• 한국철도학회논문집
• /
• 제13권3호
• /
• pp.283-289
• /
• 2010

디지털 카메라의 발전으로 인해 점차 영상을 사용한 철도의 안전관리기법이 그 사용범위를 넓히고 있다. 그러나 선로의 특성상 많은 저조도 환경에서의 영상 취득 과정에서는 심한 잡음이 영상의 화질을 떨어뜨릴 뿐만 아니라 추가적인 영상처리의 오류를 발생시킨다. 최근의 3D 잡음제거 방식은 시간적으로 연속된 영상간의 픽셀을 참조함으로 2D 잡음제거보다 더 나은 잡음 제거 결과를 얻을 수 있으나 움직임 부분에서는 오히려 모션 블러와 같은 열화가 나타나게 된다. 본 논문에서는 저조도 영상에서 적응적 가중평균필터를 이용하여 보다 정확한 움직임 검출을 구현하며, 3D 잡음제거 방식에 2D잡음 제거 방식의 결과를 적응적으로 사용하여 객관적 화질과 주관적 화질을 개선하였다.

• 한국산업정보학회논문지
• /
• 제12권3호
• /
• pp.132-140
• /
• 2007

카메라 영상은 명도의 변화와 부정확한 초점으로 인해 스캐너 영상에 비하여 화질이 저하된다. 본 연구에서는 카메라 영상에서 자주 발생하는 화질 저하에 대한 적응력을 강화하여 카메라기반 문서 인식에 적합한 이진화 방법을 제안한다. 기존의 평가에서 우수하다고 보고된 이진화 방법을 기반으로 하되, 낮은 조도와 부정확한 초점으로 인해 명도 대비가 낮은 영상에 대한 적응력을 강화하였다. 또한 이진화 시 국소 윈도우를 이용하여 기존의 방법에서 뭉개지기 쉬운 문자의 세부 구조를 섬세하게 추출하도록 개선하였다. 실험에서는 기존에 우수하다고 평가된 이진화 방법들과 제안하는 방법을 문서 인식에 적용하여 다양한 카메라 문서 영상에 대한 성능을 비교하였는데, 그 결과 제안하는 방법이 카메라로 입력받은 문서 영상의 인식에 효과적임을 확인하였다.

• 한국의류학회지
• /
• 제38권6호
• /
• pp.783-795
• /
• 2014

This study investigates how the physical attractiveness of salespeople and other customers and self-image congruence influence customer perception and brand attitude. As a result of a pretest, four types of pictorial stimuli were developed: physical attractiveness of salespeople (high/low) ${\times}$ that of other customers (high/low). Stimuli were developed using Photoshop CS. A total of 550 samples in two experiments were used and all respondents were women in their 20s and 30s. Data were analyzed using ANOVA and SEM. The result of analysis shows that the physical attractiveness of salesperson had a significant impact on perceived quality, but that of other customers did not. However, self-image congruence with other customers shows a significant effect on perceived quality; however, salespeople did not. The impact of the perceived product quality on brand attitude is higher than perceived service quality. This study adds to fashion retailing literature by demonstrating that physical attractiveness and self-image congruence can influence a customers' perception of product or service quality and brand attitude. It provides interesting insight into how retailers can use social factors as a strategic tool in a retail setting.

• 전자공학회논문지
• /
• 제51권3호
• /
• pp.105-111
• /
• 2014

영상 해상도 개선은 저해상도 획득 영상의 해상도를 개선하여 고해상도 영상을 생성하는 기술이다. 영상 해상도 개선을 위해서는 저해상도 획득 영상의 열화 과정에서 발생하는 손실된 화소 정보를 정확하게 추정하는 것이 중요하다. 따라서 본 논문에서는 영상 해상도 개선을 위한 다중 부족분 추정 방법을 제안한다. 제안하는 방법은 획득 영상의 부영상 집합에 알려진 열화 및 복원 과정을 수행하여 서로 다른 형태의 다중 부족분을 추정하고, 추정된 부족분과 획득 영상의 보간 영상의 결합을 통해서 결과 영상을 생성하고, 디블러링을 수행하여 최종 복원 영상을 생성한다. 객관적 화질 측정 지표인 PSNR, SSIM, FSIM으로 비교한 결과 제안한 방법이 보간만을 사용하는 방법들보다 높은 값을 가지는 것을 확인하였다. 또한 결과 영상의 시각적 비교 결과 주관적 관점의 화질도 가장 뛰어난 것을 알 수 있었고, 보간만을 사용하는 방법들보다 빠른 계산시간을 가지는 것을 확인할 수 있었다. 제안하는 방법은 영상 해상도 개선을 위한 응용 환경에서 유용하게 사용될 수 있다.

