• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.42 no.3 s.303
• /
• pp.107-114
• /
• 2005

Edge enhanced image is needed for processing images for special purpose such as a circuit diagram or a design composed of lines. Error diffusion halftoning, among digital halftoning methods to represent a continuous grayscale image for the binary output device such as printers, facsimiles, LCD televisions and etc. also makes edges of objects blurred. This paper proposes the method to enhance the edge of a binary image for the binary output device as well as a circuit diagram or a design. Based on that the human eyes perceive the local average luminance rather than the pixel's luminance itself, the proposed system uses a local activitymeasure (LAM), which is the difference between a pixel luminance and the average of its $3{\times}3$ neighborhood pixels' luminances weighted according to the spatial positioning. The system also usesinformation of edge enhancement(IEE), which is computed from the LAM multiplied by the average luminance. The IEE is added to the quantizer's input pixel and feeds into the halftoning quantizer. The quantizer produces the halftone image having the enhanced edge. The simulation results show that the proposed method produces more fine halftoning images than conventional methods due to the enhanced edges. Also the performance of the proposed method is compared with that of the conventional method by measuring the edge correlation and the local average accordance over a range of viewing distances.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2018.05a
• /
• pp.481-483
• /
• 2018

Recently, as the demand for electronic communication equipment increases, the importance of image and signal processing is increasing. However, noise is generated in digital signal due to various causes during transmission and reception, lowering equipment reliability and causing malfunction. Particularly, since AWGN may be found in most electronic equipments, AWGN removal is mandatorily performed as a preprocessing phase in various fields, such as image recognition, extraction, and segmentation. In the present paper, an AWGN removal algorithm which considers high frequency components is proposed. Conventional methods show relatively inadequate performance in images with high frequency components. To overcome this problem, proposed is a filter algorithm that add or subtract difference images in the local mask. And to verify performance of the proposed algorithm, PSNR and enlarged images are used to compare with the existing methods.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.9
• /
• pp.1076-1081
• /
• 2019

In response to the recent development of IT technologies, there are more demands for visual devices such as display. However, noise is generated in the process of sending video data due to various reasons. Noise is the representative noise which is commonly found. While A-TMF, CWMF, and AMF are the typical ways for removing Salt and Pepper noise, the noise is not removed well in high-density noise environment. To remove the noise in the high-density noise environment, this study suggested an algorithm which identifies whether it's noise or not. If it's not a noise, matches the original pixel. If it's a noise, divide the $3{\times}3$ local mask into the area of the element treated and the area of the element to be processed. Then, algorithm proposes to apply different weights for each element to treat it as an average filter. To analyze the performance of the algorithm, this study compared PSNR to compare the algorithm with other existing methods.

• Journal of Korea Water Resources Association
• /
• v.48 no.12
• /
• pp.1077-1087
• /
• 2015

The levee protect lifes, houses, and properties by blocking overflow of river. The revetment is forced to be covered on the slope of levee in order to prevent erosion. The stability of revetment is very important enough to directly connected to the stability of levee. In this study, the weak points of revetment on meandering channel were found by movable revetment experiment and the velocity and the water surface elevation (WSE) were measured at main points. The 3-D numerical simulations were performed under same conditions with experiment. And unclear flow characteristics by the limit of measuring instruments were analyzed through numerical simulation. Consequently, the section of large wall shear stress and the failure section are almost the same. Despite of small wall shear stress, the revetments located at right bank were carried away because of circulation zone due to secondary flow by meandering. With existing riprap design formula, the sizes of riprap determined using maximum local velocity were 1.5~4.7 times greater than them using mean velocity. As a result of this study, it is necessary to calculate the size of riprap in other ways for meandering and straight channel. At a later study, if the weighted value considered the radius of curvature and shape of hydraulic structure is applied to riprap design formula, it is expected that the size of revetment was evaluated rationally and accurately.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.41 no.3
• /
• pp.209-219
• /
• 2004

In order to implement a real time hardware for keyword spotting, we propose a Single-End-Point DTW(SEP-DTW) algorithm which is simple and less complex for computation. The SEP-DTW algorithm only needs a single end point which enables efficient applications, and it has a small wont of computations because the global search area is divided into successive local search areas. Also, we adopt new local constraints and a new distance measure for a better performance of the SEP-DTW algorithm. Besides, we make a normalization of feature same vectors so that they have the same variance in each frequency bin, and each frame has the same energy levels. To construct several reference patterns for each keyword, we use a clustering algorithm for all training patterns, and mean vectors in every cluster are taken as reference patterns. In order to detect a key word for input streams of speech, we measure the distances between reference patterns and input pattern, and we make a decision whether the distances are smaller than a pre-defined threshold value. With isolated speech recognition and keyword spotting experiments, we verify that the proposed algorithm has a better performance than other methods.

