• Journal of Korean Society of Transportation
• /
• v.29 no.4
• /
• pp.125-138
• /
• 2011

The highway visibility reduction caused by fog is one of the major elements of traffic accidents. Though the fog warning systems can lead drivers into safe driving by letting them aware dangerous situations in advance, the optical sensors, such as fog sensor, has been extremely costly. Through recent studies, it is delivered that visibility measurements have become obtainable with relatively cheap cameras and their functionality is as similar as a driver' visual sense. Those measurements however require additional signs or ROI, so it is still costly and unable to utilize the conventional images from the existing systems. This study proposes a new method to detect the visibility in real time based on the conventional images from the existing CCTV cameras. The proposed method builds a road model and extracts and applies vehicle movements and visible lines - those highlight easy and quick visibility measurements. The proposed method has advantages of both (1) having possible day and night visibility measurements similar to drivers' visual sense and (2) being easily applied to the existing CCTV system without additional devices. This paper presents field experiments using images acquired from the Central Inland Expressway and discusses future research directions.

• Journal of the Korea Computer Industry Society
• /
• v.7 no.5
• /
• pp.563-572
• /
• 2006

Advantage that light transmission treatment way of most suitable through skin can investigate light directly in part ar there is difference in ability photoelectricity month by diverse complexion of horn character department which is branch or head of a family outside part of skin and treatment according to various patient can be inappropriate. By result that this research uses color information after search each color ingredient that ingredient of HIS and YIQ that use method, color information to use skin impedance way and color information through skin area ion and difference video to do fixed measuring by light investigation way by skin impedance corresponds to skin color in an experiment though is most universal result according to patient's skin model area detection each single person's skin model through videotex automatically create and because measuring, investigate skin color, energy, wave length, approximately, transmission time, model of most suitable that draw pulse delay and so on and want and special quality, and saved standard of disease treatment pulse modeling by skin impedance, and design and manufacture light investigation pulse modeling system of most suitable by skin subordinate, and constructed suitable treatment pulse database by skin color.

• Journal of Advanced Navigation Technology
• /
• v.18 no.2
• /
• pp.151-157
• /
• 2014

Recently UAV(unmanned aerial vehicle) is frequently used not only for military purpose but also for civil purpose. UAV automatically navigates following the coordinates input in advance using GPS information. However it is impossible when GPS cannot be received because of jamming or external interference. In order to solve this problem, we propose a real-time segmentation and classification algorithm for the specific regions from UAV image in this paper. We use the super-pixels algorithm using graph-based image segmentation as a pre-processing stage for the feature extraction. We choose the most ideal model by analyzing various color models and mixture color models. Also, we use support vector machine for classification, which is one of the machine learning algorithms and can use small quantity of training data. 18 color and texture feature vectors are extracted from the UAV image, then 3 classes of regions; river, vinyl house, rice filed are classified in real-time through training and prediction processes.

• Journal of Advanced Navigation Technology
• /
• v.17 no.1
• /
• pp.115-122
• /
• 2013

Disease diagnostics based on medical imaging is getting popularity day by day. Presence of the atherosclerosis is one of the causes of narrowing of carotid arteries which may block partially or fully blood flow into the brain. Serious brain strokes may occur due to such types of blockages in blood flow. Early detection of the plaque and taking precautionary steps in this regard may prevent from such type of serious strokes. In this paper, we present an automatic image segmentation technique for carotid artery ultrasound images based on active contour approach. In our experimental study, we assume that ultrasound images are properly aligned before applying automatic image segmentation. We have successfully applied the automatic segmentation of carotid artery ultrasound images using snake based model. Qualitative comparison of the proposed approach has been made with the manual initialization of snakes for carotid artery image segmentation. Our proposed approach successfully segments the carotid artery images in an automated way to help radiologists to detect plaque easily. Obtained results show the effectiveness of the proposed approach.

