• IEIE Transactions on Smart Processing and Computing
• /
• 제2권3호
• /
• pp.140-147
• /
• 2013

We present a vaguelette-curvelet decomposition based image deblurring algorithm. We first perform denoising based on the hard-thresholding rule by estimating unknown curvelet coefficients. The proposed algorithm then calculates vaguelette functions by deconvolving the curvelet bases by the point spread function. Vaguelette transform is finally performed to make a clearly restored image. Since the proposed algorithm uses the curvelet transform to sensitively express the edges in all directions, it is possible to restore images with more naturally preserved edges in all directions.

• 한국컴퓨터정보학회논문지
• /
• 제23권4호
• /
• pp.51-56
• /
• 2018

This paper proposes the efficient preprocessing method based on curvelet transform for edge enhancement in image. The propose method is generated the edge map by using the Canny algorithm to wavelet transform, which is the sub-step of the curvelet transform. In order to improve the part of edge feature, the selective sharpening according to the generate edge map is applied. In experimental result, the propose method achieves that the enhancement of edge feature is better than conventional methods. This leads that peak to signal noise ratio, edge intensity are improvement on average about 1.92, 1.12dB respectively.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제7권11호
• /
• pp.2702-2719
• /
• 2013

A new optimal scheme based on curvelet transform is proposed for retinal image enhancement (RIE) using real-coded quantum genetic algorithm. Curvelet transform has better performance in representing edges than classical wavelet transform for its anisotropy and directional decomposition capabilities. For more precise reconstruction and better visualization, curvelet coefficients in corresponding subbands are modified by using a nonlinear enhancement mapping function. An automatic method is presented for selecting optimal parameter settings of the nonlinear mapping function via quantum genetic search strategy. The performance measures used in this paper provide some quantitative comparison among different RIE methods. The proposed method is tested on the DRIVE and STARE retinal databases and compared with some popular image enhancement methods. The experimental results demonstrate that proposed method can provide superior enhanced retinal image in terms of several image quantitative evaluation indexes.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권4호
• /
• pp.2078-2093
• /
• 2019

Visual smoke recognition is a challenging task due to large variations in shape, texture and color of smoke. To improve performance, we propose a novel smoke recognition method by combining dual-encoded features that are extracted from both spatial and Curvelet domains. A Curvelet transform is used to filter an image to generate fifty sub-images of Curvelet coefficients. Then we extract Local Binary Pattern (LBP) maps from these coefficient maps and aggregate histograms of these LBP maps to produce a histogram map. Afterwards, we encode the histogram map again to generate Dual-encoded Local Binary Patterns (Dual-LBP). Histograms of Dual-LBPs from Curvelet domain and Completed Local Binary Patterns (CLBP) from spatial domain are concatenated to form the feature for smoke recognition. Finally, we adopt Gaussian Kernel Optimization (GKO) algorithm to search the optimal kernel parameters of Support Vector Machine (SVM) for further improvement of classification accuracy. Experimental results demonstrate that our method can extract effective and reasonable features of smoke images, and achieve good classification accuracy.

• Structural Engineering and Mechanics
• /
• 제60권4호
• /
• pp.667-691
• /
• 2016

This study focuses on the damage detection of defect types in plate structures based on wavelet transform (WT) and curvelet transform (CT). In particular, for damage detection of structures these transforms have been developed since the last few years. In recent years, the CT approach has been also introduced in an attempt to overcome inherent limitations of traditional multi-scale representations such as wavelets. In this study, the performance of CT is compared with WT in order to demonstrate the capability of WT and CT in detection of defect types in plate structures. To achieve this purpose, the damage detection of defect types through defect shape in rectangular plate is investigated. By using the first mode shape of plate structure and the distribution of the coefficients of the transforms, the damage existence, the defect location and the approximate shape of defect are detected. Moreover, the accuracy and performance generality of the transforms are verified through using experimental modal data of a plate.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2011년도 춘계학술발표대회
• /
• pp.479-482
• /
• 2011

얼굴인식의 인식률 향상과 계산량을 줄이기 위한 방법으로 Curvelet 변환과 (2D)$^2$PCA(Two directional two-dimensional PCA) 를 통한 특징추출 및 차원축소 방법을 제안한다. 기존의 Wavelet 변환과 PCA 를 통한 기법들이 소개되어 인식률 향상을 이끌어 냈다. 그런데 Curvelet Transform 은 곡선의 정보를 효과적으로 표현할 수 있는 장점이 있고, (2D)$^2$PCA 는 PCA 에 비해 계산량이 적은 장점이 있기 때문에 이를 이용하여 인식률과 처리성능 측면에서 개선된 결과를 얻고자 한다.

