• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2006년도 추계학술발표대회
• /
• pp.69-72
• /
• 2006

최근 생명 정보학 기술의 발달로 마이크로 단위의 실험조작이 가능해짐에 따라 하나의 chip상에서 전체 genome의 expression pattern을 관찰할 수 있게 되었고, 동시에 수 만개의 유전자들 간의 상호작용도 연구가능하게 되었다. 이처럼 DNA 마이크로어레이 기술은 복잡한 생물체를 이해하는 새로운 방향을 제시해주게 되었다. 따라서 이러한 기술을 통해 얻어진 대량의 유전자 정보들을 효과적으로 분석하는 방법이 시급하다. 본 논문에서는 마이크로어레이 실험에서 다양한 원인에 의해 발생하는 잡음(noise)을 줄이거나 제거하는 과정인 정규화과정을 거쳐 특징 추출방법인 SVM(Support Vector Machine) 방법을 이용하여 데이터를 2개의 클래스로 나누고, 표준화 방법들의 성능 비교를 위해 Adaptive Simulated Annealing 알고리즘으로 정확도를 평가하는 분류 시스템을 설계 구현하였다.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2002년도 추계학술발표논문집 (상)
• /
• pp.763-766
• /
• 2002

레이더 영상에서의 물체 위치는 극좌표계로 주어지기 때문에 직각좌표계로 표현되는 일반적인 물체 추적에서의 클러스터링을 통한 물체 추출 방법은 비효율적이다. 본 논문에서는 이러한 레이더 영상의 특성을 고려하여 개선된 ART2클러스터링 기법을 이용하는 방법을 제안하였다. 이진화와 labeling을 통해 추적하고자 하는 물체 외의 물체나 잡영을 제거한 영상에서의 adaptive vigilance parameter를 이용한 ART2 클러스터링 기법의 적용은 추적하고자 하는 물체를 추출함에 있어 우수한 실험 결과를 보였다.

• 한국군사과학기술학회지
• /
• 제27권3호
• /
• pp.319-328
• /
• 2024

Semi-supervised learning is a good way to train a classification model using a small number of labeled and large number of unlabeled data. We applied semi-supervised learning to a synthetic aperture radar(SAR) image classification model with a limited number of datasets that are difficult to create. To address the previous difficulties, semi-supervised learning uses a model trained with a small amount of labeled data to generate and learn pseudo labels. Besides, a lot of number of papers use a single fixed threshold to create pseudo labels. In this paper, we present a semi-supervised synthetic aperture radar(SAR) image classification method that applies different thresholds for each class instead of all classes sharing a fixed threshold to improve SAR classification performance with a small number of labeled datasets.

• 제어로봇시스템학회논문지
• /
• 제3권2호
• /
• pp.179-186
• /
• 1997

In this paper we suggest a development of surface defect inspection algorithms for cold mill strip. The defects which exist in a surface of cold mill strip have a scattering or singular distribution. This paper consists of preprocessing, feature extraction and defect classification. By preprocessing, the binarized defect image is achieved. In this procedure, Top-hit transform, adaptive thresholding, thinning and noise rejection are used. Especially, Top-hit transform using local min/max operation diminishes the effect of bad lighting. In feature extraction, geometric, moment and co-occurrence matrix features are calculated. For the defect classification, multilayer neural network is used. The proposed algorithm showed 15% error rate.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 1996년도 학술대회
• /
• pp.223-227
• /
• 1996

In this paper we propose an adaptive fast image restoration filter, which is suitable for reducing the blocking artifacts in the compressed image in real-time. The proposed restoration filter is based on the observation that quantization operation in a series of coding process is a nonlinear and many-to-one mapping operator. And then we propose an approximated version of constrained optimization technique as a restoration process for removing the nonlinear and space varying degradation operator. We also propose a novel block classification method for adaptively choosing the direction of a highpass filter, which serves as a constraint in the optimization process. The proposed classification method adopts the bias-corrected maximized likelihood, which is used to determine the number of regions in the image for the unsupervised segmentation. The proposed restoration filter can be realized either in the discrete Fourier transform domain or in the spatial domain in the form of a truncated finite impulse response (FIR) filter structure for real-time processing. In order to demonstrate the validity of the proposed restoration filter experimental results will be shown.

