• The Journal of the Korea institute of electronic communication sciences
• /
• v.14 no.2
• /
• pp.439-446
• /
• 2019

In this paper, we propose a encryption method using C-MLCA and 1D CAT to secure medical image for efficiently. First, we generate a state transition matrix using a Wolfram rule and create a sequence of maximum length. By operating the complemented vector, it converts an existing sequence to a more complex sequence. Then, we multiply the two sequences by rows and columns to generate C-MLCA basis images of the original image size and go through a XOR operation. Finally, we will get the encrypted image to operate the 1D CAT basis function created by setting the gateway values and the image which is calculated by transform coefficients. By comparing the encrypted image with the original image, we evaluate to analyze the histogram and PSNR. Also, by analyzing NPCR and key space, we confirmed that the proposed encryption method has a high level of stability and security.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.06a
• /
• pp.205-208
• /
• 2022

Multi-exposure high dynamic range (HDR) image reconstruction, the task of reconstructing an HDR image from multiple low dynamic range (LDR) images in a dynamic scene, often produces ghosting artifacts caused by camera motion and moving objects and also cannot deal with washed-out regions due to over or under-exposures. While there has been many deep-learning-based methods with motion estimation to alleviate these problems, they still have limitations for severely moving scenes. They also require large parameter counts, especially in the case of state-of-the-art methods that employ attention modules. To address these issues, we propose a frequency domain approach based on the idea that the transform domain coefficients inherently involve the global information from whole image pixels to cope with large motions. Specifically we adopt Residual Fast Fourier Transform (RFFT) blocks, which allows for global interactions of pixels. Moreover, we also employ Depthwise Overparametrized convolution (DO-conv) blocks, a convolution in which each input channel is convolved with its own 2D kernel, for faster convergence and performance gains. We call this LFFNet (Lightweight Frequency Fusion Network), and experiments on the benchmarks show reduced ghosting artifacts and improved performance up to 0.6dB tonemapped PSNR compared to recent state-of-the-art methods. Our architecture also requires fewer parameters and converges faster in training.

• The Journal of Engineering Geology
• /
• v.14 no.1
• /
• pp.29-46
• /
• 2004

It is important to identify geometries of fracture that act as a conduit of fluid flow for characterization of ground water flow in fractured rock. Fracture geometries control hydraulic conductivity and stream lines in a rock mass. However, we have difficulties to acquire whole geometric data of fractures in a field scale because of discontinuous distribution of outcrops and impossibility of continuous collecting of subsurface data. Therefore, it is needed to develop a method to describe whole feature of a target fracture geometry. This study suggests a new approach to develop a method to characterize on the whole feature of a target fracture geometry based on the Fourier transform. After sampling of specimens along a target fracture from borehole cores, effective frequencies among roughness components were selected by the Fourier transform on each specimen. Then, the selected effective frequencies were averaged on each frequency. Because the averaged spectrum includes all the frequency profiles of each specimen, it shows the representative components of the fracture roughness of the target fracture. The inverse Fourier transform is conducted to reconstruct an averaged whole roughness feature after low pass filtering. The reconstructed roughness feature also shows the representative roughness of the target subsurface fracture including the geometrical characteristics of each specimen. It also means that overall roughness feature by scaling up of a fracture. In order to identify the characteristics of permeability coefficients along the target fracture, fracture models were constructed based on the reconstructed roughness feature. The computation of permeability coefficient was performed by the homogenization analysis that can calculate accurate permeability coefficients with full consideration of fracture geometry. The results show a range between $10^{-4}{\;}and{\;}10^{-3}{\;}cm/sec$, indicating reasonable values of permeability coefficient along a large fracture. This approach will be effectively applied to the analysis of permeability characteristics along a large fracture as well as identification of the whole feature of a fracture in a field scale.

• 전자공학회논문지 IE
• /
• v.43 no.4
• /
• pp.65-75
• /
• 2006

This thesis deals with a wavelet-based method of denoising of infrared images contaminated with impulse noise and Gaussian noise, he method of thresholding the wavelet coefficients using derivatives and median absolute deviations of the wavelet coefficients of the detail subbands was proposed to effectively denoise infrared images with noises. Particularly, in order to eliminate the impulse noise the method of generating binary masks indicating locations of the impulse noise was selected. By this method, the threshold values dividing edges and noises were obtained more effectively proving the validity of the denoising method compared with the conventional wavelet shrinkage method.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.6 no.2
• /
• pp.95-106
• /
• 2007

