• ETRI Journal
• /
• v.35 no.6
• /
• pp.1105-1114
• /
• 2013

Owing to the large number of video programs available, a method for accessing preferred videos efficiently through personalized video summaries and clips is needed. The automatic recognition of user states when viewing a video is essential for extracting meaningful video segments. Although there have been many studies on emotion recognition using various user responses, electroencephalogram (EEG)-based research on preference recognition of videos is at its very early stages. This paper proposes classification models based on linear and nonlinear classifiers using EEG features of band power (BP) values and asymmetry scores for four preference classes. As a result, the quadratic-discriminant-analysis-based model using BP features achieves a classification accuracy of 97.39% (${\pm}0.73%$), and the models based on the other nonlinear classifiers using the BP features achieve an accuracy of over 96%, which is superior to that of previous work only for binary preference classification. The result proves that the proposed approach is sufficient for employment in personalized video segmentation with high accuracy and classification power.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1997.04a
• /
• pp.719-723
• /
• 1997

The welding automation in ship manufacturing process,especially in the sub-assembly line is considered to be a difficult job because the welding part is too huge, various, unstructured for a welding robot to weld fully automatically. The weld orocess at the sub-assembly line for ship manufacturing is to joint the various stiffener on the base panel. In order to realize automatic robot weld in sub-assembly line, robot have to equip with the sensing system to recognize the position of the parts. In this research,we developed a vision system to detect the position of base panle for sub-assembly line is shipbuilding process. The vision system is composed of one CCD camera attached on the base of robot, 2-500W halogen lamps for active illumination. In the image processing algorithm,the base panel is represented by two set of lines located at its two corner through hough transform. However, the various noise line caused by highlight,scratches and stiffener,roller in conveyor, and so on is contained in the captured image, this nosie can be eliminated by region segmentation and threshold in hough transform domain. The matching process to recognize the position of weld panel is executed by finding patterns in the Hough transformed domain. The sets of experiments performed in the sub-assembly line show the effectiveness of the proposed algorithm.

• Journal of the Korea Society of Computer and Information
• /
• v.9 no.4 s.32
• /
• pp.133-139
• /
• 2004

In this paper. we propose a medical image processing method for detection of periodontal disease by geometric analysis on dental digital radiography. This paper proposes the method of an automatic image alignment and detection of minute changes, to overcome defects in the conventional subtraction radiography by image processing technique, that is necessary for getting subtraction image and ROI(Region Of Interest) focused on a selection method using the geometric features in target images. Therefore, we use these methods because they give accuracy, consistency and objective information or data to results. In result, easily and visually we can identify minute differences in the affected parts whether they have problems or not, and using application system.

• Journal of the Optical Society of Korea
• /
• v.14 no.4
• /
• pp.376-382
• /
• 2010

We propose a novel method to precisely detect car license plates by locating main characters, which are printed with large font size. The regions of the main characters are directly detected without detecting the plate region boundaries, so that license regions can be detected more precisely than by other existing methods. To generate a binary image, multiple thresholds are applied, and segmented regions are selected from multiple binarized images by a criterion of size and compactness. We do not employ any character matching methods, so that many candidates for main character groups are detected; thus, we use a neural network to reject non-main character groups from the candidates. The relation of the character regions and the intensity statistics are used as the input to the neural network for classification. The detection performance has been investigated on real images captured under various illumination conditions for 1000 vehicles. 980 plates were correctly detected, and almost all non-detected plates were so stained that their characters could not be isolated for character recognition. In addition, the processing time is fast enough for a commercial automatic license plate recognition system. Therefore, the proposed method can be used for recognition systems with high performance and fast processing.

• Journal of Biosystems Engineering
• /
• v.17 no.4
• /
• pp.354-362
• /
• 1992

This study was intended to examine feasibility of sizing and color grading of Fuji apple with black/white image processing system, to develop a device with which the whole surface of an apple could be captured by one camera, and to develop an algorithm for a high speed sorting. The results are summarized as follows : 1. The black/white image processing system used in this study showed a maximum error of 1.3% in area measurement with a reference figure while the focusing point of camera and location of the reference figure were changed within a certain range. 2. As the result of evaluating four automatic image segmentation algorithms with apple images, Histogram Clustering Method was the best in terms of computation time and accuracy. 3. The fast algorithm for analyzing size and coloration of apple was developed. 4. The whole surface of an apple could be captured in an image frame with two mirrors installed on the both sides of the sample. The total area of the image representing the whole surface showed a correlation of 0.995 with the weight of apple. 5. The gray level when a particular band pass filter was mounted on the camera showed high correlation with 'L' and 'a' values of Hunt color scale and could represent the coloration of apple.

