• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2021년도 춘계학술대회
• /
• pp.332-333
• /
• 2021

본 논문에서는 소규모 데이터 세트의 이미지 분류 작업에서 모델 과적 합 및 비 수렴을 해결하고 분류의 정확성을 높이는 데 주로 사용되는 몇 번의 학습을 기반으로 한 새로운 이미지 분류 방법을 제안한다.

• 복식문화연구
• /
• 제15권2호
• /
• pp.311-325
• /
• 2007

The purpose of this study is to investigate the effect of eyeshadow color(brown, purple), lipstick color(red, red purple, and yellow red), and lipstick tone(vivid, light, dull, and dark), clothing style(formal, casual), clothing tone(N9, N7, N4, N2) on image formation. Sets of stimulus and response scales(7 point semantic) were used as experimental materials. The stimuli were 128 color pictures manipulated with the combination of eyeshadow color, lipstick color, lipstick tone, clothing style, and clothing tone using computer simulation. The subjects were 768 female undergraduates living in Gyeongnam-do. Image factor of the stimulus was composed of 5 different components, attractiveness, stability, cuteness, visibility, and tenderness. In the 5 image components, clothing style and clothing tone showed independent effect. In the stability, cuteness and visibility, lipstick color showed independent effect. Eyeshadow color and lipstick tone influenced independently on the attractiveness, stability and visibility. In the coordination of achromatic clothing with makeup face, attractiveness image by the coordination of lipstick tone with clothing tone, cuteness image by the coordination of lipstick tone with clothing style or clothing style with clothing tone, visibility image can be produced by the coordination of eyeshadow color with lipstick color.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권12호
• /
• pp.6017-6037
• /
• 2017

In a multi-resolution image encoding system, the image is encoded into a single file as a layer of bit streams, and then it is transmitted layer by layer progressively to reduce the transmission time across a low bandwidth connection. This encoding scheme is also suitable for multiple decoders, each with different capabilities ranging from a handheld device to a PC. In our previous work, we proposed an edge adaptive hierarchical interpolation algorithm for multi-resolution image coding system. In this paper, we enhanced its compression efficiency by adding three major components. First, its prediction accuracy is improved using context adaptive error modeling as a feedback. Second, the conditional probability of prediction errors is sharpened by removing the sign redundancy among local prediction errors by applying sign flipping. Third, the conditional probability is sharpened further by reducing the number of distinct error symbols using error remapping function. Experimental results on benchmark data sets reveal that the enhanced algorithm achieves a better compression bit rate than our previous algorithm and other algorithms. It is shown that compression bit rate is much better for images that are rich in directional edges and textures. The enhanced algorithm also shows better rate-distortion performance and visual quality at the intermediate stages of progressive image transmission.

• 한국의상디자인학회지
• /
• 제11권1호
• /
• pp.111-121
• /
• 2009

The purpose of this study is to investigate the effect of body type(thin, standard, fat), necktie width(narrow, medium, wide), gender(man, woman) on image formation. Sets of stimulus and response scales(7 point semantic) were used as experimental materials. The stimuli were 9 color pictures manipulated with the combination of necktie-width and body type using computer simulation. The subjects were 116 male undergraduates and 99 female undergraduates living in Gyeongsangnam-do. Image factor of the stimulus was composed of 4 different components, young-activity, attractiveness-gracefulness, stability, and boldness. In the young-activity, attractiveness-gracefulness and stability, body type and necktie-width showed independent effect. In the boldness, body type showed independent effect. Significant interaction effects of body type and necktie-width on young-activity, attractiveness-gracefulness and stability were found. For young-activity image, necktie of narrow width were effective. For attractiveness-gracefulness image, body type of standard were effective. For stability image, body type of standard and body type of thin on necktie of narrow and medium width were effective. The male fashion image can be perceived differently according to combination of the clues used in body type and necktie-width. Therefore, when the male fashion to create images that will have to consider the body type and necktie width.

• 한국통신학회논문지
• /
• 제34권10B호
• /
• pp.1124-1132
• /
• 2009

본 논문은 집적영상의 요소영상을 초해상도 영상복원에 이용하여 집적영상의 해상도를 향상시키는 방법을 제안한다. 집적영상에서 전체 요소영상의 인접한 단일 요소영상들 사이에는 대상물체의 동일한 부분의 상을 포함하는 공통부분이 존재한다. 이러한 공통부분들을 초해상도 영상복원의 저해상도 영상으로 이용하게 되면 CCD(Charge Coupled Device) 등의 영상취득 장치의 제한된 해상도로 인한 집적영상의 낮은 해상도 문제를 보완 할 수 있게 된다. 전체 요소영상과 제안된 방법을 이용하여 해상도를 향상시킨 전체 요소영상을 비교하여 제안된 방법의 타당성을 증명하였다.

