• Journal of Internet Computing and Services
• /
• v.19 no.5
• /
• pp.43-54
• /
• 2018

Images taken in haze weather are characteristic of low contrast and poor visibility. The process of reconstructing clear-weather image from a hazy image is called dehazing. The main challenge of image dehazing is to estimate the transmission map or depth map for an input hazy image. In this paper, we propose a single image dehazing method by utilizing the Generative Adversarial Network(GAN) for accurate depth map estimation. The proposed GAN model is trained to learn a nonlinear mapping between the input hazy image and corresponding depth map. With the trained model, first the depth map of the input hazy image is estimated and used to compute the transmission map. Then a guided filter is utilized to preserve the important edge information of the hazy image, thus obtaining a refined transmission map. Finally, the haze-free image is recovered via atmospheric scattering model. Although the proposed GAN model is trained on synthetic indoor images, it can be applied to real hazy images. The experimental results demonstrate that the proposed method achieves superior dehazing results against the state-of-the-art algorithms on both the real hazy images and the synthetic hazy images, in terms of quantitative performance and visual performance.

• The Korean Journal of Nuclear Medicine
• /
• v.32 no.3
• /
• pp.266-275
• /
• 1998

Purpose: We measured pulmonary epithelial permeability by $^{99m}Tc-DTPA$ radioaerosol clearance in patients with diabetes and correlated with the presence of microangiopathy to understand the pathophysiology of pulmonary microangiopathy and evaluate $^{99m}Tc-DTPA$ radioaerosol clearance as a diagnostic test to assess pulmonary microangiopathy. Materials and Methods: We performed $^{99m}Tc-DTPA$ radioaerosol scan in 10 normal subjects, 10 asymptomatic smokers, 20 diabetic patients without history of smoking (10 with microangiopathy, 10 without microangiopathy). $^{99m}Tc-DTPA$ clearance half-time ($T_{1/2}$) was calculated, then compared with the result of chest radiography and pulmonary function test. Results: Chest radiography and pulmonary function test were normal in all subjects. There were no significant difference of clinical or laboratory characteristics between these groups except age. The diabetic patients with microangiopathy were significantly older (p<0.05). The $T_{1/2}$ of normal subjects and asymptomatic smokers were significantly different ($65.2{\pm}23.7min$ vs $39.6{\pm}9.8min$, p<0.05). For diabetic patients with microangiopathy, the $T_{1/2}$ was $90.5{\pm}46.5min$ and significantly delayed when compared with those of normals and asymptomatic smokers (p<0.05). However, the $T_{1/2}$ of diabetic patients without microangiopathy, $70.0{\pm}12.7min$, was not significantly different from those of normals or asymptomatic smokers (p>0.05). No significant correlation was found between the $T_{1/2}$ and spirometric parameters including DLco, FVC, $FEV_1,\;FEV_1/FVC$ (%) and $FEF_{25-75%}$ in all subjects, and between the $T_{1/2}$ and duration of diabetes ;in diabetic patients. Conclusion: Eventhough the influence of age can't be excluded, delayed $^{99m}Tc-DTPA$ clearance half-time ($T_{1/2}$) in diabetic patients with microangiopathy indicates decreased pulmonary capillary permeability as one of the pathophysiologic results of pulmonary microangiopaththy. Further studies are needed in larger number of age matched control and diabetic patients to evaluate the diagnostic efficacy.

• Annual Conference of KIPS
• /
• 2013.05a
• /
• pp.350-352
• /
• 2013

야외에서 촬영된 영상은 안개나 연무 등에 의한 화질 저하가 나타난다. 이를 해결하기 위하여 안개제거를 위한 여러 장의 영상, 추가정보를 이용하는 안개제거 방법과 한 장의 영상을 이용한 안개제거 방법들이 제안되어왔다. 본 논문에서는 한 장의 영상에서 안개나 연기를 제거하여 가시성이 향상된 영상을 제공하는 방법을 제안한다. 이를 위하여 본 논문에서는 안개영상 모델에 대한 2차원 방정식 풀이를 통해 원 영상에 안개가 어느 정도의 비율로 섞여있는지를 나타내는 전달률(transmission rate)을 계산하고, 계산된 전달률을 이용하여 안개가 제거된 영상을 구한다. 제안된 방식은 기존 방법들과 달리 필터를 사용하지 않고 화소단위의 연산만을 사용하므로 후광효과가 발생하지 않고, 연산량이 매우 적어 실시간 처리가 가능하다.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.10
• /
• pp.89-96
• /
• 2024

In this paper, we propose a deep learning model and inference pipeline that can process high-resolution fog video in real-time, addressing limitations found in classical defogging algorithms and existing deep learning-based defogging models. The key idea is separating the tasks of inferring fog color and estimating the amount of fog into two distinct models, allowing for a more efficient, lightweight design that improves inference speed. While many deep defogging models perform well on synthetic fog images, they suffer from reduced effectiveness on real-world fog images with diverse fog colors and backgrounds. We solve this problem by introducing a synthetic fog dataset generation method tailored for real-world conditions. Through experiments, we demonstrate the increase in visible distance achieved by proposed model and compare its inference speed and defogging performance against pre-trained models on real-world CCTV fog images.

