• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2009년도 IWAIT
• /
• pp.254-257
• /
• 2009

In this paper, we propose a learning-based super-resolution algorithm. In the proposed algorithm, a multi-resolution wavelet approach is adopted to perform the synthesis of local high-frequency features. To obtain a high-resolution image, wavelet coefficients of two dominant LH- and HL-bands are estimated based on wavelet frames. In order to prepare more efficient training sets, the proposed algorithm utilizes the LH-band and transposed HL-band. The training sets are then used for the estimation of wavelet coefficients for both LH- and HL-bands. Using the estimated high frequency bands, a high resolution image is reconstructed via the wavelet transform. Experimental results demonstrate that the proposed scheme can synthesize high-quality images.

• 대한의용생체공학회:의공학회지
• /
• 제29권5호
• /
• pp.348-354
• /
• 2008

Segmenting vessels in lower extremity CT images is very difficult because of gray level variation, connection to bones, and their small sizes. Instead of segmenting vessels, we propose an approach that segments bones and subtracts them from the original CT images. The subtracted images can contain not only connected vessel structures but also isolated vessels, which are very difficult to detect using conventional vessel segmentation methods. The proposed method initially grows a 3-dimensional (3D) volume with a seeded region growing (SRG) using an adaptive threshold and then detects junctions and forked branches. The forked branches are classified into either bone branches or vessel branches based on appearance, shape, size change, and moving velocity of the branch. The final volume is re-grown by collecting connected bone branches. The algorithm has produced promising results for segmenting bone structures in several tens of vessel-enhanced CT image data sets of lower extremities.

• 대한토목학회논문집
• /
• 제28권6D호
• /
• pp.915-922
• /
• 2008

불투수층은 자연적인 침투를 허용하지 않는 인위적인 토지피복상태로, 도시화율을 추정하거나 도시의 환경변화 정도를 분석하기 위한 척도로 사용되어 왔다. 수문학적인 관점에서 불투수층은 단기 유출현상에 큰 영향을 끼치는 요소로 급속한 도시화로 인해 불투수층의 영향이 더욱 커짐에 따라 불투수층의 추정에 대한 필요성이 증가하고 있다. 따라서 본 연구에서는 불투수층을 추정하기 위해 중랑천 유역을 대상지역으로 선정하고, $30m{\times}30m$ 공간해상도의 Landsat-7 ETM+ 영상과 $1m{\times}1m$의 고해상도 위성영상을 구축하였으며 tasselled cap 변환과 식생지수(NDVI) 변환을 수행하여 다양한 예측변수를 고려하였다. 수집된 학습자료에 분류 및 회귀트리를 적용하여 불투수층 추정모델을 구성하였고, 이를 지도화하여 중랑천 유역의 불투수층을 나타냈다.

• International Journal of Advanced Culture Technology
• /
• 제5권1호
• /
• pp.76-82
• /
• 2017

An image hash, a discriminative and robust summary of an image, should be robust against quality-preserving signal processing steps, while being pairwise independent for perceptually different inputs. In order to improve the hash matching performance, this paper proposes a probabilistic dissimilarity matching. Instead of extracting the binary hash from the query image, we compute the probability that the intermediate hash vector of the query image belongs to each quantization bin, which is referred to as soft quantization binning. The probability is used as a weight in comparing the binary hash of the query with that stored in a database. A performance evaluation over sets of image distortions shows that the proposed probabilistic matching method effectively improves the hash matching performance as compared with the conventional Hamming distance.

• 한국컴퓨터정보학회논문지
• /
• 제24권6호
• /
• pp.11-19
• /
• 2019

In this paper, a new Scene-based Nonuniformity Correction (SBNUC) method is proposed by applying Image Roughness-like and Spatial Noise cost functions on deep neural network structure. The classic approaches for nonuniformity correction require generally plenty of sequential image data sets to acquire accurate image correction offset coefficients. The proposed method, however, is able to estimate offset from only a couple of images powered by the characteristic of deep neural network scheme. The real world SWIR image set is applied to verify the performance of proposed method and the result shows that image quality improvement of PSNR 70.3dB (maximum) is achieved. This is about 8.0dB more than the improved IRLMS algorithm which preliminarily requires precise image registration process on consecutive image frames.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 1999년도 하계종합학술대회 논문집
• /
• pp.974-977
• /
• 1999

In this paper, we have proposed the hardware architecture which implements the algorithm for retaining the connectivity which prevents disconnecting in the gray-scale image thinning To perform the image thinning in a real time which find a skeleton in image, it is necessary to examine the connectivity of the skeleton in a real time. The proposed architecture finds the connectivity number in the 4-clock period. The architecture is consists of three blocks, PS(Parallel to Serial) Converter and State Generator and Ridge Checker. The PS Converter changes the 3$\times$3 gray level image to four sets of image pixels. The State Generator examine the connectivity of the central pixel by searching the data from the PS Converter. the 3$\times$3 gray level image determines. The Ridge Checker determines whether the central pixel is on the skeleton or not The proposed architecture finds the connectivity of the central pixel in a 3$\times$3 gray level image in the 4-clocks. The total circuits are verified by the design tools and operate correctly.

