• Journal of the Korean Physical Society
• /
• 제73권11호
• /
• pp.1644-1649
• /
• 2018

Aiming at infrared thermal images with different buried depth defects, we study a variety of image segmentation algorithms based on the threshold to develop global search ability and the ability to find the defect area accurately. Firstly, the iterative thresholding method, the maximum entropy method, the minimum error method, the Ostu method and the minimum skewness method are applied to image segmentation of the same infrared thermal image. The study shows that the maximum entropy method and the minimum error method have strong global search capability and can simultaneously extract defects at different depths. However none of these five methods can accurately calculate the defect area at different depths. In order to solve this problem, we put forward a strategy of "divide and conquer". The infrared thermal image is divided into several local thermal maps, with each map containing only one defect, and the defect area is calculated after local image processing of the different buried defects one by one. The results show that, under the "divide and conquer" strategy, the iterative threshold method and the Ostu method have the advantage of high precision and can accurately extract the area of different defects at different depths, with an error of less than 5%.

• 한국소프트웨어감정평가학회 논문지
• /
• 제17권2호
• /
• pp.1-10
• /
• 2021

감정인식은 응용 분야의 다양성으로 많은 연구가 이루어지고 있는 기술이며, RGB 영상은 물론 열화상을 이용한 감정인식의 필요성도 높아지고 있다. 열화상의 경우는 RGB 영상과 비교해 조명 문제에 거의 영향을 받지 않는 장점이 있으나 낮은 해상도로 성능 높은 인식 기술을 필요로 한다. 본 논문에서는 얼굴 열화상 기반 감정인식의 성능을 높이기 위한 Divide and Conquer 기반의 CNN 학습전략을 제안하였다. 제안된 방법은 먼저 분류가 어려운 유사 감정 클래스를 confusion matrix 분석을 통해 동일 클래스 군으로 분류하도록 학습시키고, 다음으로 동일 클래스 군으로 분류된 감정 군을 실제 감정으로 다시 인식하도록 문제를 나누어서 해결하는 방법을 사용하였다. 실험을 통하여, 제안된 학습전략이 제시된 모든 감정을 하나의 CNN 모델에서 인식하는 경우보다 모든 실험에서 높은 인식성능을 보이는 것을 확인하였다.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2022년도 춘계학술대회
• /
• pp.665-666
• /
• 2022

미분 진화는 연속 최적화 문제에 대한 효율적인 알고리즘이다. 그러나 대규모 최적화 문제를 해결하기 위해 미분 진화를 적용하면 성능이 빠르게 저하되고 런타임이 기하급수적으로 증가한다. 이 문제를 극복하기 위해 Spark(SparkDECC라고 함)를 기반으로 하는 새로운 협력 공진화 미분 진화를 제안한다. 분할 정복 전략은 SparkDECC에서 사용된다.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2021년도 추계학술대회
• /
• pp.559-560
• /
• 2021

차등 진화는 연속 최적화 문제를 해결하기 위한 효율적인 알고리즘이다. 그러나 대규모 최적화 문제를 해결하기 위해 차등 진화를 적용하면 성능이 급격히 저하되고 런타임이 기하급수적으로 증가한다. 따라서 Spark(SparkDECC로 알려짐)를 기반으로 하는 새로운 협력 공진화 차동 진화가 제안된다. 분할 정복 전략은 SparkDECC에서 사용된다.

• 대한전기학회논문지:시스템및제어부문D
• /
• 제49권6호
• /
• pp.322-329
• /
• 2000

BIST(Built-in Self Test) has been considered as the most promising DFT(design-for-test) scheme for the present and future test strategy. The most serious problem in applying BIST(Built-in Self Test) into a large circuit is the excessive increase in test time. This paper is focused on this problem. We proposed a new BIST construction scheme which uses a parallel divide-and-conquer method. The circuit division is performed with respect to some internal nodes called test points. The test points are selected by considering the nodal connectivity of the circuit rather than the testability of each node. The test patterns are generated by only one linear feedback shift register(LFSR) and they are shared by all the divided circuits. Thus, the test for each divided circuit is performed in parallel. Test responses are collected from the test point as well as the primary outputs. Even though the divide-and-conquer scheme is used and test patterns are generated in one LFSR, the proposed scheme does not lose its pseudo-exhaustive property. We proposed a selection procedure to find the test points and it was implemented with C/C++ language. Several example circuits were applied to this procedure and the results showed that test time was reduced upto 1/2151 but the increase in the hardware overhead or the delay increase was not much high. Because the proposed scheme showed a tendency that the increasing rates in hardware overhead and delay overhead were less than that in test time reduction as the size of circuit increases, it is expected to be used efficiently for large circuits as VLSI and ULSI.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제9권11호
• /
• pp.4502-4518
• /
• 2015

