. This study performed Korea Multi-Purpose Satellite-3 (KOMPSAT-3) cross calibration with Landsat-8 as the first step for fusion. Top of Atmosphere (TOA) Reflectance was compared by applying Spectral Band Adjustment Factor (SBAF) to each satellite using hyperspectral sensor band aggregation. As a result of cross calibration, KOMPSAT-3 and Landsat-8 satellites showed a difference in reflectance of less than 4% in Blue, Green, and Red bands, and 6% in NIR bands. KOMPSAT-3, without on-board calibrator, idicate lower radiometric stability compared to ladnsat-8. In the future, efforts are needed to produce normalized reflectance data through BRDF (Bidirectional reflectance distribution function) correction and SBAF application for spectral characteristics of agricultural land.

• Journal of the Korea Society of Computer and Information
• /
• v.15 no.10
• /
• pp.35-47
• /
• 2010

To complement inaccurate edge information and detect correctly the boundary of a vehicle in an image, an extended edge analysis technique is presented in this paper. The vehicle is detected using the bottom boundary generated by a vehicle and the road surface and the left and right side boundaries of the vehicle. The proposed extended edge analysis method extracts the horizontal edge by merging or dividing the nearby edges inside the region of interest set beforehand because various noises deteriorates the horizontal edge which can be a bottom boundary. The horizontal edge is considered as the bottom boundary and the vertical edges as the side boundaries of a vehicle if the extracted horizontal edge intersects with two vertical edges which satisfy the vehicle width condition at the height of the horizontal edge. This proposed algorithm is more efficient than the other existing methods when the road surface is complex. It is proved by the experiments executed on the roads having various backgrounds.

• Journal of Advanced Navigation Technology
• /
• v.23 no.5
• /
• pp.400-407
• /
• 2019

In this paper, the shadow detection and reconstruction method are proposed using intrinsic image, which does not change the essential characteristics under the influence of various illuminance, and multi-scale gamma correction. The shadow detection was estimated by the pixel change information between a grayscale and an intrinsic image of the color image, and the brightness of the image were adjusted by gamma correction in the shadow restoration process. Multi-scale gamma correction is performed for each channel of a color image due to the fact that the saturation can be changed by nonlinear adjustment to individual pixel values. Multi-scale gamma values are estimated based on the information of the crossed edge between shadows and non-shadowed regions in the color image, as a result, the shadows are reconstructed by correcting different region features with multi-scale gamma values. Experimental results show that the proposed method effectively reconstructs shadows in a single natural image.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.12
• /
• pp.505-518
• /
• 2023

3D point cloud semantic segmentation is a computer vision task that involves dividing the point cloud into different objects and regions by predicting the class label of each point. Existing 3D semantic segmentation models have some limitations in performing sufficient fusion of multi-modal features while ensuring both characteristics of 2D visual features extracted from RGB images and 3D geometric features extracted from point cloud. Therefore, in this paper, we propose MMCA-Net, a novel 3D semantic segmentation model using 2D-3D multi-modal features. The proposed model effectively fuses two heterogeneous 2D visual features and 3D geometric features by using an intermediate fusion strategy and a multi-modal cross attention-based fusion operation. Also, the proposed model extracts context-rich 3D geometric features from input point cloud consisting of irregularly distributed points by adopting PTv2 as 3D geometric encoder. In this paper, we conducted both quantitative and qualitative experiments with the benchmark dataset, ScanNetv2 in order to analyze the performance of the proposed model. In terms of the metric mIoU, the proposed model showed a 9.2% performance improvement over the PTv2 model using only 3D geometric features, and a 12.12% performance improvement over the MVPNet model using 2D-3D multi-modal features. As a result, we proved the effectiveness and usefulness of the proposed model.

• Journal of Broadcast Engineering
• /
• v.27 no.4
• /
• pp.569-580
• /
• 2022

In this study, we investigated the effect of peripheral blur on binocular fusion to resolve binocular fusion failure which is one of the 3D visual fatigues in the perspective of human visual system. With stimulus having discrete disparity change, binocular fusion failure rate for target stimulus having crossed and uncrossed disparity decreased. And target stimulus having continuous disparity also required relatively larger binocular disparity when peripheral blur was presented with target stimulus rather than when peripheral blur was not presented. These results imply that peripheral blur facilitated binocular fusion in the situation of binocular disparity change, and suggest that considering the characteristics of human three-dimensional visual systems, manipulating 3D contents can improve visual discomfort caused by binocular displays at low costs.

• KIPS Transactions on Software and Data Engineering
• /
• v.10 no.9
• /
• pp.375-384
• /
• 2021

Deep learning techniques have been proven to have high performance in image processing and are applied in various fields. The most widely used methods for validating a deep learning model include a holdout verification method, a k-fold cross verification method, and a bootstrap method. These legacy methods consider the balance of the ratio between classes in the process of dividing the data set, but do not consider the ratio of various features that exist within the same class. If these features are not considered, verification results may be biased toward some features. Therefore, we propose a deep learning model validation method based on data feature coverage for image classification by improving the legacy methods. The proposed technique proposes a data feature coverage that can be measured numerically how much the training data set for training and validation of the deep learning model and the evaluation data set reflects the features of the entire data set. In this method, the data set can be divided by ensuring coverage to include all features of the entire data set, and the evaluation result of the model can be analyzed in units of feature clusters. As a result, by providing feature cluster information for the evaluation result of the trained model, feature information of data that affects the trained model can be provided.

• Journal of Radiation Industry
• /
• v.11 no.1
• /
• pp.27-31
• /
• 2017

Reliable follow-up of bone mineral density (BMD) by dual energy X-ray absorptiometry (DXA) is essential in clinical practice. When there is a difference in the BMD values from DXA systems in the same patient, cross calibration equation is required for the reliable follow-up. Unfortunately, no equation is existed in BMD measure between GE Lunar Prodigy Advance (US, GE Healthcare; LPA) and Osteosys Dexxum T (Korea, Osteosys; ODT) DXA systems. In this study, we evaluate the agreement of BMD values between LPA and ODT and suggest the cross calibration equation using European spine phantom (ESP) with two systems. We performed BMD measurements using ten scans with ESP in each DXA systems. We compared BMD values and calculated cross calibration equation by linear regression analysis. The comparison between the LPA and ODT bone densitometers used the ESP. Compared to the ESP BMD values, ODT underestimated 14.36% and LPA overestimated 12.96%. The average of total BMD measurement values acquired with ODT were 21.44% lower than those from LPA. Cross-calibration equation for LPA and ODT was derived from ESP. We calculated simple cross calibration equation for LPA and ODT DXA systems. Cross-calibration equation is necessary for the reliable follow-up of BMD values in two different systems.