• International Journal of Internet, Broadcasting and Communication
• /
• 제13권2호
• /
• pp.7-13
• /
• 2021

Machine vision is a technology that helps the computer as if a person recognizes and determines things. In recent years, as advanced technologies such as optical systems, artificial intelligence and big data advanced in conventional machine vision system became more accurate quality inspection and it increases the manufacturing efficiency. In machine vision systems using deep learning, the image quality of the input image is very important. However, most images obtained in the industrial field for quality inspection typically contain noise. This noise is a major factor in the performance of the machine vision system. Therefore, in order to improve the performance of the machine vision system, it is necessary to eliminate the noise of the image. There are lots of research being done to remove noise from the image. In this paper, we propose an autoencoder based machine vision system to eliminate noise in the image. Through experiment proposed model showed better performance compared to the basic autoencoder model in denoising and image reconstruction capability for MNIST and fashion MNIST data sets.

• 한국통신학회논문지
• /
• 제25권4B호
• /
• pp.743-754
• /
• 2000

본 논문에서는 선형 위상 응답 특성을 갖고 쌍직교 웨이블릿 변환(biorthogonal wavelet transform)을 이용하여 영상을 다중 해상도롤 분해하고 분해된 부 밴드들을 최대 분할 이득(maximum classificatio gain)을 이용하여 분할하고 분할된 클래스 별로 최적 비트 할당을 통해 양자화 한 후, 컨텍스트 기반 비트 평면 부호화를 이용한 영상 압축 방법을 제안한다. 제안하는 방법을 양자화 된 계수들에 대해 부호변환(shuffling)과정을 거쳐 비트 평면 부호화 시, sign 비트 평면을 제거하여 부호화 성능을 향상시키며, 컨텍스트(context)기반 비트 평면 부호화 시, 각각의 클래스 및 밴드별로 균일한(uniform)컨텍스트를 부여하지 않고 현재 비트 평면과 이전 비트 평면 사이의 상관도(correlation)를 측정하여 가장 큰 상관도를 갖는 방향으로 컨텍스트를 부여하여 보다 효율적인 부호화 성능을 얻는 방법으로 부호화를 통해 제안한 방법의 우수성을 입증하고자 한다.

• 방송공학회논문지
• /
• 제16권4호
• /
• pp.669-679
• /
• 2011

본 논문에서는 불규칙한 샘플 영상에 대해 비지역적 블록 기반의 웨이블릿 영상 잡음 제거 기법을 포함하는 POCS (projection on convex sets) 보간법을 제안한다. 이 방법은 보간을 수행하기 위한 볼록 집합을 정의하고, 해당 볼록 집합으로 반복 투영하여 최종 보 간 영상을 생성한다. 우선 Delaunay 삼각화를 이용하여 불규칙한 샘플 영상을 균일 격자 영상으로 투영한다. 두 번째 단계에서 비지역 적 블록 기반의 웨이블릿 영상 잡음 제거 기법을 적용하고, 세 번째 단계에서 원본 관찰된 화소값을 주입한다. 두 번째 단계와 세 번 째 단계를 반복적으로 투영하고, 마지막 단계로 경계선 검출을 통해 비경계 영역에 비지역적 잡음 제거 기법을 수행하여 최종 보간 영 상을 생성한다. 본 논문에서는 여러 실험 영상을 사용하여 기존 제안된 기법 대비 제안한 기법의 효율성을 입증하였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권4호
• /
• pp.1166-1186
• /
• 2022

Artistic font design has become an integral part of visual media. However, without prior knowledge of the font domain, it is difficult to create distinct font styles. When the number of characters is limited, this task becomes easier (e.g., only Latin characters). However, designing CJK (Chinese, Japanese, and Korean) characters presents a challenge due to the large number of character sets and complexity of the glyph components in these languages. Numerous studies have been conducted on automating the font design process using generative adversarial networks (GANs). Existing methods rely heavily on reference fonts and perform font style conversions between different fonts. Additionally, rather than capturing style information for a target font via multiple style images, most methods do so via a single font image. In this paper, we propose a network architecture for generating multilingual font sets that makes use of geometric structures as content. Additionally, to acquire sufficient style information, we employ multiple style images belonging to a single font style simultaneously to extract global font style-specific information. By utilizing the geometric structural information of content and a few stylized images, our model can generate an entire font set while maintaining the style. Extensive experiments were conducted to demonstrate the proposed model's superiority over several baseline methods. Additionally, we conducted ablation studies to validate our proposed network architecture.