• Journal of the Institute of Convergence Signal Processing
• /
• v.22 no.4
• /
• pp.206-212
• /
• 2021

Society is developing more and more, and communication methods are developing in many ways. However, developed communication is a way for the non-disabled and has no effect on the deaf. Therefore, in this paper, a CNN-based sign language translation program is designed and implemented to help deaf people communicate. Sign language translation programs translate sign language images entered through WebCam according to meaning based on data. The sign language translation program uses 24,000 pieces of Korean vowel data produced directly and conducts U-Net segmentation to train effective classification models. In the implemented sign language translation program, 'ㅋ' showed the best performance among all sign language data with 97% accuracy and 99% F1-Score, while 'ㅣ' showed the highest performance among vowel data with 94% accuracy and 95.5% F1-Score.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.222-222
• /
• 2021

도시 내 하천 친수공간은 레저 및 여가를 위한 공간과 더불어 자연경관 및 생태체험 등의 다목적으로 활용되어 지역사회에 있어 중요한 공간으로 활용되고 있다. 과거 4대강 사업으로 국가하천 내 휴식공간을 조성하였다. 그 후 친수지구 중에서 이용도가 저조한 곳을 해제하였으며, 현재 297개의 친수지구를 중점으로 관리하고 있다. 이러한 친수지구를 유지하는데 필요한 보수 비용들을 지자체에서 담당하고 있으나, 상당한 비용이 소요되므로 친수지구 지정 후 운영단계에서 지역 주민들의 특성 및 요구를 정확히 파악할 필요가 있다. 하천 친수공간에 대한 정보구축은 조사원 조사, 유지관리 기관 조사 등 인적 조사 방식을 통한 데이터 수집으로 많은 비용이 필요할 뿐만아니라 DB 갱신 부분에도 한계가 있다. 그러므로 본 연구에서는 RS/GIS를 기반으로 친수시설에 대한 정보맵 작성 방안 연구와 친수시설 뿐만 아니라 유역조사 시 하천공간에서 수행할 수 있는 기술에 대한 연구를 수행하였다. 연구대상지역은 대저생태공원과 삼락생태공원을 대상으로 선정하였다. 해당 지역 항공영상의 정합 및 전처리를 실시한 후 QGIS를 활용하여 LSMS(Large-Scale Mean Shift) 기법으로 시설물 분류를 실시하였다. 공원 내 친수시설 분류를 위해 공간 반경(Spatial radius)를 10 ~ 25까지 변화시키면서 최적 분류 결과를 도출하는 공간 반경을 찾았으며 친수시설 규모와 시설의 특성에 따라 공간 반경을 조절하여 친수시설 분류맵을 작성하였다. 친수지구 내 친수시설 분류맵과 친수지구 내 친수시설 현황 및 친수시설별 코드와, 위치정보(위도, 경도 및 표고), 면적 및 관리현황으로 분류하여 입력 할 수 있도록 하였다. 본 연구에서 구축한 친수시설 자동분류 알고리즘을 통해 전국 단위 통합 하천관리체계 구축 및 친수시설에 대한 정보맵을 작성할 수 있는 기반 마련이 가능할 것이다.

• Journal of Space Technology and Applications
• /
• v.3 no.2
• /
• pp.154-164
• /
• 2023

In space exploration, spectroscopic observation is useful for understanding objects' composition and physical properties. There are various methods for analyzing spectral data, and there are differences depending on the object and the wavelength. This paper introduces a method for analyzing visible & nearinfrared (VNIR) spectral data, which is mainly applied in lunar exploration. The main analysis methods include false color ratio image processing, reflectance pattern analysis, integrated band depth (IBD) processing, and continuum removal as preprocessing before analysis. These spectroscopic analysis methods help to understand the mineral properties of the lunar surface in the VNIR region and can be applied to other celestial bodies such as Mars.

