• International Journal of Naval Architecture and Ocean Engineering
• /
• v.7 no.3
• /
• pp.640-653
• /
• 2015

In order to develop the challenging process of placing Aids to Navigation (AtoN), we propose performance measures which quantifies the effect of such placement. The best placement of AtoNs is that from which the navigator can best recognize the information provided by an AtoN. The visibility of AtoNs depends mostly on light sources, the weather condition and the position of the navigator. Visual recognition is enabled by achieving adequate contrast between the AtoN light source and background light. Therefore, the performance measures can be formulated through the amount of differences between these two lights. For simplification, this approach is based on the values of the human factor suggested by International Association of Marine Aids to Navigation and Lighthouse Authorities (IALA). Performance measures for AtoN placement can be evaluated through AtoN Simulator, which has been being developed by KIOST/KRISO in Korea and has been launched by Korea National Research Program. Simulations for evaluation are carried out at waterway in Busan port in Korea.

• Journal of Korea Multimedia Society
• /
• v.22 no.9
• /
• pp.1000-1010
• /
• 2019

Image semantic segmentation and dehazing are key tasks in the computer vision. In recent years, researches in both tasks have achieved substantial improvements in performance with the development of Convolutional Neural Network (CNN). However, most of the previous works for semantic segmentation assume the images are captured in clear weather and show degraded performance under hazy images with low contrast and faded color. Meanwhile, dehazing aims to recover clear image given observed hazy image, which is an ill-posed problem and can be alleviated with additional information about the image. In this work, we propose a deep multi-task network for simultaneous semantic segmentation and dehazing. The proposed network takes single haze image as input and predicts dense semantic segmentation map and clear image. The visual information getting refined during the dehazing process can help the recognition task of semantic segmentation. On the other hand, semantic features obtained during the semantic segmentation process can provide cues for color priors for objects, which can help dehazing process. Experimental results demonstrate the effectiveness of the proposed multi-task approach, showing improved performance compared to the separate networks.

• Journal of IKEEE
• /
• v.25 no.1
• /
• pp.83-87
• /
• 2021

In harsh environments such as fog or fine dust, the cameras' detection ability for object recognition may significantly decrease. In order to accurately obtain important information even in bad weather, fog removal algorithms are necessarily required. Research has been conducted in various ways, such as computer vision/data-based fog removal technology. In those techniques, estimating the amount of fog through the input image's depth information is an important procedure. In this paper, a linear model is presented under the assumption that the image dark channel dictionary, saturation ∗ value, and sharpness characteristics are linearly related to depth information. The proposed method of haze removal through a linear model shows the superiority of algorithm performance in quantitative numerical evaluation.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.6
• /
• pp.1229-1236
• /
• 2022

In this paper, we studied accident-aware smart streetlights to prevent secondary accidents when driving on highways. It used Arduino and sensors to inform drivers of weather conditions, incorporated functions such as LED brightness control according to sunlight and night driving vehicles, and used Raspberry Pi camera OpenCV to learn various traffic accidents, natural disasters, and wildlife.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.11a
• /
• pp.113-117
• /
• 2020

Images were taken under various weather such as rain, haze, snow often show low visibility, which can dramatically decrease accuracy of some tasks in computer vision: object detection, segmentation. Besides, previous work to enhance image usually downsample the image to receive consistency features but have not yet good upsample algorithm to recover original size. So, in this research, we jointly implement removal streak in heavy rain image and super resolution using a deep network. We put forth a 2-stage network: a multi-model network followed by a refinement network. The first stage using rain formula in the single image and two operation layers (addition, multiplication) removes rain streak and noise to get clean image in low resolution. The second stage uses refinement network to recover damaged background information as well as upsample, and receive high resolution image. Our method improves visual quality image, gains accuracy in human action recognition task in datasets. Extensive experiments show that our network outperforms the state of the art (SoTA) methods.

• Korean Journal of Remote Sensing
• /
• v.40 no.2
• /
• pp.167-177
• /
• 2024

With advancements in satellite technology, interest in target detection and identification is increasing quantitatively and qualitatively. Synthetic Aperture Radar(SAR) images, which can be acquired regardless of weather conditions, have been applied to various areas combined with machine learning based detection algorithms. However, conventional studies primarily focused on the detection of stationary targets. In this study, we proposed a method to identify moving targets using an algorithm that integrates sub-aperture SAR images and cosine similarity calculations. Utilizing a transformer-based deep learning target detection model, we extracted the bounding box of each target, designated the area as a region of interest (ROI), estimated the similarity between sub-aperture SAR images, and determined movement based on a predefined similarity threshold. Through the proposed algorithm, the quantitative evaluation of target identification capability enhanced its accuracy compared to when training with the targets with two different classes. It signified the effectiveness of our approach in maintaining accuracy while reliably discerning whether a target is in motion.

