• Journal of Broadcast Engineering
• /
• v.25 no.4
• /
• pp.569-575
• /
• 2020

The serial arc is one of factors causing electrical fires. Over past decades, various researches have been carried out to detect arc occurrences. Even though frequency analysis, wavelet, and statistical features have been used, additional steps such as transformation and feature extraction are required. On the contrary, deep learning models directly use the raw data without any feature extraction processes. Therefore, the usage of time-domain data is preferred, but the performance is not satisfactory. To solve this problem, subsequent 1-D signals are transformed into 2-D data that can feed into a convolutional neural network (CNN). Experiments validated that CNN model outperforms deep neural network (DNN) by the classification accuracy of 8.6%. In addition, data augmentation is utilized, resulting in the accuracy improvement by 14%.

• The Journal of the Korea Contents Association
• /
• v.11 no.8
• /
• pp.43-50
• /
• 2011

The scope of the problems that could be solved by monitoring and the improvement of the recognition time is directly correlated to the performance of the management function of the business process. However, the current event-managing monitoring system and the real-time advanced alarming business monitoring system decided whether to apply warnings or not by assuming a fixed environment and showing expressions based on the design rules. Therefore, there is a limit for distinguishing the range of occurrence and the level of severity in regard to the new external problems occurring in a complicated environment. Such problems cannot be abstracted. In this paper, evaluation model of business process contextual situations using goal scenario is suggested to provide constant services through the current monitoring process in regard to the service demands of the new scenario which occurs outside. The new demands based on the outside situation are analyzed according to the target scenario for the process activities. Also, a similar process model is found and identified by combining similarity and interrelationship. The process can be stopped in advance or adjusted to the wanted direction.

• The Journal of the Korea Contents Association
• /
• v.16 no.1
• /
• pp.52-59
• /
• 2016

For almost of concrete structures by placing in cold weather, it is very important that the selection of curing method at early aged construction stage. The Exothermic curing method with hot wire and rapid hardening cement is used mostly to prevent the initial cracks and the strength decrease. Most of the construction sites, however, have not been applied to the optimal curing method caused by the simple approaches and the empirical judgements. Therefore, this paper has proposed a evaluation algorithm of the exothermic curing method for representing heating temperature, period, position of hot wire by analyzing the transient heat transfer solution. This has been implemented, moreover, using an object oriented programming language to develop structural analysis system taking account risk parameters. This system is composed of input module, database module, database store module, analysis module, and result generation module. Linkage interface between the central database and each of the related module is implemented by the visual c# concept. Graphic user interface and the relational database table are supported for user's convenience.

• Journal of KIISE:Software and Applications
• /
• v.35 no.5
• /
• pp.306-312
• /
• 2008

In a goal-oriented dialogue, speaker's intention can be approximated by a semantic structure that consists of a pair of a speech act and a concept sequence. Therefore, it is very important to correctly identify the semantic structure of an utterance for implementing an intelligent dialogue system. In this paper, we propose a model to efficiently analyze the semantic structures based on an active teaming method. To reduce the burdens of high-level linguistic analysis, the proposed model only uses morphological features and previous semantic structures as input features. To improve the precisions of semantic structure analysis, the proposed model adopts CRFs(Conditional Random Fields), which show high performances in natural language processing, as an underlying statistical model. In the experiments in a schedule arrangement domain, we found that the proposed model shows similar performances(92.4% in speech act analysis and 89.8% in concept sequence analysis) to the previous models although it uses about a third of training data.

• Journal of Korea Multimedia Society
• /
• v.16 no.4
• /
• pp.430-443
• /
• 2013

The recognition of current position and road surrounding of car is very important to driver for safe driving. This paper presents the recognition technique of the road traveling environment using position information hiding and viewpoint transform that monitors the information of road furniture and signature and notifies them to driver. The proposed scheme generates the road images into which the position information are hided, from car camera and GPS module and provides the road information to driver through the viewpoint transformation and the road signature detection. The driving tests with camera and GPS module verified that the position information hiding takes about 66.5ms per frame, the detection rate of road signature is about 95.83%, and the road signature detection takes about 227.45ms per frame. Therefore, we know that the proposed scheme can recognize the road traveling environment on the road video with 15 frame rate.