This paper proposes an effective tempo feature extraction method for music information retrieval. The tempo information is modeled by the narrow-band temporal modulation components, which are decomposed into a modulation spectrum via joint frequency analysis. In implementation, the tempo feature is directly extracted from the modified discrete cosine transform coefficients, which is the output of partial MP3(MPEG 1 Layer 3) decoder. Then, different features are extracted from the amplitudes of modulation spectrum and applied to different music information retrieval tasks. The logarithmic scale modulation frequency coefficients are employed in automatic music emotion classification and music genre classification. The classification precision in both systems is improved significantly. The bit vectors derived from adaptive modulation spectrum is used in audio fingerprinting task That is proved to be able to achieve high robustness in this application. The experimental results in these tasks validate the effectiveness of the proposed tempo feature.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.12
• /
• pp.2641-2646
• /
• 2011

In this paper, we studied that the EZW algorithm based in HWT was proposed effective compression technique of wavelet transformed image. The proposed Haar-EZW algorithm is that image was coding by zerotree coding technique using self-similarity of HWT coefficients. If the HWT coefficient is larger than the threshold, that is coding to POS. If the HWT coefficient is smaller than the threshold, that is coding to NEG. If the HWT coefficient is larger than the root of zerotree, that is coding to ZTR. If the HWT coefficient is smaller then the threshold, and if that is not the root of zerotree, that is coding to IZ. This process is repeated until all the HWT coefficients have been encoded completely. This paper is compared Haar-EZW algorithm with Daubechies and Antonini. As the results of compare, it is shown that the PSNR of the Haar-EZW algorithm is better than Daubechies's and Antonini's.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.5
• /
• pp.152-159
• /
• 2014

In this paper, we introduce a run-length for wavelet coefficients and present a image splicing detection method using the statistical moments for the wavelet run-length. Various pre-processings for the suspicious image are performed to emphasize the discontinuous edges caused by the image splicing. The proposed scheme has the merit that can exploit the various statistical characteristics of the wavelet transform. We extracted up to 72 features, and performed training and testing using SVM(support vector machine). Experimental results showed that the proposed method generates similar detection results compared to the existing methods. In addition, we showed the wavelet domain run-length is useful to detect the spliced image.

• Journal of the Institute of Convergence Signal Processing
• /
• v.13 no.1
• /
• pp.6-11
• /
• 2012

Digital mammogram is effective for detecting the micro-calcification that is early symptom of breast cancer. In the digital mammogram, many image processing techniques have been studied for accurate diagnosis and efficient treatment of micro-calcification lesion. The wavelet based multi-scale method was mainly used to enhance the image contrast. This paper presents an advanced mammography enhancement method which is based both on the brightness and the contrast enhancement in the wavelet domain. The proposed method normalizes a dynamic range using histogram of the image. The brightness is enhanced by modifying coefficients of low frequency components, and the contrast is enhanced by coefficients of high frequency component based on the multi-scale contrast measure. The experiment results show that the proposed method yields better performance of the image enhancement over the existing methods.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.28 no.8C
• /
• pp.795-802
• /
• 2003

It is very important to construct statistical model in order to exactly estimate the signal variance from a noisy image. In order to estimate variance, information of neighboring region is used generally. The size of neighbor region is varied according to the regional characteristics of image. More accurate estimation of edge variance is due to smaller region of neighbor, on the other hands, larger region of neighbor is used to estimate the variance of flat region. By using estimated variance of original image, in general, Wiener filter is constructed, and it is applied to the noisy image. In this paper, we propose a new method for determining the range of neighbors to estimate the variance in wavelet domain. Firstly, a significance map is constructed using the parent-child relationship of wavelet domain. Based on the number of the significant wavelet coefficients, the range of neighbors is determined and then the variance of the original signal is estimated using ML(maximum likelihood method. Experimental results show that the proposed method yields better results than conventional methods for image denoising.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.10
• /
• pp.1961-1966
• /
• 2016

Compressive sensing has proved that a signal can be restored from less samples than the Nyquist rate. Reducing the required data rate is essential for a variety of fields including compression, transmission, and storage. It has been made lots of attempt to apply the compressive sensing theory into data intensive fields, such as image processing which needs to cover 4K and 8K pictures. In this paper, an image acquisition algorithm based on compressive sensing is proposed. It combines DCT, which can compact the energy of a image into a few coefficients, and the Noiselet transform, which is incoherent with DCT. The DCT coefficients represent the coarse structure of the images while the Noiselet information holds the fine details. Performance experiments with several images show that the proposed image acquisition algorithm not only outperforms the previous results, but also improves the reconstruction quality faster as the number of measurements increases.