• International Journal of Computer Science & Network Security
• /
• 제21권9호
• /
• pp.303-316
• /
• 2021

Digital content protection has recently become an important requirement in biometrics-based authentication systems due to the challenges involved in designing a feasible and effective user authentication method. Biometric approaches are more effective than traditional methods, and simultaneously, they cannot be considered entirely reliable. This study develops a reliable and trustworthy method for verifying that the owner of the biometric traits is the actual user and not an impostor. Watermarking-based approaches are developed using a combination of a color face image of the user and a mobile equipment identifier (MEID). Employing watermark techniques that cannot be easily removed or destroyed, a blind image watermarking scheme based on fast discrete curvelet transform (FDCuT) and discrete cosine transform (DCT) is proposed. FDCuT is applied to the color face image to obtain various frequency coefficients of the image curvelet decomposition, and for high frequency curvelet coefficients DCT is applied to obtain various frequency coefficients. Furthermore, mid-band frequency coefficients are modified using two uncorrelated noise sequences with the MEID watermark bits to obtain a watermarked image. An analysis is carried out to verify the performance of the proposed schema using conventional performance metrics. Compared with an existing approach, the proposed approach is better able to protect multimedia data from unauthorized access and will effectively prevent anyone other than the actual user from using the identity or images.

• 한국인터넷방송통신학회논문지
• /
• 제9권4호
• /
• pp.59-68
• /
• 2009

본 논문은 최근 소개된 curvelet 변환 구성을 사용하여 하잇브리드 다초점 이미지 융합 기법을 다룬다. 하잇브리화는 MS 융합 규칙을 새로운 "복제" 방법과 결합시킴으로써 얻어진다. 제안된 기법은 MS 규칙을 사용하여 각 분해 레벨 이미지의 스펙트럼내에 m개의 가장 두드러진 항들만을 융합시킨다. 이 기법은 이미지의 어떠한 스케일과 방향, 이동에서 변환 집합의 MSC에 충실하여 m-항 융합으로 합성이 이루어진다. 제안한 방법을 평가하기 위하여 Xydeas 와 Petrovic이 제안한 경계선에 민감한 객관적 품질 척도를 적용하였다. 실험 결과는 제안한 기법이 잉여, 쉬프트-불변 Dual-Tree 복소수 웨이블릿 변환에 대한 대안으로서의 가능성을 보여주었다. 특히, 50%의 m-항 융합은 어떤 시각적인 품질 저하를 갖지 않는 결과를 주는 것이 확인되었다.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2004년도 Proceedings of ISRS 2004
• /
• pp.224-227
• /
• 2004

From 1980's, the DWT(Discrete Wavelet Transform) is applied to the data/image processing. Many people use the DWT in remote sensing for diversity purposes and they are satisfied with the wavelet theory. Though the algorithm for wavelet is very diverse, many people use the standard wavelet such as Daubechies D4 wavelet and biorthogonal 9/7 wavelet. We will overview the wavelet theory for discrete form which can be applied to the image processing. First, we will introduce the basic DWT algorithm and review the wavelet algorithm: EZW (Embedded Zerotree Wavelet), SPIHT(Set Partitioning in Hierarchical Trees), Lifting scheme, Curvelet, etc. Finally, we will suggest the properties of wavelet algorithm; and wavelet filter for each image processing in remote sensing.

• 지구물리와물리탐사
• /
• 제26권2호
• /
• pp.37-51
• /
• 2023

그라운드-롤(ground roll)은 육상 탄성파 탐사 자료에서 가장 흔하게 나타나는 일관성 잡음(coherent noise)이며 탐사를 통해 얻고자 하는 반사 이벤트 신호보다 훨씬 큰 진폭을 가지고 있다. 따라서 탄성파 자료 처리에서 그라운드-롤 제거는 매우 중요하고 필수적인 과정이다. 그라운드-롤 제거를 위해 주파수-파수 필터링, 커브릿(curvelet) 변환 등 여러 제거 기술이 개발되어 왔으나 제거 성능과 효율성을 개선하기 위한 방법에 대한 수요는 여전히 존재한다. 최근에는 영상처리 분야에서 개발된 딥러닝 기법들을 활용하여 탄성파 자료의 그라운드-롤을 제거하고자 하는 연구도 다양하게 수행되고 있다. 이 논문에서는 그라운드-롤 제거를 위해 CNN (convolutional neural network) 또는 cGAN (conditional generative adversarial network)을 기반으로 하는 세가지 모델(DnCNN (De-noiseCNN), pix2pix, CycleGAN)을 적용한 연구들을 소개하고 수치 예제를 통해 상세히 설명하였다. 알고리듬 비교를 위해 동일한 현장에서 취득한 송신원 모음을 훈련 자료와 테스트 자료로 나누어 모델을 학습하고, 모델 성능을 평가하였다. 이러한 딥러닝 모델은 현장자료를 사용하여 훈련할 때, 그라운드-롤이 제거된 자료가 필요하므로 주파수-파수 필터링으로 그라운드-롤을 제거하여 정답자료로 사용하였다. 딥러닝 모델의 성능 평가 및 훈련 결과 비교는 정답 자료와의 유사성을 기본으로 상관계수와 SSIM (structural similarity index measure)과 같은 정량적 지표를 활용하였다. 결과적으로 DnCNN 모델이 가장 좋은 성능을 보였으며, 다른 모델들도 그라운드-롤 제거에 활용될 수 있음을 확인하였다.