• 한국멀티미디어학회논문지
• /
• 제21권12호
• /
• pp.1425-1430
• /
• 2018

In this paper, we propose a CNN based deep learning algorithm for semantic segmentation of images. In order to improve the accuracy of semantic segmentation, we combined pixel level object classification and image level object classification. The image level object classification is used to accurately detect the characteristics of an image, and the pixel level object classification is used to indicate which object area is included in each pixel. The proposed network structure consists of three parts in total. A part for extracting the features of the image, a part for outputting the final result in the resolution size of the original image, and a part for performing the image level object classification. Loss functions exist for image level and pixel level classification, respectively. Image-level object classification uses KL-Divergence and pixel level object classification uses cross-entropy. In addition, it combines the layer of the resolution of the network extracting the features and the network of the resolution to secure the position information of the lost feature and the information of the boundary of the object due to the pooling operation.

• Journal of Advanced Marine Engineering and Technology
• /
• 제28권4호
• /
• pp.611-617
• /
• 2004

Pixel classification is one of basic image processing issues. The general characteristics of the pixels belonging to various classes are discussed and the radical principles of pixel classification are given. At the same time. a pixel classification scheme based on image direction measure is proposed. As a typical application instance of pixel classification, an adaptive multi-level median filter is presented. An image can be classified into two types of areas by using the direction information measure, that is. smooth area and edge area. Single direction multi-level median filter is used in smooth area. and multi-direction multi-level median filter is taken in the other type of area. What's more. an adaptive mechanism is proposed to adjust the type of the filters and the size of filter window. As a result. we get a better trade-off between preserving details and noise filtering.

• 한국시뮬레이션학회논문지
• /
• 제10권3호
• /
• pp.59-70
• /
• 2001

Typical low-level image processing tasks require thousands of operations per pixel for each input image. Traditional general-purpose computers are not capable of performing such tasks in real time. Yet important features of traditional computers are not exploited by low-level image processing tasks. Since storage requirements are limited to a small number of low-precision integer values per pixel, large hierarchical memory systems are not necessary. The mismatch between the demands of low-level image processing tasks and the characteristics of conventional computers motivates investigation of alternative architectures. The structure of the tasks suggests employing an array of processing elements, one per pixel, sharing instructions issued by a single controller. In this paper we implemented various image processing filtering using the format converter. Also, we realized from conventional gray image process to color image process. This design method is based on realized the large processor-per-pixel array by integrated circuit technology This format converter design has control path implementation efficiently, and can be utilize the high technology without complicated controller hardware.

• 한국통신학회논문지
• /
• 제40권9호
• /
• pp.1672-1677
• /
• 2015

멀티레벨 홀로그래픽 데이터 저장장치는 한 픽셀에 1비트 보다 많은 데이터를 저장할 수 있다. 본 논문에서는 부호율이 8/6=1.33 (bit/pixel)이고 한 픽셀이 4-레벨을 갖는 6픽셀 변조부호를 제안하고 기존에 제안된 4-레벨 6/9 변조부호, 2/3 변조부호의 성능을 비교한다. 제안된 6픽셀 변조부호는 하나의 코드워드가 6픽셀로 구성되어 있고 최소 유클리디안 거리가 3 이상인 변조부호이다. 6픽셀 변조부호는 기존에 제안된 변조부호들 보다 약 1dB 정도의 성능 이득이 있었다.

• 한국정보통신학회논문지
• /
• 제17권1호
• /
• pp.175-182
• /
• 2013

아날로그 형태로 기록되어지는 조위관측 기록의 디지털화는 많은 해양 관련 연구에 도움을 줄 수 있다. 본 논문에서는 조위관측 기록 디지털화의 한 부분인 그래프 영역 분리에 관한 방법을 제안한다. 사용자가 그래프 영역으로 간주되는 하나의 픽셀을 선택하면, 선택된 픽셀의 색상을 이용해 이미지의 상당 부분을 구성하는 배경 픽셀들을 분리한다. 남아 있는 배경 픽셀들과 그래프 영역 픽셀들을 구분하기 위해, 각 열에서 하나의 그래프 픽셀을 결정하고, 이 픽셀을 중심으로 그래프 영역을 분리한다. 실험결과는 본 논문에서 제안한 방법이 이전의 그래프 영역 분리 방법의 단점을 보완하고, 원 이미지의 그래프 영역과 유사한 그래프 영역을 검출할 수 있음을 보여준다.

