• Proceedings of the IEEK Conference
• /
• 2007.07a
• /
• pp.379-380
• /
• 2007

In this paper, we propose the fast hi-linear interpolation method for SR reconstruction. This method reconstructs the HR image rapidity by considering motion vector information for each LR input image. And its calculation used normalized deviation of image data. As using the motion vector information which is obtained at registration error checking process, this proposed can be achieved the fast and simple SR reconstructed image.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.642-644
• /
• 2003

SAC is a compact electro-optical camera for imaging in visible-NIR spectral ranges. SAC provides highresolution images over the wide geometric and spectral ranges: 10 m ground sample distance (GSD) and 50 km swath width in the spectral ranges of 520 ${\sim}$ 890 nm. SAC is designed to produce high quality images: modulation transfer function (MTF) of more than 15 %; signal-to-noise ratio (SNR) of more than 100. The missions of SAC incorporate various imaging operations: multi-spectral imaging; super swath-width imaging with cameras in parallel; along-track stereo imaging with slanted 2 cameras.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2004.11a
• /
• pp.840-845
• /
• 2004

Development of the linear motors is recently required to control a high-speed and high-resolution in the high-integrated and speed process industry. This paper presents vibration analyses as well as measurement standards of the newly developed linear motors through analyzing the vibration characteristics of the advanced products. Vibration experiments are conducted for identifying vibration level during operation. They are also included in the modal test to analyze dynamic characteristics. Analytic data using Finite Element Method (FEM) are compared with the results of the modal. The FEM and experiments make it possible to understand these characteristics. Further, through computer simulation for the behavior of moving part to be vibration source, the best acceleration pattern of moving part movement can be verified to achieve effective moving part positioning and reduce the vibration due to moving part movement.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2019.06a
• /
• pp.298-301
• /
• 2019

이미지 스티칭은 다수의 이미지를 합성하여 카메라의 좁은 시야각(Field of View) 문제를 해결하는 기술이다. 최근 동영상 기반 Panorama, Super Resolution, 360 VR (Virtual Reality) 등의 컨텐츠 사용이 증가함에 따라, 보다 빠르고 정확한 이미지 스티칭 기술의 필요성이 커지고 있다. 지금까지 필요 성능을 만족시키기 위해 많은 알고리즘이 제안되고 있지만, 정확성을 측정하는 객관적 평가 방법은 표준화되지 않고 있다. 최근에서야 PSNR (Peak Signal-to-Noise Ratio) 과 SSIM (Structural Similarity index method) 측정값을 제시하는 방법이 주를 이루고 있지만, 본 논문에서는 PSNR 과 SSIM 측정 방식의 문제점을 밝히고 지역 차분 픽셀 평가 방법을 제안한다. 기하적 유사성과 광도 측정 정보를 포괄하는 LDPM(Local Differential Pixel Mean) 평가 방식을 테스트 이미지를 통해 증명하고 SSIM 과 비교를 통해 해당 평가 방법의 이점을 밝힌다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2018.11a
• /
• pp.62-65
• /
• 2018

신경망은 깊어질수록 gradient vanishing/exploding과 같은 네트워크가 불안정해지는 문제가 발생 한다. 잔차 블록을 이용하여 이러한 문제를 해결 할 수 있다. 본 논문에서는 영상 인식 분야에서 훌륭한 성능을 보여준 잔차 블록 기반의 깊은 합성곱 신경망을 통한 단일 영상 초해상도 복원 기법을 제안 한다. 제안한 알고리듬은 EDSR에 사용된 잔차 블록을 다양한 크기의 합성곱 연산을 통해 영상의 특징들을 다르게 분석하도록 수정하고 VDSR과 비슷한 수준의 복잡도로 구성하여 향상된 성능을 얻었다. 실험 결과, VDSR에 비해 PSNR이 최대 0.1dB까지 증가했다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2018.06a
• /
• pp.198-199
• /
• 2018

