• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권5호
• /
• pp.2122-2140
• /
• 2020

Proper and precise embedding of commercial ads within Webpages requires Ad-hoc analysis and understanding of their content. By the successful implementation of this step, both publishers and advertisers gain mutual benefits through increasing their revenues on the one hand, and improving user experience on the other. In this research work, we propose a novel multi-level context-based ads serving approach through which ads will be served at generic publisher websites based on their contextual relevance. In the proposed approach, knowledge encoded in domain-specific and generic semantic repositories is exploited in order to analyze and segment Webpages into sets of contextually-relevant segments. Semantically-enhanced indexes are also constructed to index ads based on their textual descriptions provided by advertisers. A modified cosine similarity matching algorithm is employed to embed each ad from the Ads repository into one or more contextually-relevant segments. In order to validate our proposal, we have implemented a prototype of an ad serving system with two datasets that consist of (11429 ads and 93 documents) and (11000 documents and 15 ads), respectively. To demonstrate the effectiveness of the proposed techniques, we experimentally tested the proposed method and compared the produced results against five baseline metrics that can be used in the context of ad serving systems. In addition, we compared the results produced by our system with other state-of-the-art models. Findings demonstrate that the accuracy of conventional ad matching techniques has improved by exploiting the proposed semantically-enhanced context-based ad serving model.

• 한국컴퓨터정보학회논문지
• /
• 제15권11호
• /
• pp.67-73
• /
• 2010

본 논문에서는 두 점의 집합들 사이의 기하학적 유사도에 근거한 Hausdorff 거리와 국지적 미세 텍스처의 분포에 근거한 Local Binary Pattern 거리가 융합된 새로운 측도를 도입함으로써 얼굴검출의 정확도를 높이는 방법을 제안하고 있다. 트레이닝 데이터를 이용해서 두 가지의 상이한 측도들을 정규화할 수 있는 매개변수와 최적화된 융합 비율을 찾는 방법을 보였다. 흔히 사용되는 얼굴 데이터베이스에 적용함으로써 제시된 방법이 두 가지 방법 각각을 이용한 방법보다 효과적이고 얼굴의 자세, 조명, 배경의 변화에 강인함을 보였다. 실험에서 사용된 데이터베이스의 경우 실제 얼굴의 위치와 검출된 얼굴의 위치 간의 평균거리오차가 LBP 방식의 47.9%, Hausdorff 방식의 22.8% 로 감소됨을 보였다.

• Journal of the Optical Society of Korea
• /
• 제20권4호
• /
• pp.515-523
• /
• 2016

Spectral reflectance is sparse in space, and while the traditional spectral-reconstruction algorithm does not make full use of this characteristic sparseness, the compressive sensing algorithm can make full use of it. In this paper, on the basis of analyzing compressive sensing based on the orthogonal matching pursuit algorithm, a new algorithm based on the Dice matching criterion is proposed. The Dice similarity coefficient is introduced, to calculate the correlation coefficient of the atoms and the residual error, and is used to select the atoms from a library. The accuracy of Spectral reconstruction based on the pseudo-inverse method, Wiener estimation method, OMP algorithm, and DOMP algorithm is compared by simulation on the MATLAB platform and experimental testing. The result is that spectral-reconstruction accuracy based on the DOMP algorithm is higher than for the other three methods. The root-mean-square error and color difference decreases with an increasing number of principal components. The reconstruction error decreases as the number of iterations increases. Spectral reconstruction based on the DOMP algorithm can improve the accuracy of color-information replication effectively, and high-accuracy color-information reproduction can be realized.

• Investigative Magnetic Resonance Imaging
• /
• 제24권3호
• /
• pp.123-131
• /
• 2020

Purpose: Joint label fusion (JLF) is a popular multi-atlas-based segmentation algorithm, which compensates for dependent errors that may exist between atlases. However, in order to get good segmentation results, it is very important to set the several free parameters of the algorithm to optimal values. In this study, we first investigate the feasibility of a JLF algorithm for prostate segmentation in MR images, and then suggest the optimal set of parameters for the automatic prostate segmentation by validating the results of each parameter combination. Materials and Methods: We acquired T2-weighted prostate MR images from 20 normal heathy volunteers and did a series of cross validations for every set of parameters of JLF. In each case, the atlases were rigidly registered for the target image. Then, we calculated their voting weights for label fusion from each combination of JLF's parameters (r_pxy, r_pz, r_sxy, r_sz, β). We evaluated the segmentation performances by five validation metrics of the Prostate MR Image Segmentation challenge. Results: As the number of voxels participating in the voting weight calculation and the number of referenced atlases is increased, the overall segmentation performance is gradually improved. The JLF algorithm showed the best results for dice similarity coefficient, 0.8495 ± 0.0392; relative volume difference, 15.2353 ± 17.2350; absolute relative volume difference, 18.8710 ± 13.1546; 95% Hausdorff distance, 7.2366 ± 1.8502; and average boundary distance, 2.2107 ± 0.4972; in parameters of r_pxy = 10, r_pz = 1, r_sxy = 3, r_sz = 1, and β = 3. Conclusion: The evaluated results showed the feasibility of the JLF algorithm for automatic segmentation of prostate MRI. This empirical analysis of segmentation results by label fusion allows for the appropriate setting of parameters.