Fisher linear discriminant analysis (LDA) is one of the most popular projection techniques for feature extraction and has been widely applied in face recognition. However, it cannot be used when encountering the single sample per person problem (SSPP) because the intra-class variations cannot be evaluated. In this paper, we propose a novel method called local similarity based linear discriminant analysis (LS_LDA) to solve this problem. Motivated by the "divide-conquer" strategy, we first divide the face into local blocks, and classify each local block, and then integrate all the classification results to make final decision. To make LDA feasible for SSPP problem, we further divide each block into overlapped patches and assume that these patches are from the same class. To improve the robustness of LS_LDA to outliers, we further propose local similarity based median discriminant analysis (LS_MDA), which uses class median vector to estimate the class population mean in LDA modeling. Experimental results on three popular databases show that our methods not only generalize well SSPP problem but also have strong robustness to expression, illumination, occlusion and time variation.

• Journal of information and communication convergence engineering
• /
• 제19권3호
• /
• pp.155-160
• /
• 2021

Differential evolution is an efficient algorithm for solving continuous optimization problems. However, its performance deteriorates rapidly, and the runtime increases exponentially when differential evolution is applied for solving large-scale optimization problems. Hence, a novel cooperative coevolution differential evolution based on Spark (known as SparkDECC) is proposed. The divide-and-conquer strategy is used in SparkDECC. First, the large-scale problem is decomposed into several low-dimensional subproblems using the random grouping strategy. Subsequently, each subproblem can be addressed in a parallel manner by exploiting the parallel computation capability of the resilient distributed datasets model in Spark. Finally, the optimal solution of the entire problem is obtained using the cooperation mechanism. The experimental results on 13 high-benchmark functions show that the new algorithm performs well in terms of speedup and scalability. The effectiveness and applicability of the proposed algorithm are verified.

• BMB Reports
• /
• 제52권4호
• /
• pp.239-249
• /
• 2019

Membrane-anchored full-length MET stimulated by its ligand HGF/SF induces various biological responses, including survival, growth, and invasion. This panel of responses, referred to invasive growth, is required for embryogenesis and tissue regeneration in adults. On the contrary, MET deregulation is associated with tumorigenesis in many kinds of cancer. In addition to its well-documented ligand-stimulated downstream signaling, the receptor can be cleaved by proteases such as secretases, caspases, and calpains. These cleavages are involved either in MET receptor inactivation or, more interestingly, in generating active fragments that can modify cell fate. For instance, MET fragments can promote cell death or invasion. Given a large number of proteases capable of cleaving MET, this receptor appears as a prototype of proteolytic-cleavage-regulated receptor tyrosine kinase. In this review, we describe and discuss the mechanisms and consequences, both physiological and pathological, of MET proteolytic cleavages.

• 한국지능시스템학회논문지
• /
• 제13권4호
• /
• pp.409-415
• /
• 2003

Recently, the modular learning are very popular and receive much attention for pattern classification. The modular learning method based on the "divide and conquer" strategy can not only solve the complex problems, but also reach a better result than a single classifier′s on the learning quality and speed. In the neural network area, some researches that take the modular learning approach also have been made to improve classification performance. In this paper, we propose a simple modular neural network for characters recognition of vehicle number plate and evaluate its performance on the clustering methods of feature vectors used in constructing subnetworks. We implement two clustering method, one is grouping similar feature vectors by K-means clustering algorithm, the other grouping unsimilar feature vectors by our proposed algorithm. The experiment result shows that our algorithm achieves much better performance.

• 로봇학회논문지
• /
• 제2권4호
• /
• pp.304-309
• /
• 2007

This paper describes an efficient path generation method for area coverage. Its applications include robots for de-mining, cleaning, painting, and so on. Our method is basically based on a divide and conquer strategy. We developed a novel cell decomposition algorithm that divides a given area into several cells. Each cell is covered by a robot motion that requires minimum time to cover the cell. Using this method, completeness and time efficiency of coverage are easily achieved. For the completeness of coverage in dynamic environments, we also propose a path following method that makes the robot cover missed areas as a result of the presence of unknown obstacles. The effectiveness of the method is verified using computer simulations.