• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.4
• /
• pp.171-176
• /
• 2022

CNN is a type of deep learning and is a neural network used to process images or image data. The filter traverses the image and extracts features of the image to distinguish the image. Deep learning has the characteristic that the more data, the better models can be made, and CNN uses a method of artificially increasing the amount of data by means of data augmentation such as rotation, zoom, shift, and flip to compensate for the weakness of less data. When learning CNN, we would like to check whether outline image learning is helpful in improving performance compared to conventional data augmentation techniques.

• Journal of IKEEE
• /
• v.27 no.4
• /
• pp.464-471
• /
• 2023

In this paper, we propose a method to reduce age distortion in facial expression image generation using StyleGAN Encoder. The facial expression image generation process first creates a face image using StyleGAN Encoder, and changes the expression by applying the learned boundary to the latent vector using SVM. However, when learning the boundary of a smiling expression, age distortion occurs due to changes in facial expression. The smile boundary created in SVM learning for smiling expressions includes wrinkles caused by changes in facial expressions as learning elements, and it is determined that age characteristics were also learned. To solve this problem, the proposed method calculates the correlation coefficient between the smile boundary and the age boundary and uses this to introduce a method of adjusting the age boundary at the smile boundary in proportion to the correlation coefficient. To confirm the effectiveness of the proposed method, the results of an experiment using the FFHQ dataset, a publicly available standard face dataset, and measuring the FID score are as follows. In the smile image, compared to the existing method, the FID score of the smile image generated by the ground truth and the proposed method was improved by about 0.46. In addition, compared to the existing method in the smile image, the FID score of the image generated by StyleGAN Encoder and the smile image generated by the proposed method improved by about 1.031. In non-smile images, compared to the existing method, the FID score of the non-smile image generated by the ground truth and the method proposed in this paper was improved by about 2.25. In addition, compared to the existing method in non-smile images, it was confirmed that the FID score of the image generated by StyleGAN Encoder and the non-smile image generated by the proposed method improved by about 1.908. Meanwhile, as a result of estimating the age of each generated facial expression image and measuring the estimated age and MSE of the image generated with StyleGAN Encoder, compared to the existing method, the proposed method has an average age of about 1.5 in smile images and about 1.63 in non-smile images. Performance was improved, proving the effectiveness of the proposed method.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.04a
• /
• pp.710-712
• /
• 2003

본 논문에서는 시각적. 공간적 정보로 멀티미디어 분야에서 다양한 응용이 가능한 이미지검색을 위해 색상특징정보와 모양특징정보를 멀티인덱스키로 구성하여 질의 이미지의 입력 시 자동으로 색상특징정보와 모양특징정보를 동시에 추출하여 유사한 이미지를 검색할 수 있는 내용기반 이미지 검색시스템을 제안하였다. 제안된 시스템은 기존의 단일 특징정보를 이용한 방법이나 2가지 이상의 특징정보를 단계적으로 검색하는 방법에 비해 향상된 효율성과 신속성을 보이고 있다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.10a
• /
• pp.322-324
• /
• 2001

기존의 이미지 검색에서는 원하는 이미지를 검색하기 위하여 사용자가 이미지의 가시적 속성을 정확히 표현하도록 요구함으로써 질의가 제한되었다. 본 논문에서는 색상으로부터 유추될 수 있는 감성 형용사를 감성 용어 시소러스로 구축하여 감성 기반의 이미지 검색이 가능하도록 하였다. 감성 용어 시소러스를 이용함으로써 ‘부드러운’, ‘세련된’등과 같은 감성 용어를 검색의 질의어로 사용할 수 있게 되어 사용자의 검색 의도를 보다 정확하게 표현할 수 있게 되고, 검색의 결과에 대한 만족도를 향상 시킬 수 있다.

• The KIPS Transactions:PartB
• /
• v.19B no.2
• /
• pp.93-98
• /
• 2012

Recently, we can easily create high resolution images with digital cameras for high-performance and make use them at variety fields. Especially, the image stitching method which adjusts couple of images has been researched. Image stitching can be used for military purposes such as satellites and reconnaissance aircraft, and computer vision such as medical image and the map. In this paper, we have proposed fast image stitching based on improved SURF algorithm using meaningful features in the process of images matching after extracting features from scenery image. The features are extracted in each image to find out corresponding points. At this time, the meaningful features can be searched by removing the error, such as noise, in extracted features. And these features are used for corresponding points on image matching. The total processing time of image stitching is improved due to the reduced time in searching out corresponding points. In our results, the processing time of feature matching and image stitching is faster than previous algorithms, and also that method can make natural-looking stitched image.