• 한국가시화정보학회지
• /
• 제20권2호
• /
• pp.78-85
• /
• 2022

In these days, the rapid development in prediction technology using artificial intelligent is being applied in a variety of engineering fields. Especially, dimensionality reduction technologies such as autoencoder and convolutional neural network have enabled the classification and regression of high-dimensional data. In particular, pixel level prediction technology enables semantic segmentation (fine-grained classification), or physical value prediction for each pixel such as depth or surface normal estimation. In this study, the pressure distribution of the ship's surface was estimated at the pixel level based on the artificial neural network. First, a potential flow analysis was performed on the hull form data generated by transforming the baseline hull form data to construct 429 datasets for learning. Thereafter, a neural network with a U-shape structure was configured to learn the pressure value at the node position of the pretreated hull form. As a result, for the hull form included in training set, it was confirmed that the neural network can make a good prediction for pressure distribution. But in case of container ship, which is not included and have different characteristics, the network couldn't give a reasonable result.

• 한국정보통신학회논문지
• /
• 제14권11호
• /
• pp.2555-2562
• /
• 2010

본 논문에서는 수작업을 최소화하면서 조위 기록지로부터 조위 기록 그래프를 추출하는 방법을 제안한다. 사용자가 그래프 상의 하나의 픽셀을 선택하면, 선택된 픽셀과 나머지 픽셀들과의 LAB 색상 공간상에서의 관계를 이용해 자동으로 대부분의 배경 픽셀들을 결정한다. 배경 픽셀들이 결정되면, 각 세로줄에서 그래프 영역으로 판단되는 한개의 픽셀을 추출하고, 이 픽셀을 시작 위치로 하여 나머지 그래프 영역을 추출한다. 실험 결과는 다양한 종류의 조위 기록지로부터 최소한의 수작업만으로 조위 기록 그래프 영역을 검출할 수 있음을 보여준다.

• 한국통신학회논문지
• /
• 제39A권10호
• /
• pp.610-612
• /
• 2014

멀티레벨 홀로그래픽 데이터 스토리지 시스템은 픽셀당 1비트 이상을 저장할 수 있기 때문에 용량에서 큰 이점을 갖는다. 만약 부호율이 2/3인 변조부호라면 2/3(symbol/pixel)은 4/3(bit/pixel)이고, 이것은 1pixel에 약 1.3개 bit를 담을 수 있다. 본 논문에서는 멀티레벨 홀로그래픽 데이터 저장장치에서 한 픽셀이 4-레벨을 가지는 경우, 최소 유클리디안 거리가 3과 4인 변조부호를 각각 제안하였다. 제안한 변조부호는 랜덤한 경우보다 훨씬 더 좋은 성능을 보였고, 거리가 클수록, 성능이 더 좋아지는 것을 보였다.

• 한국전기전자재료학회:학술대회논문집
• /
• 한국전기전자재료학회 1998년도 추계학술대회 논문집
• /
• pp.137-140
• /
• 1998

An active-matrix LCD using thin film transistors (TFT) has been widely recognized as having potential for high-quality color flat-panel displays. Pixel-Design Array Simulation Tool (PDAST) was used to profoundly understand the gate si후미 distortion and pixel charging capability. which are the most critical limiting factors for high-quality TFT-LCDs. Since PDAST can simulate the gate, data and pixel voltages of a certain pixel on TFT array at any time and at any location on an array, the effect of the resistivity of gate line material on the pixel operations can be effectively analyzed. The gate signal delay, pixel charging ratio and level-shift of the pixel voltage were simulated with varying the parameters. The information obtained from this study could be utilized to design the larger area and finer image quality panel.

• 한국전기전자재료학회:학술대회논문집
• /
• 한국전기전자재료학회 1998년도 춘계학술대회 논문집
• /
• pp.321-324
• /
• 1998

Pixel-Design Array Simulation Tool(PDAST) was used to profoundly the gate signal distortion and pixel changing capability, which are the most critical limiting factors for high-quality TFT-LCDs. Since PDAST can simulate the gate, data and pixel voltages of a certain pixel on TFT array at any time and at any location on an array, the effect of the resistivity of gate line material on the pixel operations can be effectively analyzed. The gate signal delay, pixel charging ratio, level-shift of the pixel voltage were simulated with varying the resis5tivity of the gate line material. The information obtained from this study could be utilized to design the larger area and finer image quality panel.