• KIPS Transactions on Software and Data Engineering
• /
• v.3 no.10
• /
• pp.429-440
• /
• 2014

The iris recognition is a biometrics technology to extract and code an unique iris feature from human eye image. Also, it includes the technology to compare with other's various iris stored in the system. On the other hand, eyelashes in iris image are a external factor to affect to recognition rate of iris. If eyelashes are not removed exactly from iris area, there are two false recognitions that recognize eyelashes to iris features or iris features to eyelashes. Eventually, these false recognitions bring out a lot of loss in iris informations. In this paper, in order to solve that problems, we removed eyelashes by gabor filter that using for analysis of frequency feature and improve preservation rate of iris informations. By novel method to extract various features on iris area using angle, frequency, and gaussian parameter on gabor filter that is one of the filters for analysing frequency feature for an image, we could remove accurately eyelashes with various lengths and shapes. As the result, proposed method represents that improve about 4% than previous methods using GMM or histogram analysis in iris preservation rate.

• Proceedings of the Korean Society for Agricultural Machinery Conference
• /
• 2002.02a
• /
• pp.464-470
• /
• 2002

본 연구에서는 초음파를 사과의 비파괴 품질판정에 이용하기 위한 기초 연구로서 사과 표면에 입사, 반사되는 초음파 신호를 계측 및 분석하기 위한 시스템을 구축하고 저장기간에 따른 사과의 초음파 특성을 계측하였으며, 그 결론은 다음과 같다. 1. 구성된 비 접촉식 초음파 장치를 이용하여 전압-시간 그래프에서 전압의 진폭(PTP)값을 구하였고 신호처리용 프로그램을 이용하여 획득한 신호를 주파수 영역에서 분석하여 피크주파수, 영역별 에너지 밀도함수 등을 구하였다. 2. 사과의 초음파 특성을 분석한 결과 사과의 저장기간은 초음파 응답신호의 진폭(PTP) 및 제 3영역 에너지 스펙트럼 밀도함수와 높은 상관성이 있는 것으로 나타났다. 3. 초음파를 농산물의 비파괴 품질판정에 적용시 문제가 되고 지는 센서와 시료와의 접촉응력에 의한 응답신호의 변형에 의한 문제를 해결하고자 비 접촉식 센서를 사용하였으며, 실제 농산물에 적용이 가능한 것으로 판단되었다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.10b
• /
• pp.592-594
• /
• 2003

이 논문에서는 wavelet과 sobel filter를 사용하여 영상의 객관적인 평가 점수를 계산하는 새로운 기법을 제안한다. 이 기법은 orthogonal wavelet 변환을 기초로 하고 있으며 원본 영상과 처리된 영상 데이터가 모두 가용하다는 것을 전제로 한다. Wavelet을 이용해 주파수에 따라 분할된 영상 정보를 이용해 각각의 부영역 별 차영상을 획득하고 이 획득된 영상의 에너지를 이용해 화질 평가 수치를 계산한다. 부영역 별로 획득된 영상은 일정한 크기의 블록으로 분할되어 동일한 블록 내에서 가용한 영상의 특징에 관한 정보(contrast, edge 영역의 분포 정도) 벡터와 내적하여 새로운 특징 벡터로 사용되고, 이 특징 벡터의 가중치를 최적화하여 높은 상관도의 화질평가 점수를 산출하게 된다.

• Journal of Digital Convergence
• /
• v.12 no.11
• /
• pp.309-316
• /
• 2014

In this paper, we propose an efficient detph map creation method using Graph Cut and Discrete Wavelet Transform. First, we have segmented the original image by using Graph Cut to process with its each areas. After that, the information which describes segmented areas of original image have been created by proposed labeling method for segmented areas. And then, we have created four subbands which contain the original image's frequency information. Finally, the depth map have been created by frequency map which made with HH, HL subbands and depth information calculation along the each segmented areas. The proposed method can perform efficient depth map creation process because of dynamic allocation using depth information. We also have tested the proposed method using PSNR(Peak Signal to Noise Ratio) method to evaluate ours.

• The Transactions of the Korea Information Processing Society
• /
• v.13 no.2
• /
• pp.41-47
• /
• 2024

When there were disparities in performance between models trained in the time and frequency domains, even after conducting an ensemble, we observed that the performance of the ensemble was compromised due to imbalances in the individual model performances. Therefore, this paper proposes a leakage detection technique to enhance the accuracy of pipeline leakage detection through a step-wise learning approach that extracts features from both the time and frequency domains and integrates them. This method involves a two-step learning process. In the Stage 1, independent model training is conducted in the time and frequency domains to effectively extract crucial features from the provided data in each domain. In Stage 2, the pre-trained models were utilized by removing their respective classifiers. Subsequently, the features from both domains were fused, and a new classifier was added for retraining. The proposed transfer learning-based feature fusion technique in this paper performs model training by integrating features extracted from the time and frequency domains. This integration exploits the complementary nature of features from both domains, allowing the model to leverage diverse information. As a result, it achieved a high accuracy of 99.88%, demonstrating outstanding performance in pipeline leakage detection.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.19 no.6
• /
• pp.21-27
• /
• 2019