• Science of Emotion and Sensibility
• /
• v.9 no.1
• /
• pp.63-75
• /
• 2006

This is a basic research aimed at enhancing a brand image by unifying color images of a brand at both online and offline stores. h survey was conducted on color images of online and offline stores to measure the level of consistency between consumers' perception of the color image and the brand's color strategy. A brand preference survey was also performed to shed light on the relationship between the brand preference and the consistency of color images at online and offline stores. This study is committed to building an effective color strategy, based on the brand strategy, through web color planning that is consistent with the offline color image. This paper started with a specific purpose of devising solutions in regard to the prior study, entitled 'Study on Color Strategies in Brand Coffee', which recognizes the need for consistent color images, or integrated color strategies, in online and offline stores. I specifically took the approach of looking at color planning in the stage of designing a website. Two standards were used to analyze emotional and functional aspects of color images: color grading by I.R.I and five communicators of color information by Rouge. Direct visits were made to the offline stores for surveys to address the shortcomings of the prior study, in which offline stimuli were limited to printed materials. The direct visits enabled a study of an overall color image of the offline stores, while providing a set of substantial and specific guidelines for designing colors for a website. I hope this study goes a long way toward enhancing the level of consistency between online and offline color images based on the brand's unique color strategy, and thereby improving the overall level of brand image.

• Proceedings of the KIEE Conference
• /
• 2007.10a
• /
• pp.173-174
• /
• 2007

지문인식은 생체인식에서 가장 오래되고 널리 사용된 방법이다. 지문인식이 널리 사용됨에도 불구하고, 지문의 특징점들에 대한 통계 이론이 없다. 지문 특징점들의 통계를 연구하기 전에 믿을 수 있는 특징점들을 추출하는 것이다. 그러나, 지문 이미지들은 피부와 누르는 정도의 조건에 따라서 퇴화되고, 변질된다. 따라서 지문이미지의 품질 향상은 특징추출전에 선행되어야 한다. 본 연구에서는 지문 이미지의 품질 향상을 위해 가보필터를 사용할 것이고, 가보 필터를 사용하기 위한 여러 가지 방법들에 대해서 알아 볼 것이다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2016.04a
• /
• pp.737-739
• /
• 2016

파노라마 이미지 생성 기법의 중요한 부분은 입력 영상들로부터 특징을 추출하고, 영상간의 대응점을 찾는 작업이다. 특징 추출할 때 영상의 회전, 스케일, 밝기 변화에 강건하고 수행속도가 비교적 빠른 검출 알고리즘을 사용한다. 파노라마 이미지 생성 과정에 있어서 실제 대응하는 점들을 크게 다루기 때문에 불필요한 영역의 특징들은 오히려 연산속도의 방해 요소가 된다. 본 논문에서는 특징 추출 영역을 제한함으로써 특징 매칭 횟수 감소 및 속도 향상 방법을 제안한다. 특징의 개수가 감소되면 매칭 횟수 감소되고, 이후 이루어질 여러 계산량도 줄어 속도가 향상된다. 본 연구에 SURF(Speeded-Up Robust Feature) 알고리즘을 사용하였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.05a
• /
• pp.567-569
• /
• 2023

GAN은 이미지 생성모델로서 이미지 공간에서 좋은 결과를 보여왔다. 우리는 이러한 GAN의 능력을 더욱 향상하기 위하여 본 연구에서 주파수 영역에서 이미지를 학습하고 생성하는 새로운 방법을 제안한다. 이를 위하여 먼저 학습데이터를 2D FFT로 주파수 영역으로 변환한 후 변환된 학습데이터를 GAN이 학습하게 한다. 학습 후에 GAN은 새로운 이미지를 생성하며 생성된 이미지를 2D IFFT하여 이미지 공간으로 변환한다. 이렇게 주파수 영역에서 이미지를 생성하는 방법은 이미지 공간에서 생성하는 방법보다 다양한 장점이 있다. 생성된 이미지의 품질을 평가하기 위하여 4개 데이터 셋에 4개의 평가지표를 사용하여 평가한 결과 주파수 영역에서 생성한 이미지가 IS, P&R, D&C 측면에서 더 좋은 것으로 평가되었다.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.8
• /
• pp.837-841
• /
• 2010

This paper proposes an effective similarity measure for content-based image retrieval using MPEG-7 DCD. The proposed method can measure the similarity of images with the percentage of dominant colors extracted from images. As the result of experiments, we achieved a significant improvement of 18.92% with global DCD and 47.22% with local DCD in ANMRR than the result by QHDM. This result shows that the proposed method is an effective similarity measure for content-based image retrieval. Especially, our method is useful for region-based image retrieval.