• Korean Journal of Remote Sensing
• /
• v.38 no.6_1
• /
• pp.1343-1356
• /
• 2022

Clouds or shadows are the most problematic when monitoring crops using optical satellite images. To reduce this effect, a composite algorithm was used to select the maximum Normalized Difference Vegetation Index (NDVI) for a certain period. This Maximum NDVI Composite (MNC) method reduces the influence of clouds, but since only the maximum NDVI value is used for a certain period, it is difficult to show the phenomenon immediately when the NDVI decreases. As a way to maintain the spectral information of crop as much as possible while minimizing the influence of clouds, a Score-Based Composite (SBC) algorithm was proposed, which is a method of selecting the most suitable pixels by defining various environmental factors and assigning scores to them when compositing. In this study, the Sentinel-2A/B Level 2A reflectance image and cloud, shadow, Aerosol Optical Thickness(AOT), obtainging date, sensor zenith angle provided as additional information were used for the SBC algorithm. As a result of applying the SBC algorithm with a 15-day and a monthly period for Dangjin rice fields and Taebaek highland cabbage fields in 2021, the 15-day period composited data showed faster detailed changes in NDVI than the monthly composited results, except for the rainy season affected by clouds. In certain images, a spatially heterogeneous part is seen due to partial date-by-date differences in the composited NDVI image, which is considered to be due to the inaccuracy of the cloud and shadow information used. In the future, we plan to improve the accuracy of input information and perform quantitative comparison with MNC-based composite algorithm.

• Proceedings of the Korea Contents Association Conference
• /
• 2009.05a
• /
• pp.179-184
• /
• 2009

Robust face recognition under various illumination environments is difficult to achieve. For face recognition robust to illumination changes, usually face images are normalized with respect to illumination as a preprocessing step before face recognition. The anisotropic smoothing-based illumination normalization method, known to be one of the best illumination normalization methods, cannot handle casting shadows. In this paper, we present an efficient illumination normalization method for face recognition. The proposed illumination normalization method separates the effect of illumination from eigenfaces and constructs an illumination-separated eigenface subspace. Then, an incoming face image is projected into the subspace and the obtained projected face image is rendered so that illumination effects including casting shadows are reduced as much as possible. Application to real face images shows the proposed illumination normalization method.

• Journal of the Korean Society of Costume
• /
• v.60 no.3
• /
• pp.84-98
• /
• 2010

This study aims to present basic materials to promote the importance of make-up design in advertising images and designing effective perfume advertisement by looking at features such as colors, forms, patterns, and material feels, the elements of make-up designs shown magazine advertising images for various perfume brands. The features of make-up designs shown in Anna Sui and Bvlgari perfume advertisements are as the following. Among 9 images in Anna Sui advertisements, 5 displayed mat-like skin feels. Among shadow colors, brown colors accounted for over half, 5 out of 9 while pink colors accounted 5 out of 9 for cheek make-up. Among 7 images in Bvlgari advertisements, pale blue skin colors accounted most, 5 out 7. 4 out of 7 displayed mat-like skin feels and 4 out 7 were in pink colors for lip make-up. Anna Sui brand was found to use a bold and unique makeup design for each perfume advertisement according to the designer's fashion orientation tendency to pursue unique, vintage-style, and romantic images whereas Bvlgari focused on make-up to emphasize elegant and refined female beautify according to the company philosophy of pursuing elegant and high-class images. Namely, it was determined that the BI(brand image), the images a company pursues is related to the MI(make up image), the make-up images in advertisements.

• Journal of radiological science and technology
• /
• v.18 no.2
• /
• pp.49-59
• /
• 1995

Recently, general radiography became to variety because of the continuous development of Inverter type generator and ortho chromatic system. Therefore, we must re-evaluate about patient exposure dose and image quality. I studied about chest radiography which has frequency among general radiography being used during FEB. to AUG., 1994 over 151 medical facilities from medical facilities that are located in Seoul area. The result obtained were as follows ; 1) The rectification method of the generator were employing mainly single phase full wave in 82.8 %, three phase full wave in 11.26 % and Inverter type in 4.64 % and condenser type is 1.32 %. 2) Exposure kV was used below 80 kV in most medical facilities and 21.8 % of the medical facilities was using high tube voltage higher than 120 kV. 3) The exposure time was used below the 0.05 sec in 28.4 %, in 29.8 % of the medical facilities used above 0.1 sec. 4) The usage frequency of the added filter is 15.3 %, and among them compound filter was used only 4 %. 5) Rare earth screen was used in 37.7 %. 6) The average skin entrance dose was 0.25 mSv, the range of dose is $0.05{\sim}0.79\;mSv$ in each medical facilities. 7) The average density of the lung field is 1.76, 0.49 in lung side, 0.30 in mediastinum and 0.37 in heart shadow. Therefore it is required to improve all of these for increasing image quality and reducing patient exposure dose as soon as possible.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.44 no.6
• /
• pp.84-92
• /
• 2007

The gait recognition is presented for human identification from a sequence of noisy silhouettes segmented from video by capturing at a distance. The proposed gait recognition algorithm gives better performance than the baseline algorithm because of segmentation of the object by using multiple modules; i) motion detection, ii) object region detection, iii) head detection, and iv) active shape models, which solve the baseline algorithm#s problems to make background, to remove shadow, and to be better recognition rates. For the experiment, we used the HumanID Gait Challenge data set, which is the largest gait benchmarking data set with 122 objects, For realistic simulation we use various values for the following parameters; i) viewpoint, ii) shoe, iii) surface, iv) carrying condition, and v) time.

