• 한국멀티미디어학회논문지
• /
• 제20권8호
• /
• pp.1166-1174
• /
• 2017

In general, the detection of the vehicle license plate is a previous step of license plate recognition and has been actively studied for several decades. In this paper, we propose an algorithm to detect a license plate area of a moving vehicle from a video captured by a fixed camera installed on the road using the Convolution Neural Network (CNN) technology. First, license plate images and non-license plate images are applied to a previously learned CNN model (AlexNet) to extract and classify features. Then, after detecting the moving vehicle in the video, CNN detects the license plate area by comparing the features of the license plate region with the features of the license plate area. Experimental result shows relatively good performance in various environments such as incomplete lighting, noise due to rain, and low resolution. In addition, to protect personal information this proposed system can also be used independently to detect the license plate area and hide that area to secure the public's personal information.

• 디지털산업정보학회논문지
• /
• 제8권1호
• /
• pp.9-15
• /
• 2012

Multiple convenience features have been getting installed in recently released cars. However, the control of them has many uncomfortable matters yet. To resolve them, it is needed to study how to use easier the convenience features and control them remotely. Currently, wide range of convergence services are being released in various industries by using smartphone and smartphones with its state-of-the-art functions also are being released. In this paper, we design and implement smartphone-based applications for controling the vehicle's convenience features to control the vehicle convenience features with smartphone. To do this, we configure CAN (Controller Area Network) communication between the vehicle's various convenience features, and establish MCU (Micro Controller Unit) to control each feature. We also connect between the MCU and smartphones to make them available for the remote control. We can control lights, turn signals, audio, windows, air conditioner, and so on with the implemented smartphone-based control service of vehicle's convenience features using CAN remotely.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2007년도 Proceedings of ISRS 2007
• /
• pp.548-551
• /
• 2007

Among components of digital topographic maps used officially in Korea, only contours have 3D values except buildings and trees that are demanded in landscape planning. This study presented a series of processes for 3Dreconstructing landscape features such as terrain, buildings and standing trees using LiDAR (Light Detection And Ranging) data and aerial digital photo graphs. The 3D reconstructing processes contain 1) building terrain model, 2) delineating outline of landscape features, 3) extracting height values, and 4) shaping and coloring landscape features using aerial photograph and 3-D virtual data base. LiDAR data and aerial photograph was taken in November 2006 for $50km^{2}$ area in Sorak National Park located in eastern part of Korea. The average scanning density of LiDAR pulse was 1.32 points per square meter, and the aerial photograph with RGB bands has $0.35m{\times}0.35m$ spatial resolution. Using reconstructed 3D landscape features, visibility with the growing trees with time and at different viewpoints was analyzed. Visible area from viewpoint could be effectively estimated considering 3D information of landscape features. This process could be applied for landscape planning like building scale with the consideration of surrounding landscape features.

• 한국환경복원기술학회지
• /
• 제13권5호
• /
• pp.135-145
• /
• 2010

This study was carried out to investigate the current urban forest network and future plan of forest framework using nearest features program where stepping stones within 5km from the core forest were selected in the program. We found several conclusions as follows: First, we found that cities in inland area including Daegu, Dajeon and Gwangju have quite different types of forest network with comparing to the cities nearby coastline including Busan, Incheon and Ulsan. The cities in inland area have large mountain patch around each city. However they have small and lower number of island forests within their urban area. Otherwise, cities nearby coastline have more forest patches than in the cities in inland area, and Busan and Incheon especially have strong forest network using various size of forest patches. Second, Daegu and Daejeon have much smaller forest patches distributed in each urban area. So additional forest patches should be added to have highly strong forest network within urban area. Third, Ulsan and Gwangju have most stepping-stone forests close to the large mountain patches in suburban area, which are not able to connect to the forest patches in the central area of each city So additional forest patches are needed to be added in the central area of each city for reinforcing the effect of stepping stone in the central area. Though there should be an addition approach except for forest size and its isolation to construct the ecological network in an urban area, this indices can be a good method to check an environmental and ecological status in an urban area.

• 한국컴퓨터정보학회논문지
• /
• 제20권12호
• /
• pp.29-36
• /
• 2015

In this paper, we propose the horizontal line detection using the Haar-like features and linear regression in infrared images. In the marine environment horizon image is very useful information on a variety of systems. In the proposed method Haar-like features it was noted that the standard deviation be calculated in real time on a static area. Based on the pixel position, calculating the standard deviation of the around area in real time and, if the reaction is to filter out the largest pixel can get the energy map of the area containing the straight horizontal line. In order to select a horizontal line of pixels from the energy map, we applied the linear regression, calculating a linear fit to the transverse horizontal line across the image to select the candidate optimal horizontal. The proposed method was carried out in a horizontal line detecting real infrared image experiment for day and night, it was confirmed the excellent detection results than the legacy methods.