본 논문은 초해상도 기술을 위한 CNN 구조를 제안한다. 제안하는 Gradient-SR 은 고해상도 영상이 고주파 신호와 저주파 신호로 분리될 수 있다는 점을 바탕으로 고역 통과 필터인 Sobel Operator 를 CNN 기반으로 구성한다. Gradient-SR 로부터 생성된 고주파 신호는 목표 크기로 보간 된 저해상도 입력 영상과 더해짐으로 고해상도 영상을 생성한다. 실험 영상은 VDSR 이 사용 한 291 개의 영상과 B100 영상을 이용한다. 제안하는 방법은 스케일 팩터 2 에 대한 초해상도 영상 생성 실험에서 약 200%의 속도 향상을 보인다.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2019.11a
• /
• pp.81-82
• /
• 2019

In this study, the Proctor penetration resistance test value and the hardness value of the improved Durometer were compared and analyzed for the change in the FA replacement rate of the concrete using CGS fine aggregate. The results are summarized as follows. It is. 1) The Proctor penetration resistance test value and the hardness value of the improved Durometer showed a high correlation in the form of a curve. 2) About surface finish Super-resolution with the improved Durometer is considered to be useful when using about 50 HD and about 80 HD in the end to determine the surface finishing work time.

• Journal of Semiconductor Engineering
• /
• v.2 no.1
• /
• pp.125-129
• /
• 2021

As the interest of deep learning, techniques to control the color of images in image processing field are evolving together. However, there is no clear standard for color, and it is not easy to find a way to represent only the color itself like the color-palette. In this paper, we propose a novel color palette extraction system by chroma fine-tuning with reinforcement learning. It helps to recognize the color combination to represent an input image. First, we use RGBY images to create feature maps by transferring the backbone network with well-trained model-weight which is verified at super resolution convolutional neural networks. Second, feature maps are trained to 3 fully connected layers for the color-palette generation with a generative adversarial network (GAN). Third, we use the reinforcement learning method which only changes chroma information of the GAN-output by slightly moving each Y component of YCbCr color gamut of pixel values up and down. The proposed method outperforms existing color palette extraction methods as given the accuracy of 0.9140.

• Journal of Information Processing Systems
• /
• v.18 no.3
• /
• pp.293-301
• /
• 2022

Direction of arrival (DOA) estimation of space signals is a basic problem in array signal processing. DOA estimation based on the multiple signal classification (MUSIC) algorithm can theoretically overcome the Rayleigh limit and achieve super resolution. However, owing to its inadequate real-time performance and accuracy in practical engineering applications, its applications are limited. To address this problem, in this study, a DOA estimation algorithm with high parallelism and precision based on an analysis of the characteristics of complex matrix eigenvalue decomposition and the coordinate rotation digital computer (CORDIC) algorithm is proposed. For parallel and single precision, floating-point numbers are used to construct an orthogonal identity matrix. Thus, the efficiency and accuracy of the algorithm are guaranteed. Furthermore, the accuracy and computation of the fixed-point algorithm, double-precision floating-point algorithm, and proposed algorithm are compared. Without increasing complexity, the proposed algorithm can achieve remarkably higher accuracy and efficiency than the fixed-point algorithm and double-precision floating-point calculations, respectively.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.11a
• /
• pp.44-46
• /
• 2022

본 논문에서는 TMIV 부호화 과정에서 개선된 압축성능을 위해 딥러닝을 이용한 초해상화 기술을 적용하는 방식을 제안한다. 제안 방식에서는 TMIV 인코더에서 아틀라스 생성한 후, 해당 아틀라스의 패킹된 뷰들을 downsampling하여 뷰들이 축소된 아틀라스를 생성하는 방식을 사용한다. 생성된 아틀라스는 기존의 방식 그대로 VVC를 이용하여 부복호화를 한다. 복호화된 아틀라스를 렌더링을 위해 뷰로 만드는 과정 중에 딥러닝을 이용한 초해상화 기술을 적용하여 줄어든 뷰들을 원래의 크기로 복원시킨다. 제안 기술을 통해 복원된 뷰의 화질을 유지시킨 채 많은 비트율을 감소시킬 수 있음이 확인된다.