• Journal of Korean Society of Environmental Engineers
• /
• v.27 no.8
• /
• pp.877-885
• /
• 2005

The Froude scaling between the prototype and the model was tried to estimate the necessary ventilation rate for non-isothermal concentrated fume from the semi-closed inner space. Based on the non-dimensional similitude equations derived from the Zukoski plume rise analysis, the scaling experiments were done to verify the relationship of the non-dimensional energy release rate and the non-dimensional mass flow rate by using two different scaled volume models, model A ($1\;m{\times}1\;m{\times}1\;m$) and model B ($0.5\;m{\times}0.5\;m{\times}0.5\;m$). The experimental results showed that the theoretical similitude between the models is acceptable for the prediction of ventilation rate of the concentrated fume. The maximum energy release rate used for the experiments was $20\;kW/m^3$. In the experimental range, the similitude between the energy release rate and the ventilation mass flow rate was well defined and the necessary ventilation rates were 20-30% higher than the stoichiometric ventilation mass flow rate. Based on results of current study, the design of the local air ventilation system can be improved by correcting the effects of buoyancy and diffusion of the non-isothermal concentrated fume.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.4
• /
• pp.163-170
• /
• 2013

In general, images of outdoor scenes often contain degradation due to dust, water drop, haze, fog, smoke and so on, as a result they cause the contrast reduction and color fading. Haze removal is not easier problem due to the inherent ambiguity between the haze and the underlying scene. So, we propose a novel method to solve single scene dehazing problem using the region segmentation based on graph algorithm that has used a gradient value as a cost function. We segment the scene into different regions according to depth-related information and then estimate the global atmospheric light. The medium transmission can be directly estimated by the threshold function of graph-based segmentation algorithm. After estimating the medium transmission, we can restore the haze-free scene. We evaluated the degree of the visibility restoration between the proposed method and the existing methods by calculating the gradient of the edge between the restored scene and the original scene. Results on a variety of outdoor haze scene demonstrated the powerful haze removal and enhanced image quality of the proposed method.

• Annual Conference of KIPS
• /
• 2016.10a
• /
• pp.658-660
• /
• 2016

최근 디지털 기술의 발달로 인하여 실외 환경에서 획득된 영상은 민수분야 및 군사분야 등과 같이 다양한 목적에 따라 활용되는 분야의 폭이 넓어지고 있다. 교통정보 수집장치, 차량 블랙박스, 산불 및 지진관측, 선락/해안경비 감시, 국경 및 군사표적이동 감시 등의 목적에 의해 촬영된 영상들은 대부분 획득된 영상을 통해서 분석 및 판독의 과정을 거쳐서 각 원하는 정보 획득에 목적을 두고 있다. 하지만 실외에서 촬영된 영상은 실내에서 촬영된 영상에 비해서 기상에 따른 환경적인 요인에 노출이 쉽게 됨으로써 영상에 대한 화질 저하가 발생하는 문제점이 존재한다. 본 논문에서는 화질저하의 원인이 되는 다양한 요인 중에서도 대기중에 존재하는 먼지, 물방울 연무, 안개, 연기 등으로 인해 빛이 산란됨으로써 밝기 값을 왜곡시키는 문제점에 대한 해결 방법을 제안한다.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.49 no.1
• /
• pp.67-72
• /
• 2012

In this paper, we present an image enhancement method for atmospherically degraded images using atmospheric light and transmission based on color analysis. We first generate a normalized image using maximum value of each RGB color channel. Then, each atmospheric light is estimated from RGB color channel respectively by calculating reflectance of an image. We also, generate a transmission using gamma coefficients from the Y channel of the image. We can significantly enhance the visibility of an image by using the estimated atmospheric light and the transmission. The proposed algorithm can remove atmospheric degradation components better than existing techniques because the color prevents color distortion which is common problem of existing techniques. Experimental results demonstrate that the proposed algorithm can improve visibility be removing fog, smoke, and dust.

• The Korean Journal of Nuclear Medicine
• /
• v.28 no.3
• /
• pp.338-342
• /
• 1994

Pulmonary clearance of Tc-99m-DTPA(PCD) has been used for the measurement of polmonary epithelial permeability. It has been reported to be increased not only in variety of polmonary diseases including ARDS, interstitial fibrosis, and smokers, but also in normal subjects on positive end expiratory pressure respirator, or after exercise. It was also noted that decrease of pulmonary blood flow due to pulmonary arterial obstruction results in delayed PCD. Normal range of PCD varies with institutes. We prospectively measured PCD in 17 normals (5 males and 12 females) consisted of staffs and trainees in the department of radiology of Kangnam St. Mary's hospital using original Bark Nebulizer (India). Age ranged from 32 to 43 years. 370 MBq of Tc-99m-DTPA was inhaled in supine position and supine posterior images were subsequently obtained with 1 min/frame, $64{\times}64$ matrix and word mode for 30min. Regions of interest were set on each lung, whole lungs, and upper, middle and lower thirds of right lung, respectively. Best fit regression curve was obtained by least square method from initial 7min after peak activity on each curve and time for half clearance of maximum activity (t1/2) was calculated. Mean t1/2 was $51{\pm}11.2min$ for whole lung. There was no significant difference between t1/ 2 of right and left lungs. Initial uptake was higher in the lower third and t1/2 was shorter in the lower third than in the upper third(P<0.05). We reviewed several reports on PCD and compared our data with the others. In this study, faster clearance in the lower third may be due to the position imaged with or the environment the subjects belong to, and further investigation is under way.