• Korean Journal of Remote Sensing
• /
• v.35 no.3
• /
• pp.457-470
• /
• 2019

Automatic Target Recognition (ATR) using Synthetic Aperture Radar (SAR) has been attracted attention in the fields of surveillance, reconnaissance, and national security due to its advantage of all-weather and day-and-night imaging capabilities. However, there have been some difficulties in automatically identifying targets in real situation due to various observational and environmental conditions. In this paper, ATR problems in Extended Operating Conditions (EOC) were investigated. In particular, we considered partial occlusions of the target (10% to 50%) and differences in the depression angle between training ($17^{\circ}$) and test data ($30^{\circ}$ and $45^{\circ}$). To simulate various occlusion conditions, SARBake algorithm was applied to Moving and Stationary Target Acquisition and Recognition (MSTAR) images. The ATR accuracies were evaluated by using the template matching and Adaboost algorithms. Experimental results on the depression angle showed that the target identification rate of the two algorithms decreased by more than 30% from the depression angle of $45^{\circ}$ to $30^{\circ}$. The accuracy of template matching was about 75.88% while Adaboost showed better results with an accuracy of about 86.80%. In the case of partial occlusion, the accuracy of template matching decreased significantly even in the slight occlusion (from 95.77% under no occlusion to 52.69% under 10% occlusion). The Adaboost algorithm showed better performance with an accuracy of 85.16% in no occlusion condition and 68.48% in 10% occlusion condition. Even in the 50% occlusion condition, the Adaboost provided an accuracy of 52.48%, which was much higher than the template matching (less than 30% under 50% occlusion).

• Journal of Korea Water Resources Association
• /
• v.37 no.7
• /
• pp.569-579
• /
• 2004

Recently, the extreme drought is often occurred due to the global warming and the serious weather changes. Also, the problems of the water pollution In the developed areas, the oppositions from people in the upper stream area and water concession from the local governments affect the national request to get more clean water resources in upper stream of the undeveloped areas. It also brings on the necessity of recognition for water supply managements. Therefore, as the water demand is rapidly changes in the metropolitan areas, the capability of water supply from the north Han river basin dams should be appropriately investigated. In this study, we developed a simulation system using STELLA (equation omitted) software environment, a shared vision model, to analyze the possibility of the stable water supply from north Han river basin dams. Also, three different rules are applied on this model by dividing the water level to minimum(Rule 1), medium(Rule 2) and maximum(Rule 3). Using the rules, the safety yield changes are analyzed for dam rule curve of the reservoir and hydropower release.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.25 no.3
• /
• pp.217-222
• /
• 2015

Clustering is one of important data mining techniques known as exploratory data analysis and is being applied in various engineering and scientific fields such as pattern recognition, remote sensing, and so on. The method organizes data by abstracting underlying structure either as a grouping of individuals or as a hierarchy of groups. Weather radar observes atmospheric objects by utilizing reflected signals and stores observed data in corresponding coordinate. To analyze the radar data, it is needed to be separately organized precipitation and non-precipitation echo based on similarities. Thus, this paper studies to apply clustering method to radar data. In addition, in order to solve the problem when precipitation echo locates close to non-precipitation echo, fuzzy logic based clustering method which can consider both distance and other properties such as reflectivity and Doppler velocity is suggested in this paper. By using actual cases, the suggested clustering method derives better results than previous method in near-located precipitation and non-precipitation echo case.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.20 no.2
• /
• pp.183-192
• /
• 2009

SAR(Synthetic Aperture Radar) is an all-weather imaging radar and is widely used in military and civil application. Especially high-resolution SAR images are very important in military purpose because it can be used at target recognition application. LIG Nex1 developed a SAR system called NexSAR with bandwidth of 600 MHz and resolution of 30 cm to obtain technologies required for high-resolution SAR. To achieve 600 MHz bandwidth of waveform generator, two DDSs are used and its output signals are SSB modulated. And deramp technique is used to reduce the sampling rate of ADC at high resolution mode. NexSAR has stripmap and spotlight modes and its functionality and performances are evaluated through ground and flight tests.