• Journal of Korea Society of Industrial Information Systems
• /
• v.28 no.6
• /
• pp.45-52
• /
• 2023

Recently, as the importance of walking in daily life has been emphasized, projects to guarantee walking rights and create a pedestrian environment are being promoted throughout the region. In previous studies, a pedestrian environment assessment was conducted using Jeonju-si road images, and an image comparison pair data set was constructed. However, data sets expressed in numbers have difficulty in generalizing the judgment criteria of pedestrian environment assessors or visually identifying the pedestrian environment preferred by pedestrians. Therefore, this study proposes a method to interpret the results of the pedestrian environment assessment through data visualization by building a web application. According to the semantic segmentation result of analyzing the walking environment components that affect pedestrian environment assessors, it was confirmed that pedestrians did not prefer environments with a lot of "earth" and "grass," and preferred environments with "signboards" and "sidewalks." The proposed study is expected to identify and analyze the results randomly selected by participants in the future pedestrian environment evaluation, and believed that more improved accuracy can be obtained by pre-processing the data purification process.

• Journal of Korea Society of Industrial Information Systems
• /
• v.29 no.1
• /
• pp.91-105
• /
• 2024

Car accidents occur as a result of collisions between vehicles, leading to both vehicle damage and personal and material losses. This study developed a vehicle accident detection model based on 2,550 image frames extracted from car accident videos uploaded to YouTube, captured by CCTV. To preprocess the data, bounding boxes were annotated using roboflow.com, and the dataset was augmented by flipping images at various angles. The You Only Look Once version 8 (YOLOv8) model was employed for training, achieving an average accuracy of 0.954 in accident detection. The proposed model holds practical significance by facilitating prompt alarm transmission in emergency situations. Furthermore, it contributes to the research on developing an effective and efficient mechanism for vehicle accident detection, which can be utilized on devices like smartphones. Future research aims to refine the detection capabilities by integrating additional data including sound.

• Journal of Korea Multimedia Society
• /
• v.15 no.5
• /
• pp.595-605
• /
• 2012

Various methods for measuring strabismus have been developed and used in clinical diagnosis. However, most of them are based on the visual inspection by clinicians. For this reason, there is a high possibility of subjective evaluation in clinical decisions and they are only useful for cooperative patients. Therefore, the development of a more objective and reproducible method for measuring strabismus is needed. In this paper, we introduce a new software to complement the limitations of previous diagnostic methods. Firstly, we simply obtained facial images of patients and performed several preprocessing steps based on the spherical RGB color model with them. Then, the measurement of strabismus was performed automatically by using our 3D eye model and mathematical algorithm. To evaluate the validity of our software, we performed statistical correlation analysis of the results of the proposed method and the Krimsky test by two clinicians for ten patients. The coefficients of correlation for two clinicians were very high, 0.955 and 0.969, respectively. The coefficient of correlation between two clinicians also showed 0.968. We found a statistically significant correlation between two methods from our results. The newly developed software showed a possibility that it can be used as an alternative or effective assistant tool of previous diagnostic methods for strabismus.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.2
• /
• pp.95-101
• /
• 2017

In this paper, the sizing removal condition for the pretreatment of composite materials is obtained numerically by applying an image processing algorithm and nickel-plated carbon fiber is fabricated by a dry process method to enhance its electromagnetic shielding performance. Sizings that are wrapped in a polymer type material during the manufacturing of carbon fiber should be removed for dry coating. A numerical value, that is the correlation, can be obtained by determining the regular pattern of the carbon fiber in the image taken by a scanning electron microscope (SEM) after the sizing is removed. The application of the proposed numerical method to the SEM image of the fiber after the sizing is removed with solution, compressed air, solution and compressed air (hybrid), showed that this method of eliminating the sizing is superior to the hybrid method. Then, by spreading the carbon fiber roll with the sizing removed, we were able to produce nickel plated carbon fiber by the roll-to-roll sputtering method. The electromagnetic shielding performance of the fabricated 30, 40 and 100 nickel coated carbon fibers was measured. The Korea Advanced Institute of Science and Technology evaluated the electromagnetic shielding performance of the 100 nickel-coated carbon fiber to have a maximum value of 73.2 (dB) and a minimum value of 66.7 (dB). This is similar to the electromagnetic shielding rate of copper and shows that this material can be used as a cable for EV / HEV automobiles.