• Journal of the Korean Society for Precision Engineering
• /
• v.16 no.5 s.98
• /
• pp.48-57
• /
• 1999

Applications of thresholding technique are based on the assumption that object and background pixels in a digital image can be distinguished by their gray level values. For the segmentation of more complex images, it is necessary to resort to multiple threshold selection techniques. This paper describes a new method for multiple threshold selection of gray level images which are not clearly distinguishable from the background. The proposed method consists of three main stages. In the first stage, a probability distribution function for a gray level histogram of an image is derived. Cluster points are defined according to the probability distribution function. In the second stage, fuzzy partition matrix of the probability distribution function is generated through the fuzzy clustering process. Finally, elements of the fuzzy partition matrix are classified as clusters according to gray level values by using max-membership method. Boundary values of classified clusters are selected as multiple threshold. In order to verify the performance of the developed algorithm, automatic inspection process of ball grid array is presented.

• Journal of KIISE:Computing Practices and Letters
• /
• v.7 no.6
• /
• pp.582-592
• /
• 2001

Video indexing plays an important role in the applications such as digital video libraries or web VOD which archive large volume of digital videos. Video indexing is usually based on video segmentation. In this paper, we propose a software tool called V2Web Studio which can generate video clips utilizing shot boundary detection algorithm. With the V2Web Studio, the process of clip generation consists of the following four steps: 1) Automatic detection of shot boundaries by parsing the video, 2) Elimination of errors by manually verifying the results of the detection, 3) Building a modeling structure of logical hierarchy using the verified shots, and 4) Generating multiple video clips corresponding to each logically modeled segment. The aforementioned steps are performed by shot detector, shot verifier, video modeler and clip generator in the V2Web Studio respectively.

• Phonetics and Speech Sciences
• /
• v.7 no.2
• /
• pp.149-156
• /
• 2015

We propose a new method for automatic fluency scoring of English speaking tests spoken by nonnative speakers in a free-talking style. The proposed method is different from the previous methods in that it does not require the transcribed texts for spoken utterances. At first, an input utterance is segmented into a phone sequence by using a phone recognizer trained by using native speech databases. For each utterance, a feature vector with 6 features is extracted by processing the segmentation results of the phone recognizer. Then, fluency score is computed by applying support vector regression (SVR) to the feature vector. The parameters of SVR are learned by using the rater scores for the utterances. In computer experiments with 3 tests taken by 48 Korean adults, we show that speech rate, phonation time ratio, and smoothed unfilled pause rate are best for fluency scoring. The correlation of between the rater score and the SVR score is shown to be 0.84, which is higher than the correlation of 0.78 among raters. Although the correlation is slightly lower than the correlation of 0.90 when the transcribed texts are given, it implies that the proposed method can be used as a preprocessing tool for fluency evaluation of speaking tests.

• ETRI Journal
• /
• v.33 no.3
• /
• pp.423-433
• /
• 2011

This paper introduces an adaptive and integrated utterance verification (UV) framework using minimum verification error (MVE) training as a new set of solutions suitable for real applications. UV is traditionally considered an add-on procedure to automatic speech recognition (ASR) and thus treated separately from the ASR system model design. This traditional two-stage approach often fails to cope with a wide range of variations, such as a new speaker or a new environment which is not matched with the original speaker population or the original acoustic environment that the ASR system is trained on. In this paper, we propose an integrated solution to enhance the overall UV system performance in such real applications. The integration is accomplished by adapting and merging the target model for UV with the acoustic model for ASR based on the common MVE principle at each iteration in the recognition stage. The proposed iterative procedure for UV model adaptation also involves revision of the data segmentation and the decoded hypotheses. Under this new framework, remarkable enhancement in not only recognition performance, but also verification performance has been obtained.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.3
• /
• pp.871-879
• /
• 1997

The comversion of printed maps into computerixed data bases is an enormous rask. Thus the autmaotion of the conversion process is essential. Efficient computer representation of printed maps and line drawings depends on codes assigened to chracaters, symbools, and vestor representation of the graphics. In many cases, maps ard constructed in a number of layers, where each layer is printed in a distinct color, and it represents a subste of the map infromation. In order to properly repressnet road information from color map images, an automatic road extraction algorithm is proposed. Road image is separated from graghics by color segmentation, and then restored by the proposed concurrent conditional dilation operation. The internal and external noise of the road image is eliminated by opening and closing operation. By thining and vectorizing line segments, the desited road information is extracted.