• Journal of Korea Multimedia Society
• /
• v.11 no.11
• /
• pp.1471-1479
• /
• 2008

In this paper, we propose a technique to improve Out-Of-Vocabulary(OOV) rejection algorithms in variable vocabulary recognition system which is much used in ASR(Automatic Speech Recognition). The rejection system can be classified into two categories by their implementation method, keyword spotting method and utterance verification method. The utterance verification method uses the likelihood ratio of each phoneme Viterbi score relative to anti-phoneme score for deciding OOV. In this paper, we add speaker verification system before utterance verification and calculate an speaker verification probability. The obtained speaker verification probability is applied for determining the proposed variable-confidence threshold. Using the proposed method, we achieve the significant performance improvement; CA(Correctly Accepted for keyword) 94.23%, CR(Correctly Rejected for out-of-vocabulary) 95.11% in office environment, and CA 91.14%, CR 92.74% in noisy environment.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.19 no.3
• /
• pp.651-658
• /
• 2015

Recently, lots of research works have been actively focused on the self-driving car. In order to implement the self-driving car, lots of fusion techniques should be merged and, specially, it is noted that a vehicle-attached camera can provide several useful functionalities such as traffic lights recognition, pedestrian detection, stop-line recognition including simple driving records. Accordingly, as one of the efficient tools for the self-driving car implementation, this paper proposes a mathematical model for estimating effectively the road distance with a vehicle-attached black box camera. The proposed model can be effectively used for estimating the road distance by using the height of black box camera or the widths of the referenced road line and the observed road line. Through several simulations, it is shown that the proposed model is effective in estimating the road distance.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.1
• /
• pp.237-243
• /
• 2017

In this paper, the detection performance of the electro-optical telescopes which observes and surveils space objects including artificial satellites, is analyzed. To perform the Modeling & Simulation(M&S) based analysis, satellite orbit model, telescope model, and the atmospheric model are constructed and a detection scenario observing the satellite is organized. Based on the organized scenario, pointing accuracy is analyzed according to the Field of View(FOV), which is one of the key factors of the telescope, considering pointing angle command error. In accordance with the preceding result, detection possibility according to the pixel-count of the detector and the FOV of the telescope is analyzed by discerning detection by Signal-to-Noise Ratio(SNR). The result shows that pointing accuracy increases with larger FOV, whereas the detection probability increases with smaller FOV and higher pixel-count. Therefore, major specification of the telescope such as FOV and pixel-count should be determined considering the result of M&S based analysis performed in this paper and the operational circumstances.

• Proceedings of the Korea Contents Association Conference
• /
• 2005.11a
• /
• pp.633-638
• /
• 2005

The extraction of a face is a very important part for human interface, biometrics and security. In this paper, we applies DCM(Dilation of Color and Motion) filter and Active Contour Models to extract facial outline. First, DCM filter is made by applying morphology dilation to the combination of facial color image and differential image applied by dilation previously. This filter is used to remove complex background and to detect facial outline. Because Active Contour Models receive a large effect according to initial curves, we calculate rotational degree using geometric ratio of face, eyes and mouth. We use edgeness and intensity as an image energy, in order to extract outline in the area of weak edge. We acquire various head-pose images with both eyes from five persons in inner space with complex background. As an experimental result with total 125 images gathered by 25 per person, it shows that average extraction rate of facial outline is 98.1% and average processing time is 0.2sec.

• The KIPS Transactions:PartB
• /
• v.12B no.3 s.99
• /
• pp.265-272
• /
• 2005

In this paper, we propose a technique which is speedily searching for correspondent points of two images using Voronoi-Distance, as an image registration method for feature based image mosaics. It extracts feature points in two images by the SUSAN corner detector, and then create not only the Voronoi Surface which has distance information among the feature points in the base image using a priority based Voronoi distance algorithm but also select the model area which has the maximum variance value of coordinates of the feature points in the model image. We propose a method for searching for the correspondent points in the Voronoi surface of the base image overlapped with the model area by use of the partitive search algorithm using queues. The feature of the method is that we can rapidly search for the correspondent points between adjacent images using the new Voronoi distance algorithm which has $O(width{\times}height{\times}logN)$ time complexity and the the partitive search algerian using queues which reduces the search range by a fourth at a time.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.4C
• /
• pp.388-396
• /
• 2009

This paper proposes a new fire-smoke detection method by using extracted features from camera images and pattern recognition technique. First, moving regions are detected by analyzing the frame difference between two consecutive images and generate candidate smoke regions by applying smoke color model. A smoke region generally has a few characteristics such as similar color, simple texture and upward motion. From these characteristics, we extract brightness, wavelet high frequency and motion vector as features. Also probability density functions of three features are generated using training data. Probabilistic models of smoke region are then applied to observation nodes of our proposed Dynamic Bayesian Networks (DBN) for considering time continuity. The proposed algorithm was successfully applied to various fire-smoke tasks not only forest smokes but also real-world smokes and showed better detection performance than previous method.