• Journal of Korea Multimedia Society
• /
• v.16 no.9
• /
• pp.1031-1043
• /
• 2013

Extracting hand region from images is the first part in the process to recognize hand posture and gesture interaction. Therefore, a good segmenting method is important because it determines the overall performance of hand recognition systems. Conventional hand segmentation researches were prone to changing illumination conditions or limited to the ability to detect multiple people. In this paper, we propose a robust technique based on the fusion of skin-color data and depth information for hand segmentation process. The proposed algorithm uses skin-color data to localize accurate seed location for region-growing from a complicated background. Based on the seed location, our algorithm adjusts each detected blob to fill up the hole region. A region-growing algorithm is applied to the adjusted blob boundary at the detected depth image to obtain a robust hand region against illumination effects. Also, the resulting hand region is used to train our skin-model adaptively which further reduces the effects of changing illumination. We conducted experiments to compare our results with conventional techniques which validates the robustness of the proposed algorithm and in addition we show our method works well even in a counter light condition.

• Journal of The Korean Association of Information Education
• /
• v.8 no.1
• /
• pp.67-77
• /
• 2004

Recently, the rapid progress of internet technology stirs the widespread development in the educational areas. Internet has been exploited as an effective means to support school information management as well as to supplement traditional off-line teaching-learning methods. Science instruments are very important tools for science-related courses in the elementary schools. However, managing science instruments effectively has not been concerned greatly so far. Only off-line management systems were presented. In this paper, a Web-based science instrument management system is proposed. The proposed Web-based science instrument management system has the following characteristics. First, an extensive model for science instrument management is presented. The proposed model can be used as a standard model for effective management. Second, the system can provide inventory information as well as standard instrument requirements to teachers anytime anywhere. The system helps teachers make decisions on procurement plan. Finally, the user interface is designed to support easy navigation and simple use.

• Journal of Korea Multimedia Society
• /
• v.9 no.11
• /
• pp.1435-1444
• /
• 2006

We present an effective and real-time face detection method based on Principal Component Analysis(PCA) and Support Vector Machines(SVMs). We extract simple Haar-like features from training images that consist of face and non-face images, reinterpret the features with PCA, and select useful ones from the large number of extracted features. With the selected features, we construct a face detector using an SVM appropriate for binary classification. The face detector is not affected by the size of a training data set in a significant way, so that it showed 90.1 % detection rates with a small quantity of training data. it can process 8 frames per second for $320{\times}240$ pixel images. This is an acceptable processing time for a real-time system.

• Journal of Korea Multimedia Society
• /
• v.11 no.1
• /
• pp.53-60
• /
• 2008

Fingerprint verification method based on minutiae has been widely used for its speed and size stemming from utilizing only a few data, but it is vulnerable to some errors caused by the false minutiae extractions. A number of suggestions have been made to correct these problems. However, because it is very difficult to avoid all the external factors, such as noises that occur when fingerprints are collected, and all the internal factors that occur during the process of them, there is a limit in minimizing errors by improving the whole process. On the other hand, a fingerprint verification method based on images yields fewer errors because it makes direct comparison between images not using minutiae. Image-based fingerprint verification method has its limitations in aligning images with accuracy. Therefore, this paper proposes a new methodology to improve the performance utilizing both image-based and minutiae-based verification methods effectively. As the result of experimenting with both images and minutiae, the false accept rate and the false reject rate have been improved from 2.7% to 0.8% and from 6.5% to 5.5%, respectively.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.3
• /
• pp.556-563
• /
• 2006

We should extract lines and data which consist of the table in order to classify the table region and analyze its structure in document image. But it is difficult to extract lines and data exactly because the lines are cut and their lengths are changed, or characters or noises are merged to the table lines. These problems result from the error of image input device or image reduction. In this paper, we propose the better method of extracting lines and data for table region classification and structure analysis than the previous ones including commercial softwares. The prposed method extracts horizontal and vertical lines which consist of the table by the use of one dimensional median filter. This filter not only eliminates the noises which attach to the line and the lines which are orthogonal to the filtering direction, but also connects the cut line of which the gap is shorter than the length of the filter tap in the process of extracting lines to the filtering direction. Furthermore, texts attached to the line are separated in the process of extracting vertical lines. This is an example of ABSTRACT format.