• Journal of KIISE:Software and Applications
• /
• v.36 no.2
• /
• pp.169-177
• /
• 2009

In this paper, we propose a new shot boundary detection method which is optimized for news video story parsing. This new news shot boundary detection method was designed to satisfy all the following requirements: 1) minimizing the incorrect data in data set for anchor shot detection by improving the recall ratio 2) detecting abrupt cuts and gradual transitions with one single algorithm so as to divide news video into shots with one scan of data set; 3) classifying shots into static or dynamic, therefore, reducing the search space for the subsequent stage of anchor shot detection. The proposed method, based on singular value decomposition with incremental clustering and mercer kernel, has additional desirable features. Applying singular value decomposition, the noise or trivial variations in the video sequence are removed. Therefore, the separability is improved. Mercer kernel improves the possibility of detection of shots which is not separable in input space by mapping data to high dimensional feature space. The experimental results illustrated the superiority of the proposed method with respect to recall criteria and search space reduction for anchor shot detection.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.48 no.4
• /
• pp.116-128
• /
• 2011

In this paper, we propose the structure of a high-performance face-detection engine that responds well to facial rotating changes using rotation transformation which minimize the required memory usage compared to the previous face-detection engine. The validity of the proposed structure has been verified through the implementation of FPGA. For high performance face detection, the MCT (Modified Census Transform) method, which is robust against lighting change, was used. The Adaboost learning algorithm was used for creating optimized learning data. And the rotation transformation method was added to maintain effectiveness against face rotating changes. The proposed hardware structure was composed of Color Space Converter, Noise Filter, Memory Controller Interface, Image Rotator, Image Scaler, MCT(Modified Census Transform), Candidate Detector / Confidence Mapper, Position Resizer, Data Grouper, Overlay Processor / Color Overlay Processor. The face detection engine was tested using a Virtex5 LX330 FPGA board, a QVGA grade CMOS camera, and an LCD Display. It was verified that the engine demonstrated excellent performance in diverse real life environments and in a face detection standard database. As a result, a high performance real time face detection engine that can conduct real time processing at speeds of at least 60 frames per second, which is effective against lighting changes and face rotating changes and can detect 32 faces in diverse sizes simultaneously, was developed.

• Korean Journal of Cognitive Science
• /
• v.16 no.4
• /
• pp.305-322
• /
• 2005

Watermarking is a widely employed method tn protecting copyright of a digital image, the owner's unique image is embedded into the original image. Strengthened level of watermark insertion would help enhance its resilience in the process of extraction even from various distortions of transformation on the image size or resolution. However, its level, at the same time, should be moderated enough not to reach human visibility. Finding a balance between these two is crucial in watermarking. For the algorithm for watermarking, the predefined strength of a watermark, computed from the physical difference between the original and embedded images, is applied to all images uniformal. The mean brightness or contrast of the surrounding images, other than the absolute brightness of an object, could affect human sensitivity for object detection. In the present study, we examined whether the detectability for watermark noise might be attired by image statistics: mean brightness and contrast of the image. As the first step to examine their effect, we made rune fundamental images with varied brightness and control of the original image. For each fundamental image, detectability for watermark noise was measured. The results showed that the strength ot watermark node for detection increased as tile brightness and contrast of the fundamental image were increased. We have fitted the data to a regression line which can be used to estimate the strength of watermark of a given image with a certain brightness and contrast. Although we need to take other required factors into consideration in directly applying this formula to actual watermarking algorithm, an adaptive watermarking algorithm could be built on this formula with image statistics, such as brightness and contrast.

• KIPS Transactions on Software and Data Engineering
• /
• v.3 no.9
• /
• pp.361-368
• /
• 2014

The LPR(License plate recognition) system has been developed to efficient control for complex traffic environment and currently be used in many places. However, because of light, noise, background changes, environmental changes, damaged plate, it only works limited environment, so it is difficult to use in real-time. This paper presents a heuristic segmentation algorithm for robust to noise and illumination changes and introduce a real-time license plate recognition system using it. In first step, We detect the plate utilized Haar-like feature and Adaboost. This method is possible to rapid detection used integral image and cascade structure. Second step, we determine the type of license plate with adaptive histogram equalization, bilateral filtering for denoise and segment accurate character based on adaptive threshold, pixel projection and associated with the prior knowledge. The last step is character recognition that used histogram of oriented gradients (HOG) and multi-layer perceptron(MLP) for number recognition and support vector machine(SVM) for number and Korean character classifier respectively. The experimental results show license plate detection rate of 94.29%, license plate false alarm rate of 2.94%. In character segmentation method, character hit rate is 97.23% and character false alarm rate is 1.37%. And in character recognition, the average character recognition rate is 98.38%. Total average running time in our proposed method is 140ms. It is possible to be real-time system with efficiency and robustness.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.22 no.2
• /
• pp.233-242
• /
• 2018

Previous works for detecting arrhythmia have mostly used nonlinear method such as artificial neural network, fuzzy theory, support vector machine to increase classification accuracy. Most methods require higher computational cost and larger processing time. Therefore it is necessary to design efficient algorithm that classifies PVC(premature ventricular contraction) and decreases computational cost by accurately detecting feature point based on only R peak through optimal R wave. For this purpose, we detected R wave through optimal threshold value and extracted RR interval and R peak pattern from noise-free ECG signal through the preprocessing method. Also, we classified PVC in realtime through RR interval and R peak pattern. The performance of R wave detection and PVC classification is evaluated by using 9 record of MIT-BIH arrhythmia database that included over 30. The achieved scores indicate the average of 99.02% in R wave detection and the rate of 94.85% in PVC classification.