• 천문학회보
• /
• 제44권2호
• /
• pp.70.4-70.4
• /
• 2019

Image super-resolution (SR) is a technique that enhances the resolution of a low resolution image. In this study, we use three SR models (RCAN, ProSRGAN and Bicubic) for enhancing solar SDO/HMI magnetograms using deep learning. Each model generates a high resolution HMI image from a low resolution HMI image (4 by 4 binning). The pixel resolution of HMI is about 0.504 arcsec. Deep learning networks try to find the hidden equation between low resolution image and high resolution image from given input and the corresponding output image. In this study, we trained three models with HMI images in 2014 and test them with HMI images in 2015. We find that the RCAN model achieves higher quality results than the other two methods in view of both visual aspects and metrics: 31.40 peak signal-to-noise ratio(PSNR), Correlation Coefficient (0.96), Root mean square error (RMSE) is 0.004. This result is also much better than the conventional bi-cubic interpolation. We apply this model to a full-resolution SDO/HMI image and compare the generated image with the corresponding Hinode NFI magnetogram. As a result, we get a very high correlation (0.92) between the generated SR magnetogram and the Hinode one.

• 한국해양공학회지
• /
• 제36권1호
• /
• pp.32-40
• /
• 2022

Underwater optical images face various limitations that degrade the image quality compared with optical images taken in our atmosphere. Attenuation according to the wavelength of light and reflection by very small floating objects cause low contrast, blurry clarity, and color degradation in underwater images. We constructed an image data of the Korean sea and enhanced it by learning the characteristics of underwater images using the deep learning techniques of CycleGAN (cycle-consistent adversarial network), UGAN (underwater GAN), FUnIE-GAN (fast underwater image enhancement GAN). In addition, the underwater optical image was enhanced using the image processing technique of Image Fusion. For a quantitative performance comparison, UIQM (underwater image quality measure), which evaluates the performance of the enhancement in terms of colorfulness, sharpness, and contrast, and UCIQE (underwater color image quality evaluation), which evaluates the performance in terms of chroma, luminance, and saturation were calculated. For 100 underwater images taken in Korean seas, the average UIQMs of CycleGAN, UGAN, and FUnIE-GAN were 3.91, 3.42, and 2.66, respectively, and the average UCIQEs were measured to be 29.9, 26.77, and 22.88, respectively. The average UIQM and UCIQE of Image Fusion were 3.63 and 23.59, respectively. CycleGAN and UGAN qualitatively and quantitatively improved the image quality in various underwater environments, and FUnIE-GAN had performance differences depending on the underwater environment. Image Fusion showed good performance in terms of color correction and sharpness enhancement. It is expected that this method can be used for monitoring underwater works and the autonomous operation of unmanned vehicles by improving the visibility of underwater situations more accurately.

• 여성건강간호학회지
• /
• 제13권4호
• /
• pp.290-298
• /
• 2007

Purpose: This study was to measure the quality of life(QOL) and to identify the related factors in gynecological cancer patients during chemotherapy. Method: The subjects of this study were the patients who had undergone a hysterectomy and were admitted for chemotherapy at S university hospital between November 2006 and April 2007. Data was collected from 106 gynecological cancer patients with the use of a structured questionnaire which measured the QOL(FACT-G), body image, the presence of anxiety & depression, uncertainty, and family support. The data was analyzed by the SPSS win 12.0 program. Results: The mean FACT-total score was 62.1$({\pm}16.7)$ (range; 26-107). Positive correlations were found between QOL and body image(r= .67, p= .00), and QOL and family support(r= .32, p= .00), whereas there were negative correlations between QOL and anxiety(r= -.54, p= .00), QOL and depression(r= -.70, p= .00), and QOL and uncertainty(r= -.59, p= .00). Fifty seven pre cent of the variance in subjective overall QOL can be explained by depression, body image, and uncertainty(Adj $R^2$= .57, F=47.00, p= .00). Conclusion: Our patients had a relatively low QOL score. Factors significantly affecting quality of life were depression, body image and uncertainty. Nursing interventions, therefore, should be focused on improving QOL in gynecological cancer patients during chemotherapy, particularly so in patients with depression, uncertainty or poor body image.

