• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.10
• /
• pp.955-961
• /
• 2010

In this paper, we describe a study of a vehicle detection method based on a Boosting Classifier which uses Histogram of Oriented Gradient (HOG) features and Image Segmentation techniques. An input image is segmented by means of a split and merge algorithm. Then, the two largest segmented regions are removed in order to reduce the search region and speed up processing time. The HOG features are then calculated for each pixel in the search region. In order to detect the vehicle region we used the AdaBoost (adaptive boost) method, which is well known for classifying samples with two classes. To evaluate the performance of the proposed method, 537 training images were used to train and learn the classifier, followed by 500 non-training images to provide the recognition rate. From these experiments we were able to detect the proper image 98.34% of the time for the 500 non-training images. In conclusion, the proposed method can be used for detecting the location of a vehicle in an intelligent vehicle control system.

• The KIPS Transactions:PartB
• /
• v.10B no.7
• /
• pp.829-836
• /
• 2003

In this paper, we propose a stochastic model for sentence speech understanding using dictionary and thesaurus. The proposed model extracts words from an input speech or text into a sentence. A computer is sellected category of dictionary database compared the word extracting from the input sentence calculating a probability value to the compare results from stochastic model. At this time, computer read out upper dictionary information from the upper dictionary searching and extracting word compared input sentence caluclating value to the compare results from stochastic model. We compare adding the first and second probability value from the dictionary searching and the upper dictionary searching with threshold probability that we measure the sentence understanding rate. We evaluated the performance of the sentence speech understanding system by applying twenty questions game. As the experiment results, we got sentence speech understanding accuracy of 79.8%. In this case, probability ($\alpha$) of high level word is 0.9 and threshold probability ($\beta$) is 0.38.

• Preventive Nutrition and Food Science
• /
• v.1 no.1
• /
• pp.134-142
• /
• 1996

Diabetes carries an increased risk of atherosclerotic disease that is not fully explained by known car-diovascular risk factors. There is accumulating evidence that advanced glycation of structural proteins, and oxidation and glycation of circulating lipoproteins, are implicated in the pathogenesis of diabetic ather-osclerosis. Reactions involving glycation and oxidation of proteins and lipids are believed to contribute to atherogenesis. Glycation, the nonenzymatic binding of glucose to protein molecules, can increase the ather-ogenic potential of certain plasma constituents, including low density lipoptotein(LDL). Glycation of LDL is significant increased in diabetic patients compared with normal subjects, even in the presence of good glycemic control. Metabolic abnormalities associated with glycation of LDL include diminished recognition of LDL by the classic LDL receptor; increased covalent binding of LDL in vessel walls ; enhanced uptake of LDL by the macrophages, thus stimulating foam cell formation ; increased platelet aggregation; formation of LDL-immune complexes ; and generation of oxygen free radicals, resulting on oxidative damage to both the lipid and protein components of LDL and to any nearby macromolecules. Oxidized lipoproteins are characterzied by cytotoxicity, potent stimulation of foam cell formation by macrophages, and procoagulant effects. Combined glycation and oxidation, "glycoxidation" occurs when oxidative reactions affect the initial products of glycation, and results in irreversible structural alterations of proteins. Glycoxidation is of greatest significance in long lived proteins such as collagen. In these proteins, glycoxidation products, believed to be atherogenic, accumulate with advancing age : in diabetes, their rate of accumulate is accelerated. Inhibition of glycation, oxidation and glycoxidation may form the basis of future antiaterogenic strategies in both diabetic and nondiabetic individuals.dividuals.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.12 no.4
• /
• pp.182-190
• /
• 2009

To solve the problems of traffic congestion, air pollution and energy derived from increased car consumption, people are in recognition of the importance of green mode, bicycle. Bicycle use rate in Busan is lower due to the terrain and limited public transportation accessibility. Therefore, geographical conditions and use activation should be initially considered in the bicycle route planning. We calculated a weight using AHP(Analytic Hierarchy Process), make a database using GIS tool and deduced the routes applying calculated weight in this study. The result of this study, We could get reliable data as inspecting consistency of the research. Routes are deducted in the place where using demand is higher than arbitrarily chosen routes. Therefore, the route planning through AHP is expected to be utilized in area-specialty-reflected route planning or bicycle road alternatives testing.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.7 no.5
• /
• pp.53-63
• /
• 2008

In this paper, we propose the regional travel time measurement algorithm using the sequence pattern matching to the type of vehicles between the origin of the region and the end of the region, that could be able to overcome the limit of conventional method such as Probe Car Method or AVI Method by License Plate Recognition. This algorithm recognizes the vehicles as a sequence group with a definite length, and measures the regional travel time by searching the sequence of the origin which is the most highly similar to the sequence of the end. According to the assumption of similarity cost function, there are proposed three types of algorithm, and it will be able to estimate the average travel time that is the most adequate to the information providing period by eliminating the abnormal value caused by inflow and outflow of vehicles. In the result of computer simulation by the length of region, the number of passing cars, the length of sequence, and the average maximum error rate are measured within 3.46%, which means that this algorithm is verified for its superior performance.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.11 no.10
• /
• pp.1992-1998
• /
• 2007

It is important to obtain conn cytodiagnosis to classify background, cytoplasm, and nucleus from the diagnostic image. This study mose an algorithm that detects and classifies carcinoma cells of the uterine cervix in Pap smear using features of cervical cancer. It applies Median filter and Gaussian filter to get noise-removed nucleus area and also applies Kapur method in binarization of the resultant image. We apply 8-directional contour tracking algorithm and stretching technique to identify and revise clustered cells that often hinder to obtain correct analysis. The resulted nucleus area has distinguishable features such as cell size, integration rate, and directional coefficient from normal cells so that we can detect and classify carcinoma cells successfully. The experiment results show that the performance of the algorithm is competitive with human expert.