• Smart Structures and Systems
• /
• v.20 no.4
• /
• pp.427-440
• /
• 2017

One of the suitable structural damage detection methods using vibrational characteristics are damage-index-based methods. In this study, a damage index for identifying damages in plate structures using frequency response function (FRF) data has been provided. One of the significant challenges of identifying the damages in plate structures is high number of degrees of freedom resulting in decreased damage identifying accuracy. On the other hand, FRF data are of high volume and this dramatically decreases the computing speed and increases the memory necessary to store the data, which makes the use of this method difficult. In this study, FRF data are compressed using two-dimensional principal component analysis (2D-PCA), and then converted into damage index vectors. The damage indices, each of which represents a specific condition of intact or damaged structures are stored in a database. After computing damage index of structure with unknown damage and using algorithm of lookup tables, the structural damage including the severity and location of the damage will be identified. In this study, damage detection accuracy using the proposed damage index in square-shaped structural plates with dimensions of 3, 7 and 10 meters and with boundary conditions of four simply supported edges (4S), three clamped edges (3C), and four clamped edges (4C) under various single and multiple-element damage scenarios have been studied. Furthermore, in order to model uncertainties of measurement, insensitivity of this method to noises in the data measured by applying values of 5, 10, 15 and 20 percent of normal Gaussian noise to FRF values is discussed.

• Journal of Advanced Navigation Technology
• /
• v.14 no.4
• /
• pp.553-561
• /
• 2010

This paper proposed the face tracking method which can be effectively applied to the robot's vision system. The proposed algorithm tracks the facial areas after detecting the area of video motion. Movement detection of video images is done by using median filter and erosion and dilation operation as a method for removing noise, after getting the different images using two continual frames. To extract the skin color from the moving area, the color information of sample images is used. The skin color region and the background area are separated by evaluating the similarity by generating membership functions by using MIN-MAX values as fuzzy data. For the face candidate region, the eyes are detected from C channel of color space CMY, and the mouth from Q channel of color space YIQ. The face region is tracked seeking the features of the eyes and the mouth detected from knowledge-base. Experiment includes 1,500 frames of the video images from 10 subjects, 150 frames per subject. The result shows 95.7% of detection rate (the motion areas of 1,435 frames are detected) and 97.6% of good face tracking result (1,401 faces are tracked).

• Journal of KIISE:Software and Applications
• /
• v.31 no.9
• /
• pp.1171-1179
• /
• 2004

We describe word boundary detection that extracts the boundary between speech and non-speech. The proposed method uses two features. One is the normalized root mean square of speech signal, which is insensitive to white noises and represents temporal information. The other is the normalized met-frequency band energy of voice signal, which is frequency information of the signal. Our method detects word boundaries using a recurrent fuzzy associative memory(RFAM) that extends FAM by adding recurrent nodes. Hebbian learning method is employed to establish the degree of association between an input and output. An error back-propagation algorithm is used for teaming the weights between the consequent layer and the recurrent layer. To confirm the effectiveness, we applied the suggested system to voice data obtained from KAIST.

• Journal of Korea Multimedia Society
• /
• v.15 no.11
• /
• pp.1305-1318
• /
• 2012

Image segmentation is one of the difficult research problems in machine vision and pattern recognition field. A commonly used segmentation method is the Otsu method. It is simpler and easier to implement but it fails if the histogram is unimodal or similar to unimodal. And if some target area is smaller than background object, then its histogram has the distribution close to unimodal. In this paper, we proposed an improved image segmentation method based on 1D Otsu method for a small target detection. To overcome drawbacks by unimodal histogram effect, we depressed the background histogram using a logarithm function. And to improve a signal to noise ratio, we used a local average value by the neighbor window for thresholding using 1D Otsu method. The experimental results show that our proposed algorithm performs better segmentation result than a traditional 1D Otsu method, and needs much less computational time than that of the 2D Otsu method.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.29 no.10
• /
• pp.791-798
• /
• 2018

The CFAR algorithm is mainly used for target detection in radar systems. In particular, OS CFAR is used in a non-uniform noise environment. However, it requires a large amount of computation, because it should sort reference cells in ascending order. This makes it difficult to apply the radar system in real time. In this paper, we describe how to reduce the computational burden of OS CFAR. We compared the power of the test cell and reference cell to determine only the presence or absence of target detection. The common reference cells overlapping in the reference cells of the three test cells are obtained. We first compare the test cell with the highest power value among the three test cells to the common reference cells. Next, we compare each test cell to general reference cells, excluding the common reference cells. The computation time is shortened by reducing the power comparison computation amounts.