• The International Journal of Costume Culture
• /
• 제5권1호
• /
• pp.58-66
• /
• 2002

The objectives of this study are to develope sensibility image scales for Korean traditional motifs by quantitatively measuring their images and preference and to classify them into clusters. Data were collected via a questionnaire from seven hundred twenty five Korean undergraduate students. Re experimental materials were forty eight stimuli of Korean traditional motifs with different categories, interpretation types, composition types, and application objects. The instruments consisted of 7-point polar semantic differential scales of twenty three bipolar adjectives including preference. Data were analyzed by correspondence analysis, cluster analysis, ANOVA and Duncan's multiple range test. Re major results are as follows; image scales for textile patterns and dress designs using Korean traditional motifs were constructed. The axes of sensibility image scales for both textile patterns and dress designs were defined by quality level and degree of simplicity. Second, four clusters on the scale of textile patterns and two clusters on the scale dress designs were identified. Third, in the case of textile Patterns, the preferred cluster had high-quality and classical images, while the cluster that was not preferred had a complex image. In the case of dress designs, the preferred cluster had simple and high-quality images, while the cluster that was not preferred had complex and low-quality images.

• Journal of Information Processing Systems
• /
• 제19권4호
• /
• pp.417-426
• /
• 2023

To improve the effect of image restoration and solve the image detail loss, an image dehazing enhancement algorithm based on mean guided filtering is proposed. The superpixel calculation method is used to pre-segment the original foggy image to obtain different sub-regions. The Ncut algorithm is used to segment the original image, and it outputs the segmented image until there is no more region merging in the image. By means of the mean-guided filtering method, the minimum value is selected as the value of the current pixel point in the local small block of the dark image, and the dark primary color image is obtained, and its transmittance is calculated to obtain the image edge detection result. According to the prior law of dark channel, a classic image dehazing enhancement model is established, and the model is combined with a median filter with low computational complexity to denoise the image in real time and maintain the jump of the mutation area to achieve image dehazing enhancement. The experimental results show that the image dehazing and enhancement effect of the proposed algorithm has obvious advantages, can retain a large amount of image detail information, and the values of information entropy, peak signal-to-noise ratio, and structural similarity are high. The research innovatively combines a variety of methods to achieve image dehazing and improve the quality effect. Through segmentation, filtering, denoising and other operations, the image quality is effectively improved, which provides an important reference for the improvement of image processing technology.

• 한국컴퓨터정보학회논문지
• /
• 제11권2호
• /
• pp.53-63
• /
• 2006

본문에서는 Discrete Wavelet Transform(DWT)를 기반으로 한 워터마킹 알고리즘을 제안하였다. 제안한 방법은 원 영상을 네 개의 영상으로 분할하고, 각 분할된 영상들을 2단계 DWT한다. 네 개의 DWT된 영상 중 하나의 영상에 에지 테이블(Edge Table) 이라는 새로운 개념으로 영상의 고유한 에지 특성을 고려해 워터마크를 삽입하였다. 워터마크 추출 시에는 워터마크가 삽입된 한 개의 분할 영상과 나머지 분할 영상들을 비교하여 원 영상 없이 워터마크를 추출하였다. 기존의 Blind 워터마킹의 문제점 중의 하나인 명암이 급격히 변하는 에지 영역에서의 부정확한 추정을 본 논문에서는 에지 테이블을 사용하여 극복하였다. 뿐만 아니라, 저주파(Low frequency) 영역에 워터마크를 삽입하여, 영상의 품질(Quality)를 유지하였고, 영상의 평가 방법인 PSNR 테스트 및 인간의 눈에서 느껴지는 주관적인 화질도 향상됨을 보였다.