• Advances in aircraft and spacecraft science
• /
• v.3 no.1
• /
• pp.61-75
• /
• 2016

Climate changes brought on by human interventions have proved to be more devastating than predicted during the recent decades. Recognition of seriousness of the situation has led regulatory organisations to impose strict targets on allowable carbon dioxide emissions from automotive vehicles. As a possible solution, it has been proposed that Fibre Metal Laminate (FML) system is used to reduce the weight of future vehicles. To facilitate this investigation, FML based on steel and self-reinforced polypropylene was stamp formed into dome shapes under different blank holder forces (BHFs) at room temperature and its forming behaviour analysed. An open-die configuration was used in a hydraulic press so that a 3D photogrammetric measurement system (ARAMIS) could capture real-time surface strains. This paper presents findings on strain evolutions at different points along and at $45^{\circ}$ to fibre directions of circular FML blank, through various stages of forming. It was found initiation and rate of deformation varied with distance from the pole, that the mode of deformations range from biaxial stretching at the pole to drawing towards flange region, at decreasing magnitudes away from the pole in general. More uniform strain distribution was observed for the FML compared to that of plain steel and the most significant effects of BHF were its influence on forming depth and level of strain reached before failure.

• Annals of Surgical Treatment and Research
• /
• v.95 no.6
• /
• pp.324-332
• /
• 2018

Purpose: We present our experience involving the management of this disease, identifying prognostic factors affecting treatment outcomes. Methods: The patients treated for Fournier gangrene at our institution were retrospectively reviewed. Data collected included demographics, extent of soft tissue necrosis, predisposing factors, etiological factors, laboratory values, and treatment outcomes. The severity index and score were calculated. Multivariate regression analysis was used to determine the association between potential predictors and clinical outcomes. Results: A total of 41 patients (male:female = 33:8) were studied. The mean age was 54.4 years (range, 24-79 years). The most common predisposing factor was diabetes mellitus (n = 19, 46.3%). Sixteen patients (39.0%) were current smokers. Seven patients had chronic kidney disease. The most frequent etiology was urogenital lesion (41.5%). The mortality rate was 22.0% (n = 9). Multivariate regression analyses showed that extension of necrosis beyond perineal/inguinal area and pre-existing chronic kidney disease were significant and independent predictors of mortality. Extension of necrosis beyond perineal/inguinal area was a significant predictor of increased duration in the intensive care unit and hospital stay. In addition, pre-existing chronic kidney disease was a significant predictor of flap reconstruction in the wound. Conclusion: Fournier gangrene with extensive soft tissue necrosis and pre-existing chronic kidney disease was associated with poor prognosis and complexity of patient management. Early recognition of dissemination and premorbid renal function is essential to reduce mortality and establish a management plan for this disease.

• Journal of the Institute of Convergence Signal Processing
• /
• v.19 no.4
• /
• pp.180-185
• /
• 2018

We have developed an interactive signage system based on floating hologram by combining hologram technology and ICT technology, which can be competitive to small businesses that have excellent products and services. The developed interactive signage system can be used for publicity and marketing of small business owners at low cost, introducing menus with 3D hologram images, and providing various contents responding to user's hand movements. The developed system is able to detect 10 finger movements at a rate of 290 frames per second in a range of 60cm and a range of 150 degrees. We also confirmed that the virtual touch function operates normally by dividing the user's motion recognition into the hover zone and the touch zone by the physical motion experiment of the leap motion object.

• The Journal of Korean Institute of Information Technology
• /
• v.16 no.12
• /
• pp.75-83
• /
• 2018

Recently, autonomous vehicles have been actively studied. Pedestrian detection and recognition technology is important in autonomous vehicles. Pedestrian detection using CNN(Convolutional Neural Netwrok), which is mainly used recently, generally shows good performance, but there is a performance degradation depending on the environment of the image. In this paper, we propose a pedestrian detection system applying long-term memory structure of hippocampal neural network based on CNN network with LGP-FL (Local Gradient Pattern-Feature Layer) added. First, change the input image to a size of $227{\times}227$. Then, the feature is extracted through a total of 5 layers of convolution layer. In the process, LGP-FL adds the LGP feature pattern and stores the high-frequency pattern in the long-term memory. In the detection process, it is possible to detect the pedestrian more accurately by detecting using the LGP feature pattern information robust to brightness and color change. A comparison of the existing methods and the proposed method confirmed the increase of detection rate of about 1~4%.