• 한국측량학회지
• /
• 제33권3호
• /
• pp.211-218
• /
• 2015

The analysis of remote sensing data depends on sensor specifications that provide accurate and consistent measurements. However, it is not easy to establish confidence and consistency in data that are analyzed by different sensors using various radiometric scales. For this reason, the cross-calibration method is used to calibrate remote sensing data with reference image data. In this study, we used an airborne hyperspectral image in order to calibrate a multispectral image. We presented an automatic cross-calibration method to calibrate a multispectral image using hyperspectral data and spectral mixture analysis. The spectral characteristics of the multispectral image were adjusted by linear regression analysis. Optimal endmember sets between two images were estimated by spectral mixture analysis for the linear regression analysis, and bands of hyperspectral image were aggregated based on the spectral response function of the two images. The results were evaluated by comparing the Root Mean Square Error (RMSE), the Spectral Angle Mapper (SAM), and average percentage differences. The results of this study showed that the proposed method corrected the spectral information in the multispectral data by using hyperspectral data, and its performance was similar to the manual cross-calibration. The proposed method demonstrated the possibility of automatic cross-calibration based on spectral mixture analysis.

• 패션비즈니스
• /
• 제12권1호
• /
• pp.30-44
• /
• 2008

The purpose of this study is to investigate the effect of eyeshadow color(brown, purple), lipstick color(red, red purple, and yellow red), and lipstick tone(vivid, light, dull, and dark), clothing tone(vivid, light, dull, and dark) on image formation. Sets of stimulus and response scales(7 point semantic) were used as experimental materials. The stimuli were 64 color pictures manipulated with the combination of eyeshadow color, lipstick color, lipstick tone, and clothing tone using computer simulation. The subjects were 384 female undergraduates living in Gyeongnam-do. Image factor of the stimulus was composed of 4 different components (attractiveness, visibility, gracefulness, and tenderness). In the 4 image components, eyeshadow color and clothing tone showed independent effect. Lipstick tone influenced independently on the visibility and tenderness. In the contrast color coordination of clothing and makeup, visibility image by the coordination of lipstick color with lipstick tone, lipstick color with clothing tone or lipstick tone with clothing tone, gracefulness image by the coordination of eyeshadow color with lipstick color, tenderness image can be produced by the coordination of eyeshadow color with lipstick color, eyeshadow color with lipstick tone or eyeshadow color with clothing tone.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권6호
• /
• pp.2527-2545
• /
• 2016

Measuring the similarity of given samples is a key problem of recognition, clustering, retrieval and related applications. A number of works, e.g. kernel method and metric learning, have been contributed to this problem. The challenge of similarity learning is to find a similarity robust to intra-class variance and simultaneously selective to inter-class characteristic. We observed that, the similarity measure can be improved if the data distribution and hidden semantic information are exploited in a more sophisticated way. In this paper, we propose a similarity learning approach for retrieval and recognition. The approach, termed as LDA-FEK, derives free energy kernel (FEK) from Latent Dirichlet Allocation (LDA). First, it trains LDA and constructs kernel using the parameters and variables of the trained model. Then, the unknown kernel parameters are learned by a discriminative learning approach. The main contributions of the proposed method are twofold: (1) the method is computationally efficient and scalable since the parameters in kernel are determined in a staged way; (2) the method exploits data distribution and semantic level hidden information by means of LDA. To evaluate the performance of LDA-FEK, we apply it for image retrieval over two data sets and for text categorization on four popular data sets. The results show the competitive performance of our method.

• Spatial Information Research
• /
• 제12권1호
• /
• pp.111-125
• /
• 2004

위성영상 분류에 관한 주요 주제 중 하나는 분류 정확도 향상에 있다. 동일지역에 대한 동일시기의 위성영상을 취득할 수 있는 기회가 많아지는 현실을 감안할 때, 복수의 위성영상 데이터를 이용하여 분류정확도가 향상된 분류결과를 도출하는 것은 의미 있는 일일 것이다. 본 연구 주제는 최대우도법을 사용하여 계산된 데이터의 사후확률 및 분류 불확실도를 Dempster-Shafer의 증거이론에 적용하여 분류정확도를 향상시키고자 하는 것이다. 분석결과 개별적인 데이터 분류나 데이터간 융합에 의한 분류보다 본 연구에서 제안한 방법이 전체정확도와 Kappa 지수 모두 높은 정확도를 나타냈으며, 정확도 차에 대한 검정을 실시하여 본 연구에서 제안한 방법이 다른 방법에 비해 우수함을 통계적으로 증명하였다.

• 융합보안논문지
• /
• 제6권2호
• /
• pp.53-60
• /
• 2006

CAD나 화상 처리와 같은 응용에서는 다각형으로 구성된 다양한 기하 객체를 다룬다. x축과 y축에 평행한 변을 가지는 다각형은 효율적인 조작을 위해 단순 다각형인 직사각형으로 흔히 분할된다. 그러나 동일한 기하 객체 또는 영역이라 할지라도 분할 방법에 따라 직사각형의 수, 크기 및 모양에 있어 전혀 다르게 된다. 따라서 CAD 또는 화상과 같은 장면으로부터 추출된 두 개의 직사각형 집합이 동일한 영역을 나타내는지 판단하는 알고리즘이 요구된다. 본 고에서는 직사각형들로 구성된 두 집합을 효율적으로 비교하는 알고리즘을 제안한다. 제안된 알고리즘은 기존의 훑기에 근거한 알고리즘에 비해 간단할 뿐만 아니라, 균형이진트리를 이용한 알고리즘에서의 중첩 직사각형 수 $O(n^2)$을 O(nlogn)으로 줄인다.