• 정보처리학회논문지B
• /
• 제12B권5호
• /
• pp.577-586
• /
• 2005

영상 처리와 패턴 인식 그리고 컴퓨터 비젼 분야의 가장 성공적인 응용들 중 하나인 얼굴 인식을 위해 많은 알고리즘이 제안되었고, 최근에는 얼굴의 어떤 속성이 대상을 인식하는 것을 더 쉽거나 어렵게 만드는지에 대한 연구가 진행되고 있다. 본 논문에서는 얼굴의 속성(조명, 표정)에 따라 각각의 알고리즘의 인식 성능이 달라지는 점에 착안해서, 얼굴 데이터와 적용된 알고리즘과의 연관성을 이용하여 인식 성능을 높이는 클러스터링 방법을 제안하였다. 실험에서는 인식 알고리즘으로 n-tuple, PCA 그리고 가보 웨이블릿이 사용되었고, 세 가지 벡터화 방법이 제안되었다. 우선 학습 데이터를 k-means 알고리즘을 이용하여 클러스터링하고 각각의 클러스터에 대한 세 가지 인식 알고리즘의 적합도를 평가한 후, 같은 알고리즘을 선택한 클러스터들을 통합하여 새로운 클러스터를 구성한다. 그리고 테스트 데이터에서 새로운 클러스터에 대한 유사도를 평가하여 가장 가까운 클러스터가 선택한 알고리즘으로 인식을 수행한다. 그 결과 클러스터링 과정을 거치지 않고 단일 알고리즘을 사용하여 인식했을 때보다 인식 성능이 향상된 것을 관찰할 수 있다.

• 한국표면공학회:학술대회논문집
• /
• 한국표면공학회 2017년도 춘계학술대회 논문집
• /
• pp.159-159
• /
• 2017

Ti-6Al-4V alloy have been used for dental implant because of its excellent biocompatibility, corrosion resistance, and mechanical properties. However, the integration of such implant in bone was not in good condition to achieve improved osseointergraiton. For solving this problem, calcium phosphate (CaP) has been applied as coating materials on Ti alloy implants for hard tissue applications because its chemical similarity to the inorganic component of human bone, capability of conducting bone formation and strong affinity to the surrounding bone tissue. Various metallic elements are known to play an important role in the bone formation and also affect bone mineral characteristics. Especially, Zn is essential for the growth of the human and Zn coating has a major impact on the improvement of corrosion resistance. Plasma electrolytic oxidation (PEO) is a promising technology to produce porous and firmly adherent inorganic Zn containing TiO2(Zn-TiO2)coatings on Ti surface, and the a mount of Zn introduced in to the coatings can be optimized by altering the electrolyte composition. In this study, effect of Zn content on the corrosion behavior of Ti-6Al-4V alloy after plasma electrolytic oxidation were studied by SEM, EDS, XRD, AC impedance, and potentiodynamic polarization test. The potentiodynamic polarization and AC impedance tests for corrosion behaviors were carried out in 0.9% NaCl solution at similar body temperature using a potentiostat with a scan rate of 1.67 mV/s and potential range from -1500 mV to +2000 mV. Also, AC impedance was performed at frequencies ranging from 10 MHz to 100 kHz for corrosion resistance.