The video compression standard required a processing technique for a high resoluion image and increased the coding size to increase the resolution of the image. Accurate motion estimation and increased coding size provide high accuracy and compression rate, but there is a problem of increased computational complexity. In this paper, we use DCT - based motion estimation in the frequency domain to reduce complexity. However, we found that the DCT and quantization process used in a general video encoder are applied to the frequency domain encoder, resulting in problems caused by the scaling process. Therfore, in this paper, we extract the scaling matrix that can be applied in the DCT step and resolve the, and improve the performance of motion estimation using increased coding size.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.24 no.9B
• /
• pp.1795-1805
• /
• 1999

In an OFDM (Orthogonal Frequency-Division Multiplexing) system, the sampling frequency offset between the transmitter and receiver is known to cause the interchannel interference (ICI), resulting in performance degradation. In this paper, we propose two time-domain techniques to estimate the sampling frequency offset, especially for a high data-rate OFDM system. The first technique estimates the sampling frequency offset by using the phase difference between two received samples with a fixed amount of time interval, corresponding to the transmitted training symbol, under the assumption of perfect symbol and carrier offset synchronization. The second technique estimates the sampling frequency offset and carrier frequency offset jointly, when the two offsets exist together, by using two training symbols with different frequency components and using a sample algebraic calculation. The proposed estimation techniques for sampling frequency offset cause no time delay due to all time-domain processing, and have a good performance due to no ICI effect. The performances of the proposed techniques are demonstrated by various simulations.

• The Journal of the Acoustical Society of Korea
• /
• v.38 no.5
• /
• pp.543-548
• /
• 2019

The tonal signal caused by the machinery component of a vessel such as an engine, gearbox, and support elements, can be modeled as a sparse signal in the frequency domain. Recently, compressive sensing based techniques that recover an original signal using a small number of measurements in a short period of time, have been applied for the tonal frequency detection. These techniques, however, cannot avoid a basis mismatch error caused by the discretization of the frequency domain. In this paper, we propose a method to detect the tonal frequency with a small number of measurements in the continuous domain by using the atomic norm minimization technique. From the simulation results, we demonstrate that the proposed technique outperforms conventional methods in terms of the exact recovery ratio and mean square error.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2015.11a
• /
• pp.161-164
• /
• 2015

본 논문에서는 고정된 카메라에서 초기 프레임을 참조하여 현재 프레임에 새롭게 유입된 물체의 실시간 분류 방법을 제안한다. 제안한 알고리즘의 실시간 분류 처리를 위하여 복잡도를 최소화 하였다. 먼저 전경과 배경을 구분하는 이진 분류 영상을 추출하기 위하여 DCT(Discrete Cosine Transform) 기법을 사용한다. DCT는 기존 공간영역에서 Texture를 분석하는 방식보다 더욱 정확하게 Texture를 분석할 수 있다. 이는 주파수 영역에서 Texture 특징 분석이 더욱 용이하고 각 요소 안에 intensity와 texture 정보를 종합적으로 고려할 수 있기 때문이다. 또한 DCT 계산 복잡도를 최소화하기 위하여 DCT 수행 전에 색 정보를 이용하여 미리 분류 영역을 분석함으로써 처리 효율을 극대화 하였다. 마지막으로 생성된 이진 분류 영상을 자연스럽게 matting하기 위하여 Guided 필터 사용을 제안한다. Guided 필터는 guidance 정보를 통해 입력 영상을 전반적으로 개선할 수 있지만 intensity가 평탄한 영역 등에서 그 한계를 보여주므로 본 논문에서는 Guided 필터의 단점을 개선하는 방법을 추가적으로 제안한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2004.11a
• /
• pp.871-874
• /
• 2004

멀티미디어 컨텐츠의 저작권 보호를 위한 워터마크 기법은 현재 많은 연구가 진행 중에 있다. 본 논문은 MPEG 비디오의 압축 과정에서 마지막 단계인 VLC 영역에 DCT된 워터마크의 키 값을 적용하여 얻어낸 신호를 run-level 쌍의 level 값의 부호에 따라 선택적으로 삽입하는 방범을 제안하였다. 양자화 된 DCT계수들의 고주파수 영역은 level 값을 거의 가지지 않기 때문에 고주파수영역의 계수 값이 제거되어도 워터마크의 정보는 대부분 살아남아 공격에 강인함을 보이며 압축의 마지막단계에 정보를 삽입하여 실시간 적용이 가능하다는 장점을 보여 주었다.