• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.9C
• /
• pp.939-948
• /
• 2005

In EBCOT, the context extraction process takes excessive calculation time and this paper proposed a method to reduce this calculation time. That is, if a coefficient is less than a pre-defined threshold value the coefficient and its descendents skip the context extraction process. There is a trade-off relationship between the calculation time and the image quality or the amount of output data such that as this threshold value increases, the calculation time and the amount of output data decreases, but the image degradation increases. Therefore, by deciding this threshold value according to the network environments or conditions, it is possible to establish a network-adaptive context extraction method. The experimental results showed that the range of the threshold values for acceptable image quality(better than 30dB) is from 0 to 4. The experimental results showed that in this range the Resulting reduction rate in calculation time was from $3\%\;to\;64\%$ in average, the reduction rate in output data was from $32\%$ to $73\%$ in average, which means that large reduction in calculation time and output data can be obtained with a cost of an acceptable image quality degradation. Therefore, the proposed method is expected to be used efficiently in the application area such as the real-time image/video data communication in wireless environments, etc.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.49 no.2
• /
• pp.46-54
• /
• 2012

We propose a continuously varying contact pressure(CVCP)-adaptive feature extraction algorithm for pulse diagnostic analysis. The CVCP method measures the pulse waveform with continuously increasing contact pressure(CP). This method offer a high resolution signal of the pulse waveform amplitude(PWA) as a function of the contact pressure. Therefore it enables us to overcome the limitation of commercially available pulse-taking devices whose analysis rely on a few number of PWA-CP pairs. We show that an efficient feature extraction algorithm which covers the features of the CVCP-method can be developed by sequentially applying Fast Fourier Transform, peak detection by center-to-edges method, baseline drift removal, detection of the percussion wave upstroke by intersecting tangent method and detection of the analysis region. Finally, by a clinical study with 30 subjects, we show that our CVCP-adaptive feature extraction algorithm detected the upstroke with accuracy of 99.46% and sensitivity of 99.51%, which were about 4.82% and 2.46% increases respectively, compared to a conventional feature extraction method. The proposed CVCP method and the CVCP-adaptive feature extraction algorithm are expected to improve the accuracy in the pulse diagnostic algorithms such as floating/sunken pulse qualities and deficient/excess pulse qualities.

• Journal of the Korea Computer Graphics Society
• /
• v.29 no.1
• /
• pp.1-11
• /
• 2023

Deep-learning-based image segmentation is one of the most widely employed lane detection approaches, and it requires a post-process for extracting the key points on the lanes. A general approach for key-point extraction is using a fixed threshold defined by a user. However, finding the best threshold is a manual process requiring much effort, and the best one can differ depending on the target data set (or an image). We propose a novel key-point extraction algorithm that automatically adapts to the target image without any manual threshold setting. In our adaptive key-point extraction algorithm, we propose a line-level normalization method to distinguish the lane region from the background clearly. Then, we extract a representative key point for each lane at a line (row of an image) using a kernel density estimation. To check the benefits of our approach, we applied our method to two lane-detection data sets, including TuSimple and CULane. As a result, our method achieved up to 1.80%p and 17.27% better results than using a fixed threshold in the perspectives of accuracy and distance error between the ground truth key-point and the predicted point.

• Journal of Industrial Technology
• /
• v.31 no.B
• /
• pp.3-7
• /
• 2011

This paper addresses a computationally simple adaptive beamforming method to cancel interferences arriving onto a sensor array. In the proposed method, an estimate of the interference subspace is extracted from a submatrix of the sample covariance matrix and an orthonormal basis for the estimated subspace is efficiently found, one basis vector being updated every sample. Its computational burden is just $O(M{\eta})$ in an M-sensor array when ${\eta}$ directional signals are present. The new method does not make any premises of the geometrical structure of arrays, and can be applied to arbitrary arrays.