• 한국지능시스템학회논문지
• /
• 제19권3호
• /
• pp.297-303
• /
• 2009

본 논문에서는 비선형 변환에 의해 입력신호를 고차원의 확장공간으로 변환한 후, 주성분분석기법(PCA)에 의해 신호의 특징을 추출하는 기법을 제안한다. 특징추출을 위해 사용되는 기존의 주성분분석기법은 입력데이터가 비선형 특성을 갖는 경우 최적의 변환행렬을 구할 수 없다는 문제점을 가지고 있다. 이러한 문제점을 해결하기 위해, 확장공간상에서 구간별로 입력데이터를 분할한 후 주성분분석기법에 의해 구간별 특징을 추출하는 서브패턴기반 주성분분석기법(SpPCA)을 적용하였다. 다음 단계인 분류단계에서는 MLP 비선형분류기를 이용하여 구간마다 추출된 특징벡터를 이용하여 기준패턴과의 유사도를 산출한다. 최종 분류단계에서는 MLP에 의해서 산출된 유사도에 기반을 둔 융합법칙에 의하여 생체 스펙트럼 패턴을 분류한다. 제안된 방법의 유용성을 보이기 위한 실험결과에서 기존의 방법들에 비해서 향상된 인식결과를 보임을 확인하였다.

• 한국작물학회지
• /
• 제59권3호
• /
• pp.359-363
• /
• 2014

Soybean seed is a good source of plant protein in human consumables such as baby formula and protein concentrate. The seeds contain an abundance of storage proteins, namely ${\beta}$-conglycin and glycinin that account for ~ 70-80% of the total seed protein content. Proteome profiling has been proved to be an efficient way that can help us to investigate the seed storage proteins. In the present study, the seeds were removed from the pods and the cotylendonary tissues were separated from the testa for proteome analysis in order to investigate the seed storage proteins. A systematic proteome profiling was conducted through one-dimensional gel electrophoresis followed by MALDI-TOF-TOF mass spectrometry in the seeds (cotyledonary tissue) of soybean genotypes. Two dimensional gels stained with CBB, a total of 10 proteins were identified and analyzed using MASCOT search engine according to the similarity of sequences with previously characterized proteins along with the UniProt database. A total of ten proteins such as glycinin Gy4 precursor, glycinin G3 precursor, glycinin G1 precursor, glycinin chain A2B1a precursor, glycinin chain A2B1a precursor were identified in our investigation. However, the glycinin subunit may be considered to play important roles in soybean breeding and biochemical characterization. In addition, the improved technique will be useful to dissect the genetic control of glycinin expression in soybean.

• 대한원격탐사학회지
• /
• 제31권5호
• /
• pp.433-440
• /
• 2015

본 연구에서는 초분광영상을 이용한 표적탐지에 있어 배경 신호 특징에 포함되는 표적 신호가 탐지성능에 미치는 영향을 살펴보고, 분광각을 기준으로 표적과 유사한 분광반사 특성을 가지는 화소들을 배경 특징화 과정에서 제외함으로써 표적탐지 성능을 향상시킬 수 있는 방법을 제안하였다. 초분광 표적탐지를 위해 가장 흔히 이용되는 matched Filter와 adaptive cosine estimator 기법에 대해 실제 항공 초분광영상 자료와 여기에 인공표적을 삽입하여 생성한 모의 자료를 이용한 실험 결과, 배경 특징화를 위한 공분산행렬 계산 시 표적 스펙트럼과 유사도가 높은 표적 유사화소들을 제외함으로써 탐지 성능이 크게 향상될 수 있음이 확인되었다. 분광각외에 다양한 유사도 판정 기준들에 대한 적용성 연구와 함께, 제외되는 표적 유사화소들의 양이 최적으로 결정될 수 있는 방법에 대한 추가 연구가 이루어진다면 사용이 간편하고 성능이 우수한 초분광 표적탐지 기법으로 활용될 수 있을 것으로 기대된다.

• 한국물환경학회지
• /
• 제26권1호
• /
• pp.10-18
• /
• 2010

Water quality and the distribution of ammonia oxidizing bacteria were characterized in constructed wetland of Shihwa lake. Both physico-chemical parameters and fecal indicator microorganisms including total coliforms, E.coli, Enterococcus spp. were measured. In addition, denaturant gradient gel electrophoresis (DGGE) was carried out after PCR amplification of amoA gene from input, output, and wetland sites of the Banwol, Donghwa, and Samhwa stream in Shihwa lake area. Physico-chemical parameters were in proper range for typical nitrifying bacteria to grow and perform their biological activities. Average concentrations of fecal indicator microorganisms of wetland samples were lower than those of input sites. These results suggested that microbial water quality improved by the process of constructed wetland. According to phylogenetic information obtained from DGGE from study sites, distribution of nitrifying bacteria from each of input, output, and wetland were generally distinctive one another. In addition, distribution of nitrifying bacteria between Banwol and Donghwa streams showed higher similarity (52.6%) than this of Samhwa stream (15.2%). These results indicated that characteristics of ammonia oxidizing bacteria in Samhwa were unique in comparison with those of Banwol and Donghwa stream.