• Journal of Digital Contents Society
• /
• v.18 no.8
• /
• pp.1635-1640
• /
• 2017

In this paper, a hierarchical image tree model for weather classification is defined in a single outdoor image, and a weather classification algorithm using image intensity and k-mean segmentation image is proposed. In the first level of the hierarchical image tree model, the indoor and outdoor images are distinguished. Whether the outdoor image is daytime, night, or sunrise/sunset image is judged using the intensity and the k-means segmentation image at the second level. In the last level, if it is classified as daytime image at the second level, it is finally estimated whether it is sunny or foggy image based on edge map and fog rate. Some experiments are conducted so as to verify the weather classification, and as a result, the proposed method shows that weather features are effectively detected in a given image.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.46 no.1
• /
• pp.10-21
• /
• 2009

In order to enhance the contrast in the regions where the pixels have similar intensities, this paper presents a new histogram equalization scheme. Conventional global equalization schemes over-equalizes those regions so that too bright or dark pixels are resulted and local equalization schemes produce unexpected discontinuities at the boundaries of the blocks. The proposed algorithm segments the original histogram into sub-histograms with reference to brightness level and equalizes each sub-histogram with the limited extents of equalization considering its mean and variance. The final image is determined as the weighted sum of the equalized images obtained by using the sub-histogram equalizations. By limiting the maximum and minimum ranges of equalization operations on individual sub-histograms, the over-equalization effect is eliminated. Also the result image does not miss feature information in low density histogram region since the remaining these area is applied separating equalization. This paper includes how to determine the segmentation points in the histogram. The proposed algorithm has been tested with more than 100 images having various contrast in the images and the results are compared to the conventional approaches to show its superiority.

• Journal of the Korean earth science society
• /
• v.30 no.3
• /
• pp.344-353
• /
• 2009

By measuring the brightness of night sky, we have investigated light pollution around the observatory in the College of Education, Seoul National University. As a result of measuring the extinction coefficient and photometric constants by standardization, in January 28, 2009, the extinction coefficient found to be $k_B$=0.359 and photometric constant was $C_B$=4.397. In March 27, 2009, extinction coefficients were $k_B$=0.896 and $k_V$=0.725, and photometric constants were $C_B$=6.235 and $C_V$=6.027. Brightness of the night sky was measured from east, west, south, and north each by altitude of $20^{\circ}$, $40^{\circ}$, $60^{\circ}$, $75^{\circ}$ and $90^{\circ}$. Data reduction and analysis was based on IRAF. Seeings for the each day of observation were 5.1 and 5.7 arcseconds in January 28 and March 27, respectively. Night sky spanned the magnitude range of $16{\leq}m_V$, $m_B{\leq}18$ We found that the brightness of night sky located at downtown was twice to four times brighter. On these observational conditions, limiting magnitude within 40cm-telescope becomes 11-13 magnitudes. Compared with Jan 28 and Mar 27, night sky brightness of January is 1 magnitude fainter than that of March in B filter.

• Journal of KIISE:Software and Applications
• /
• v.27 no.10
• /
• pp.1046-1053
• /
• 2000

본 논문에서는 정규화된 웨이블렛 방법에 기반한 얼굴 인식 방법을 제안한다. 추출된 얼굴 영역의 크기를 정규화하고. 배경 영역을 제거하기 위해 이진화된 가우시안 윈도우를 사용하였으며, 또한 조명의 영향을 줄이기 위해 얼굴 영역의 히스토그램을 상세화하고, 왼쪽 부분의 밝기와 오른쪽 부분의 밝기를 평균하였다. 정규화된 얼굴 영역은 극좌표계로 표현하여 사각형의 형태를 가지게 하였다. 특징으로는 가보 웨이블렛 계수를 사용한다. 가보 웨이블렛 변환은 매개 변수와 정규화된 얼굴 영역의 해상도를 바꾸어 가며 여러 차례 적용하였다. 인식에 유용한 계수들을 선택하기 위해 FD 분석을 수행하였다. 선택된 특정들은 FD 값과 함께 인식에 사용되었다. 실험 결과를 보면 제안된 방법이 매우 유망하다는 것을 알 수 있다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2012.06b
• /
• pp.348-350
• /
• 2012

본 논문에서는 열 영상카메라를 통해 입력 받은 영상을 CS-LBP(Center-symmetric LBP)와 랜덤 포레스트(Random forest)를 이용하여 보행자 휴먼 객체를 검출하는 방법을 제안한다. 우선 불필요한 후보영역을 줄이기 위해 열 영상의 표준편차, 밝기 평균, 밝기 최대값을 이용하여 이진화하고, 신체부위 중 가장 발열이 강한 얼굴부위를 핫스팟 영역으로 설정한다. 그 후, 핫스팟 영역에서 CS-LBP특징을 추출하여 결정 트리의 앙상블인 랜덤 포레스트 분류기를 이용하여 최종적인 보행자 휴먼 객체를 검증한다. CS-LBP와 랜덤 포레스트 분류기를 통해 실시간 보행자 객체의 검출이 가능하고, 높은 검출 성능을 나타내었다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.06a
• /
• pp.217-220
• /
• 2022