• Journal of Korea Multimedia Society
• /
• v.8 no.11
• /
• pp.1449-1461
• /
• 2005

In this paper, we propose a vehicle recognition system based on the classification of vehicle identification mark and recognition of vehicle license plate. In the proposed algorithm, From the input vehicle image, we first simulate preprocessing procedures such as noise reduction, thinning etc., and detect vehicle identification mark and license plate region using the frequency distribution of intensity variation. And then, we classify extracted vehicle candidate region into identification mark, character and number of vehicle by using structural feature informations of vehicle. Lastly, we recognize vehicle informations with recognition of identification mark, character and number of vehicle using hybrid and vertical/horizontal pattern vector method. In the proposed algorithm, we used three properties of vehicle informations such as Independency property, discriminance property and frequency distribution of intensity variation property. In the vehicle images, identification mark is generally independent of the types of vehicle and vehicle identification mark. And also, the license plate region between character and background as well as horizontal/vertical intensity variations are more noticeable than other regions. To show the efficiency of the propofed algorithm, we tested it on 350 vehicle images and found that the propofed method shows good Performance regardless of irregular environment conditions as well as noise, size, and location of vehicles.

• 한국해양학회지
• /
• v.22 no.3
• /
• pp.143-152
• /
• 1987

The digital data were obtained using Kennedy 9000 magnetic tape deck which was connected to the SMS960 side scan sonar during the field operations. The data of three consecutive survey tracks near Seongsan-po, Cheju were used for the development of this study. The softwares were mainly written in Fortran-77 using VAX 11/780 MINI-COMPUTER (CPU Memory; 4MB). The established mapping system consists of the pretreatment and the digital processing of seafloor image data. The pretreatment was necessary because the raw digital data format of the field magnetic tapes was not compatible to the VAX system. Therefore the raw data were read by the personal computer using the Assembler language and the data format was converted to IBM compatible, and next data were communicated to the VAX system. The digital processing includes geometrical correction for slant range, statistical analysis and cartography of the seafloor image. The sound speed in the water column was assumed 1,500 m/sec for the slant range correction and the moving average method was used for the signal trace smoothing. Histograms and cumulative curves were established for the statistical analysis, that was purposed to classify the backscattering strength from the sea-bottom. The seafloor image was displayed on the color screen of the TEKTRONIX 4113B terminal. According to the brief interpretation of the result image map, rocky and sedimentary bottoms were very well discriminated. Also it was shown that the backscattered acoustic pressurecorrelateswith the grain size and sorting of surface sediments.

• Journal of IKEEE
• /
• v.9 no.1 s.16
• /
• pp.7-18
• /
• 2005

A key component of an autonomous mobile robot is to localize itself accurately and build a map of the environment simultaneously. In this paper, we propose a vision-based mobile robot localization and mapping algorithm using scale-invariant features. A camera with fisheye lens facing toward to ceiling is attached to the robot to acquire high-level features with scale invariance. These features are used in map building and localization process. As pre-processing, input images from fisheye lens are calibrated to remove radial distortion then labeling and convex hull techniques are used to segment ceiling region from wall region. At initial map building process, features are calculated for segmented regions and stored in map database. Features are continuously calculated from sequential input images and matched against existing map until map building process is finished. If features are not matched, they are added to the existing map. Localization is done simultaneously with feature matching at map building process. Localization. is performed when features are matched with existing map and map building database is updated at same time. The proposed method can perform a map building in 2 minutes on $50m^2$ area. The positioning accuracy is ${\pm}13cm$, the average error on robot angle with the positioning is ${\pm}3$ degree.