• Journal of the Korea Convergence Society
• /
• v.4 no.1
• /
• pp.33-41
• /
• 2013

This paper presents an adaptive recursive least squares algorithm (ARLS) for detecting voltage sag and voltage swell events in power systems. Different methods have been developed to detect voltage sag and voltage swell. Some of them use window techniques, which are too slow when voltage sag or swell mitigation is required. Others depend on the extraction of a single non-stationary sinusoidal signal out of a given multi-components input signal, and therefore they don't consider the harmonic components in calculating the voltage root mean square value (rms). The method, proposed in this paper, is capable of estimating the voltage rms taking into account all harmonic components. The method is tested by applying it to different, simulated signals using ATP program, and compared with voltage sag detection algorithms.

• Structural Engineering and Mechanics
• /
• v.45 no.6
• /
• pp.779-802
• /
• 2013

In this paper, adaptive neuro-fuzzy inference system (ANFIS) and artificial neural networks (ANNs) techniques are developed and applied to identify damage in a model steel girder bridge using dynamic parameters. The required data in the form of natural frequencies are obtained from experimental modal analysis. A comparative study is made using the ANNs and ANFIS techniques and results showed that both ANFIS and ANN present good predictions. However the proposed ANFIS architecture using hybrid learning algorithm was found to perform better than the multilayer feedforward ANN which learns using the backpropagation algorithm. This paper also highlights the concept of ANNs and ANFIS followed by the detail presentation of the experimental modal analysis for natural frequencies extraction.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.51 no.3
• /
• pp.108-111
• /
• 2002

This paper proposes an new water mark embedding and extraction technique which extends the direct sequence spread spectrum technique. The proposed technique approximates the complexity of image and block in spatial domain using Laplacian filtering and watermark is adaptively embedded in the mid-frequency DCT components. Local parity bits are attached to higher-frequency DCT components and they are used to detect extraction errors and correct those errors. In extraction process the proposed method boosts the higher frequency components of image and extracts the watermark by demodulation and this information is verified and adjusted by parity bits. Experimental results show it is invisible and robust to several external attacks.

• Proceedings of the IEEK Conference
• /
• 2003.11a
• /
• pp.521-524
• /
• 2003

In this paper, we propose a face region extraction algorithm using color information and projection. After the extraction of face candidate image using adaptive color information, we project it into vertical direction to estimate the width of the face. Then the redundant parts of the face are efficiently removed by using the estimated face width. And the width information of the face is used at the horizontal projection step to extract the height of the face, and non-face region such as the neck and some background regions, which are represented as the similar skin color, effectively eliminated. From the experiment results for the various images, the proposed algorithm shows more accurate results than the conventional algorithm.

• MALSORI
• /
• no.59
• /
• pp.13-25
• /
• 2006

In previous study, the linlog(linear log) RASTA(J-RASTA) approach based on PLP was proposed to deal with both the channel effect and the additive noise. The extraction of PLP required generally more steps and computation than the extraction of widely used MFCC. Thus, in this paper, we apply the linlog function to the MFCC for investigating the possibility of simple compensation method that removes both distortion. With the experimental results, the proposed method shows the similar tendency to the linlog RASTA-PLP_ When the J value is set to le-6, the best ERR(Error Reduction Rate) of 33% is obtained. For applying the linlog function to the feature extraction process, the J value plays a very important role in compensating the corruption. Thus, the study for the adaptive J or noise dependent J estimation is further required.

• Journal of Korea Society of Industrial Information Systems
• /
• v.16 no.2
• /
• pp.31-38
• /
• 2011

Skin color is a very important information for an automatic face recognition. In this paper, we proposed a skin region extraction method using the MLP(Multi-Layer Perceptron) and skin color model. We use the adaptive lighting compensation technique for improved performance of skin region extraction. Also, using an preprocessing filter, normally large areas of easily distinct non-skin pixels, are eliminated from further processing. Experimental results show that the proposed method has better performance than the conventional methods, and reduces processing time by 31~49% on average.