• 한국정보통신학회논문지
• /
• 제9권6호
• /
• pp.1282-1287
• /
• 2005

수 마이크로 단위로 계측되는 반도체 COG의 불량 검사에 있어서 칩 얼라인은 검사의 정확성을 높이는데 매우 중요한 역할을 한다. 본 논문에서는 칩 얼라인의 정확성을 높이기 위해서 영역분할 패턴매칭 방법을 제안한다. 영역분할 패턴매칭 방법은 세분화 된 영역 내의 특징치들과 영역들 간의 상관관계를 비교하여 매칭된다. 그리고 불량 패턴으로 인한 매칭오류를 최소화 하기 위해서 패턴 주위의 3영역을 학습시킨다. 제안된 방법은 분할 된 영역에서 특징치를 찾기 때문에 매칭 시간을 단축시키는 효과와 정확성을 높일 수 있는 이점을 가지고 있다.

• 한국환경복원기술학회지
• /
• 제8권2호
• /
• pp.65-75
• /
• 2005

A green area plan is getting attention as a way to solve the recent urban problems such as the rise of environment problems. To correspond to this change, there were dramatic amendments for the green area related laws. These amendments are appraised as the epochal turning point for the green area policies. This study is to introduce the main contents of amended urban green area law and urban park law in Japan, to compare with the Korean green area related laws, and to summarize the special features of both countries' green policies and the comments about the Korean green area policy structures. As a result, this amendment of Japan established the unified green area policy structure supporting the green area policies of municipal governments and is inducing living environment improvement by securing green area in the center of city, support, and the participation of residents. On the other hands, this amendment of Korea is a lack of the systemization of green area policy and the phased establishments of green area plan in spite of the scope of whole city because of absence of the unified high level plan.

• 한국가구학회지
• /
• 제21권1호
• /
• pp.54-61
• /
• 2010

This study was carried out to identify the differences of some macro and microanatomical features between Eunsasi poplar (Populus tomentiglandulosa T. Lee) and Konara oak (Quercus serrata Thunb.). Anatomical features such as vessel and fiber diameter, cross sectional area of ray parenchyma and endwall pit aperture diameter in ray parenchyma were compared. Differences of anatomical features between two species were found statistically significant. Earlywood vessel diameter in poplar and oak increased from pith to bark while it was found almost stable in latewood. Fiber diameter in poplar was higher than oak fiber. Ray area and the pit aperture diameter in endwall of ray parenchyma were found higher in oak than those in poplar.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권1호
• /
• pp.245-265
• /
• 2022

In response to problems such as insufficient extraction information, low detection accuracy, and frequent misdetection in the field of Thangka image defects, this paper proposes a YOLOv5 prediction algorithm fused with the attention mechanism. Firstly, the Backbone network is used for feature extraction, and the attention mechanism is fused to represent different features, so that the network can fully extract the texture and semantic features of the defect area. The extracted features are then weighted and fused, so as to reduce the loss of information. Next, the weighted fused features are transferred to the Neck network, the semantic features and texture features of different layers are fused by FPN, and the defect target is located more accurately by PAN. In the detection network, the CIOU loss function is used to replace the GIOU loss function to locate the image defect area quickly and accurately, generate the bounding box, and predict the defect category. The results show that compared with the original network, YOLOv5-SE and YOLOv5-CBAM achieve an improvement of 8.95% and 12.87% in detection accuracy respectively. The improved networks can identify the location and category of defects more accurately, and greatly improve the accuracy of defect detection of Thangka images.

• 한국해양학회지
• /
• 제11권1호
• /
• pp.9-17
• /
• 1976

Multispectral scanner data collected by LANDSAT-1 over the mid West Sea of Korea were analyzed and interpreted for delineation of coastal features and turbidity distribution patterns during different portions of the tidal cycle. Imagery from two successful LANDSAT-1 overpasses of the area in October 1972 and in October 1973 had been used to prepare schematic maps of coastal features and turbidity distributions. Color composite imagery of LANDSAT MSS 4, 5 and 7 gave the best representation of shorelines, coastlines and tidal flats. MSS 5 imagery was most effective in differentiating relative turbidity levels through density slicing techniques. Referring to the tidal power development of Garolim Bay, the basin area measurements assuming dyke construction at the bay entrance, have been carried out on the coastal reature maps comiled from LANDSAT imagery, and those results were correlated with existing data. General areal patterns of surface turbidity distribution in the study area revealed close similarity with bathymetry of the area. Synoptic circulation patterns were also well discriminated from the LANDSAT imagery using the suspended sediment as a tracer.