• Journal of Korea Multimedia Society
• /
• v.14 no.6
• /
• pp.742-751
• /
• 2011

This paper presents a robust 3D model hashing dependent on key and parameter by using heat kernel signature (HKS), which is special shape feature descriptor, In the proposed hashing, we calculate HKS coefficients of local and global time scales from eigenvalue and eigenvector of Mesh Laplace operator and cluster pairs of HKS coefficients to 2D square cells and calculate feature coefficients by the distance weights of pairs of HKS coefficients on each cell. Then we generate the binary hash through binarizing the intermediate hash that is the combination of the feature coefficients and the random coefficients. In our experiment, we evaluated the robustness against geometrical and topological attacks and the uniqueness of key and model and also evaluated the model space by estimating the attack intensity that can authenticate 3D model. Experimental results verified that the proposed scheme has more the improved performance than the conventional hashing on the robustness, uniqueness, model space.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.37 no.1
• /
• pp.1-10
• /
• 2000

In this paper, a new digital watermarking algorithm using wavelet transform is proposed. Wavelet transform is widely used for image processing, because of its multiresolution characteristic which conforms to the principles of the human visual system(HVS). It is also very efficient for localizing images in the spatial and frequency domain. Since wavelet coefficients can be characterized by the gaussian distribution, the proposed algorithm uses a gaussian distributed random vector as the watermark in order to achieve invisibility and robustness. After the original image is transformed using DWT(Discrete Wavelet Transform), the coefficients of all subbands including LL subband are utilized to equally embed the watermark to the whole image. To select perceptually significant coefficients for each subband, we use level-adaptive thresholding. The watermark is embedded to the selected coeffocoents, using different scale factors according to the wavelet characteristics. In the process of watermark detection, the similarity between the original watermark and the extracted watermark is calculated by using vector projection method. We analyze the performance of the proposed algorithm, compared with other transform-domain watermarking methods. The experimental results tested on various images show that the proposed watermark is less visible to human eyes and more robust to image compressions, image processings, geometric transformations and various noises, than the existing methods.

• Journal of Broadcast Engineering
• /
• v.20 no.6
• /
• pp.818-826
• /
• 2015

Recently, objective image quality assessment (IQA) methods that elaborately reflect the visual quality perception characteristics of human visual system (HVS) have actively been studied. Among those characteristics of HVS, luminance adaptation (LA) effect, indicating that HVS has different sensitivities depending on background luminance values to distortions, has widely been reflected into many existing IQA methods via Weber's law model. In this paper, we firstly reveal that the LA effect based on Weber's law model has inaccurately been reflected into the conventional IQA methods. To solve this problem, we firstly derive a new LA effect-based Local weight Function (LALF) that can elaborately reflect LA effect into IQA methods. We validate the effectiveness of our proposed LALF by applying LALF into SSIM (Structural SIMilarity) and PSNR methods. Experimental results show that the SSIM based on LALF yields remarkable performance improvement of 5% points compared to the original SSIM in terms of Spear rank order correlation coefficient between estimated visual quality values and measured subjective visual quality scores. Moreover, the PSNR (Peak to Signal Noise Ratio) based on LALF yields performance improvement of 2.5% points compared to the original PSNR.

• Proceedings of the Korean Society for Agricultural Machinery Conference
• /
• 2017.04a
• /
• pp.142-142
• /
• 2017

본 연구에서는 파프리카를 자동 수확하기 위한 시스템 개발의 일환으로 파프리카 재배환경에서 획득한 영상 내에 존재하는 파프리카 영역과 비 파프리카 영역의 RGB 정보를 입력으로 하는 인공신경망을 설계하고 학습을 수행하고자 하였다. 학습된 신경망을 이용하여 영상 내 파프리카 영역과 비 파프리카 영역의 구분이 가능 할 것으로 사료된다. 심층 신경망을 설계하기 위하여 MS Visual studio 2015의 C++, MFC와 Python 및 TensorFlow를 사용하였다. 먼저, 심층 신경망은 입력층과 출력층, 그리고 은닉층 8개를 가지는 형태로 입력 뉴런 3개, 출력 뉴런 4개, 각 은닉층의 뉴런은 5개로 설계하였다. 일반적으로 심층 신경망에서는 은닉층이 깊을수록 적은 입력으로 좋은 학습 결과를 기대 할 수 있지만 소요되는 시간이 길고 오버 피팅이 일어날 가능성이 높아진다. 따라서 본 연구에서는 소요시간을 줄이기 위하여 Xavier 초기화를 사용하였으며, 오버 피팅을 줄이기 위하여 ReLU 함수를 활성화 함수로 사용하였다. 파프리카 재배환경에서 획득한 영상에서 파프리카 영역과 비 파프리카 영역의 RGB 정보를 추출하여 학습의 입력으로 하고 기대 출력으로 붉은색 파프리카의 경우 [0 0 1], 노란색 파프리카의 경우 [0 1 0], 비 파프리카 영역의 경우 [1 0 0]으로 하는 형태로 3538개의 학습 셋을 만들었다. 학습 후 학습 결과를 평가하기 위하여 30개의 테스트 셋을 사용하였다. 학습 셋을 이용하여 학습을 수행하기 위해 학습률을 변경하면서 학습 결과를 확인하였다. 학습률을 0.01 이상으로 설정한 경우 학습이 이루어지지 않았다. 이는 학습률에 의해 결정되는 가중치의 변화량이 너무 커서 비용 함수의 결과가 0에 수렴하지 않고 발산하는 경향에 의한 것으로 사료된다. 학습률을 0.005, 0.001로 설정 한 경우 학습에 성공하였다. 학습률 0.005의 경우 학습 횟수 3146회, 소요시간 20.48초, 학습 정확도 99.77%, 테스트 정확도 100%였으며, 학습률 0.001의 경우 학습 횟수 38931회, 소요시간 181.39초, 학습 정확도 99.95%, 테스트 정확도 100%였다. 학습률이 작을수록 더욱 정확한 학습이 가능하지만 소요되는 시간이 크고 국부 최소점에 빠질 확률이 높았다. 학습률이 큰 경우 학습 소요 시간이 줄어드는 반면 학습 과정에서 비용이 발산하여 학습이 이루어지지 않는 경우가 많음을 확인 하였다.