• 대한기계학회논문집A
• /
• 제30권5호
• /
• pp.520-527
• /
• 2006

The design of major industrial facilities for the prevention of fatigue failure is customarily done by defining a set of transients and performing a calculation of cumulative usage factor. However, sometimes, the inherent conservatism or lack of details as well as unanticipated transients in old plant may cause maintenance problems. Even though several famous on-line monitoring and diagnosis systems have been developed world-widely, in this paper, a new system fur fatigue monitoring and life evaluation of crane is proposed to reduce customizing effort and purchasing cost. With regard to the system, at first, comprehensive operating transient data has been acquired at critical locations of crane. The real-time data were classified, by using adaptive resonance theory that is one of typical artificial neural network, into representative stress groups. Then the each classified stress pattern was mapped to calculated cumulative usage factor in accordance with ASME procedure. Thereby, promising results were obtained fur the crane and it is believed that the developed system can be applicable to other major facilities extensively.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권3호
• /
• pp.1264-1286
• /
• 2018

In this paper, a novel background prior-based salient object detection framework is proposed to deal with images those are more complicated. We take the superpixels located in four borders into consideration and exploit a mechanism based on image boundary information to remove the foreground noises, which are used to form the background prior. Afterward, an initial foreground prior is obtained by selecting superpixels that are the most dissimilar to the background prior. To determine the regions of foreground and background based on the prior of them, a threshold is needed in this process. According to a fixed threshold, the remaining superpixels are iteratively assigned based on their proximity to the foreground or background prior. As the threshold changes, different foreground priors generate multiple different partitions that are assigned a likelihood of being foreground. Last, all segments are combined into a saliency map based on the idea of similarity voting. Experiments on five benchmark databases demonstrate the proposed method performs well when it compares with the state-of-the-art methods in terms of accuracy and robustness.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권7호
• /
• pp.3286-3300
• /
• 2016

High Efficiency Video Coding (HEVC) Standard, as the latest coding standard, introduces satisfying compression structures with respect to its predecessor Advanced Video Coding (H.264/AVC). The new coding standard can offer improved encoding performance compared with H.264/AVC. However, it also leads to enormous computational complexity that makes it considerably difficult to be implemented in real time application. In this paper, based on machine learning, a fast partitioning method is proposed, which can search for the best splitting structures for Intra-Prediction. In view of the video texture characteristics, we choose the entropy of Gray-Scale Difference Statistics (GDS) and the minimum of Sum of Absolute Transformed Difference (SATD) as two important features, which can make a balance between the computation complexity and classification performance. According to the selected features, adaptive decision trees can be built for the Coding Units (CU) with different size by offline training. Furthermore, by this way, the partition of CUs can be resolved as a binary classification problem. Experimental results have shown that the proposed algorithm can save over 34% encoding time on average, with a negligible Bjontegaard Delta (BD)-rate increase.

• 정보처리학회논문지B
• /
• 제18B권4호
• /
• pp.173-180
• /
• 2011

본 논문은 GPU(Graphics Processing Unit) 에서 CUDA(Compute Unified Device Architecture)를 사용하여 실시간으로 객체를 분할하는 방법을 소개한다. 최근에 감시 시스템, 오브젝트 추적, 모션 분석 등의 많은 응용 프로그램들은 실시간 처리가 요구된다. 이러한 단계의 선행부분인 객체 분할 기법은 기존 CPU 기반의 시스템으로는 실시간 처리에 제약이 발생한다. NVIDIA에서는 Parallel Processing for General Computation 을 위해 그래픽 하드웨어 제약을 개선한 CUDA platform을 제공하고 있다. 본 논문에서는 객체 추출 단계에 대표적인 적응적 가우시안 혼합 배경 모델링(Adaptive Gaussian Mixture Background Modeling) 알고리즘과 Classification 기법으로 사용되는 CCL (Connected Component Labeling) 알고리즘을 적용하였다. 본 논문은 2.4GHz를 갖는 Core2 Quad 프로세서와 비교하여 평가하였고 그 결과 3~4배 이상의 성능향상을 확인할 수 있었다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2003년도 ICCAS
• /
• pp.1315-1318
• /
• 2003

In this paper, a neural network based approach using a self-organizing feature map is proposed for the segmentation of X ray images. A number of algorithms based on such approaches as histogram analysis, region growing, edge detection and pixel classification have been proposed for segmentation of general images. However, few approaches have been applied to X ray image segmentation because of blur of the X ray image and vagueness of its edge, which are inherent properties of X ray images. To this end, we develop a new model based on the neural network to detect objects in a given X ray image. The new model utilizes Mumford-Shah functional incorporating with a modified adaptive SOFM. Although Mumford-Shah model is an active contour model not based on the gradient of the image for finding edges in image, it has some limitation to accurately represent object images. To avoid this criticism, we utilize an adaptive self organizing feature map developed earlier by the authors.[1] It's learning rule is derived from Mumford-Shah energy function and the boundary of blurred and vague X ray image. The evolution of the neural network is shown to well segment and represent. To demonstrate the performance of the proposed method, segmentation of an industrial part is solved and the experimental results are discussed in detail.