• Journal of Digital Contents Society
• /
• v.11 no.4
• /
• pp.553-559
• /
• 2010

By applying the various reformation techniques by using a circle raw data of after computed tomography image in the patient enforcing the spine screw fixation, this research tried to look into the various information including the exact location of the position of the screw spike and accuracy of an operation. In a clinical, by applying the or multi planar reformatting(MPR), that is the re-composition technique used mainly, maximum intensity projection (MIP), and volume rendering technique(VRT) and transformation removal from a register modifying VRT, video data were compared and were analyzed by the quantitative method and qualitative method. It is judged as the transformation volume rendering technique of the re-composition technique which is most useful in minimizing the artifact shadow by the exact location of the position of a screw and metal among the analysis and evaluation result computed tomography image reformation technique of the reformation image after the spine screw fixation.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.38 no.6
• /
• pp.70-81
• /
• 2001

This paper presents a 3D face modeling method using a CCD camera and a projector (LCD projector or Slide projector). The camera faces the human face and the projector casts white stripe patterns on the human face. The 3D shape of the face is extracted from spatial and temporal locations of the white stripe patterns on a series of image frames. The proposed method employs region segmentation and multi-beam techniques for efficient 3D modeling of hair region and faster 3D scanning respectively. In the proposed method, each image is segmented into face, hair, and shadow regions, which are independently processed to obtain the optimum results for each region. The multi-beam method， which uses a number of equally spaced stripe patterns, reduces the total number of image frames and consequently the overall data acquisition time. Light beam calibration is adopted for efficient light plane measurement, which is not influenced by the direction (vertical or horizontal) of the stripe patterns. Experimental results show that the proposed method provides a favorable 3D face modeling results, including the hair region.

• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.38 no.6
• /
• pp.458-469
• /
• 1989

A measurement method is proposed which finds 3-D position and attitude of a known polyhedra utilizing shading information. Through the systematic interpretation of relations between polyhedra and its image as well as shadow image and also the determination of candidate position, 3-D information with respect to vertex of polyhedra is extracted. Following preprocessing of this information, the image of polyhedra is represented in terms of the scene with positioned object and the correspondence is searched by means of matching process between a scene description of the object and the correspondence is searched by means of matching process between a scene description of the object and a model description stored in data-base. In the experiments, initially 3-D information is employed to select several model regions, and objects are recognized through matching process with respect to scene regions. The results demonstrate that the recognition system performs with a high efficiency by proper selection of the threshold values.

• The KIPS Transactions:PartB
• /
• v.16B no.1
• /
• pp.17-24
• /
• 2009

This paper suggests the background image and the algorism of detecting an accident at crossroads by using the sequence of traffic light at crossroads, which is installed within the crossroads, in order to detect an accident within crossroads. A method of using the existing image contains a problem that the accident-detection ratio gets lower in a situation that noise occurs loudly given using new accident model, the confused situation, or sound source. This study used the accident detection by developing a filter of using the property of histogram in the sequence of traffic light at crossroads and the background image, in order to reduce misjudgment of an accident caused by external shadow, vehicle stoppage, vehicle headlight, and externally environmental influence. As a result of experimenting by acquiring 15 actual accident images in order to examine the performance of the suggested algorism, the accident was detected in all the 15 videos. Even as for a new accident model, the accident within crossroads could be detected.

• Journal of KIISE:Computing Practices and Letters
• /
• v.5 no.6
• /
• pp.748-757
• /
• 1999

본 논문에서는 차량 영역의 추출을 이용한 효율적인 교통 혼잡도 측정 시스템을 설계하고 구현한다. 차량 영역 정보의 추출은 첫째 영역 분할, 둘째 작은 영역의 제거와 영역의 직사각형화, 셋째 영역의 병합 및 삭제의 단계로 나눌 수 있다. 영역 분할 단계에서는 획득한 도로 영상을 영역 기반 영역 분할에 의해 영역으로 분할한다. 그 다음 영역 분할 후의 영역 정보 중 차량 영역을 추출하는데 영향을 미치지 않는 작은 영역들을 제거하고, 남은 영역들을 직사각형화한다. 마지막으로 차선 별로 남은 영역들을 병합, 삭제함으로써 각 차선마다 차량 영역 정보를 추출할 수 있다. 이러한 방법은 배경 영상과 같은 부가적인 정보를 사용하지 않고 도로 자체 영상만으로 교통 혼잡도를 측정할 수 있으며, 그림자의 영향이 없을 경우 적용할 수 있는 기법이다.Abstract In this paper, we designed and implemented an efficient road congestion analysis system using regional information. To extract vehicle regions from a road image, the system process the image in five steps: segmentation, small region elimination, region rectangularization, region merging and region deletion. First, we segment road image by a threshold value. Then, we eliminate useless small regions to extract vehicle region, and perform region rectangularization. Finally, we extract vehicle region of each lane of the road by region merging and deletion. This method has the advantage of measuring road congestion without additional information such as background images. But this method must be applied to road images without shadow.