본 논문에서는 CCTV 영상 화질을 향상하고 해상도를 높이기 위해 딥 러닝(Deep Learning)을 이용하여 잡음 제거(Denoising) 와 초해상도(Super-resolution) 작업을 수행한다. 데이터 증강(Data Augmentation)을 통한 초해상도 성능 향상을 위해서 잡음 제거 네트워크의 출력 영상을 초해상도 네트워크의 입력으로 사용하는 순차적 작업을 사용한다. 또한 딥 러닝을 이용한 영상처리에서 발생하는 평균 밝기 오차 문제를 해결하기 위한 손실함수(Loss Function)와 두 가지 이상의 순차적인 딥 러닝 작업에서 발생하는 문제점을 극복하기 위한 손실함수를 제안한다. 제안하는 손실함수는 네트워크의 출력 영상과 타겟 영상의 밝기 오차를 줄이는 것이 가능하고, 순차적 작업에서 보다 정확한 모델 성능 판단이 가능하다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2008.05a
• /
• pp.84-87
• /
• 2008

본 논문에서는 Q.Wang & R.K.Ward 가 제안한 WTHE(weighted and thresholded histogram equalization)방법의 enhancement parameters를 주어진 영상의 히스토그램 분포에 따라 적응적으로 제공하는 방법을 제안한다. WTHE는 영상의 히스토그램을 weight와 threshold를 이용하여 변형한 후 히스토그램 평활화(histogram equalization : HE)방법을 수행 함으로써 화질을 개선하는 방법이다. 이 방법은 두 가지 parameters 제어로 기존의 히스토그램 평활화 방법의 단점인 과도한 밝기 변화와 불필요한 artifacts를 줄일 수 있다. 본 논문에서는 WTHE 방법을 좀 더 간편하면서 다양한 분야에 적용하기 위해서 입력 영상에 따라 달라지는 parameters 값을 자동으로 제공하는 적응형 WTHE(Adaptive WTHE : AWTHE) 방법을 제안하고, 제안된 방법의 성능을 실험으로 제시한다.

• The KIPS Transactions:PartB
• /
• v.18B no.5
• /
• pp.265-270
• /
• 2011

In this paper, we propose a method for intensity correction using binarization-based region segmentation in 3D stereoscopic images. In the proposed method, 3D stereoscopic right image is segmented using binarizarion. Small regions in the segmented image are eliminated. For each region in right image, a corresponding region in left image is decided through region matching using correlation coefficient. When region-based matching, in order to prevent overlap between regions, we remove a portion of the area closed to the region boundary using morphological filter. The intensity correction in left and right image can be performed through histogram specification between the corresponding regions. Simulation results show the proposed method has the smallest matching error than the conventional method when we generate the right image from the left image using block based motion compensation.

• Korean Journal of Remote Sensing
• /
• v.31 no.6
• /
• pp.583-598
• /
• 2015

This paper presents derivation of background temperature from geostationary satellite and its validation based on ground measurements and Geographic Information System (GIS) for future use in weather and surface heat variability. This study only focuses on daily and monthly brightness temperature in 2012. From the analysis of COMS Meteorological Data Processing System (CMDPS) data, we have found an error in cloud distribution of model, which used as a background temperature field, and in examining the spatial homogeneity. Excessive cloudy pixels were reconstructed by statistical reanalysis based on consistency of temperature measurement. The derived Brightness temperature has correlation of 0.95, bias of 0.66 K and RMSE of 4.88 K with ground station measurements. The relation between brightness temperature and both elevation and vegetated land cover were highly anti-correlated during warm season and daytime, but marginally correlated during cold season and nighttime. This result suggests that time varying emissivity data is required to derive land surface temperature.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2013.05a
• /
• pp.820-823
• /
• 2013

Recently, the need for improving image quality of the image is increasing in various fields smartphones, cameras, and portable devices. How a significant impact on improving image quality of the image is a contrast enhancement, as a representative method to improve the contrast, the process of histogram equalization, various studies have been made. However, the method of histogram equalization general, by readjusting the only brightness, when the image histogram is biased to one side, due to changes in the excess brightness, distortions such as blocking phenomenon occurs. In this paper, we provide a contrast enhancement techniques through the compression and re-distribution of a well-balanced average brightness of the histogram distribution. By be differential compression histogram based on the histogram frequency in order to suppress the supersaturation phenomenon due to the increase in contrast ratio excessive repositioning well-balanced histogram lopsided, the proposed method, the balance of the brightness of the image I want to to take. The experimental results, the image brightness is balanced manner compared to conventional methods, the proposed method showed a good effect to improve the contrast without supersaturation phenomenon as compared with the conventional methods.