• Journal of Multimedia Information System
• /
• v.7 no.1
• /
• pp.1-10
• /
• 2020

The image analysis is an important and predominant task for classifying the different parts of the image. The analysis of complex image analysis like histopathological define a crucial factor in oncology due to its ability to help pathologists for interpretation of images and therefore various feature extraction techniques have been evolved from time to time for such analysis. Although deep reinforcement learning is a new and emerging technique but very less effort has been made to compare the deep learning and deep reinforcement learning for image analysis. The paper highlights how both techniques differ in feature extraction from complex images and discusses the potential pros and cons. The use of Convolution Neural Network (CNN) in image segmentation, detection and diagnosis of tumour, feature extraction is important but there are several challenges that need to be overcome before Deep Learning can be applied to digital pathology. The one being is the availability of sufficient training examples for medical image datasets, feature extraction from whole area of the image, ground truth localized annotations, adversarial effects of input representations and extremely large size of the digital pathological slides (in gigabytes).Even though formulating Histopathological Image Analysis (HIA) as Multi Instance Learning (MIL) problem is a remarkable step where histopathological image is divided into high resolution patches to make predictions for the patch and then combining them for overall slide predictions but it suffers from loss of contextual and spatial information. In such cases the deep reinforcement learning techniques can be used to learn feature from the limited data without losing contextual and spatial information.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.589-591
• /
• 2023

현재 수치예보 시스템은 항공기, 위성 등 다양한 센서에서 얻은 다종 관측 데이터를 동화하여 대기 상태를 추정하고 있지만, 관측변수 또는 물리량이 서로 다른 관측들을 처리하기 위한 계산 복잡도가 매우 높다. 본 연구에서 기존 시스템의 계산 효율성을 개선하여 관측을 평가하거나 전처리하는 데에 효율적으로 활용하기 위해, 각 관측의 특성을 고려한 자기 지도학습 방법을 통해 멀티모달 기상관측으로부터 실제 대기 상태를 추정하는 방법론을 제안하고자 한다. 비균질적으로 수집되는 멀티모달 기상관측 데이터를 융합하기 위해, (i) 기상관측의 heterogeneous network를 구축하여 개별 관측의 위상정보를 표현하고, (ii) pretext task 기반의 self-supervised learning을 바탕으로 개별 관측의 특성을 표현한다. (iii) Graph neural network 기반의 예측 모델을 통해 실제에 가까운 대기 상태를 추정한다. 제안하는 모델은 대규모 수치 시뮬레이션 시스템으로 수행되는 기존 기술의 한계점을 개선함으로써, 이상 관측 탐지, 관측의 편차 보정, 관측영향 평가 등 관측 전처리 기술로 활용할 수 있다.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2001.01a
• /
• pp.65-70
• /
• 2001

We consider a hyper-redundant system that consists of many uniform units. The hyper-redundant system has many degrees of freedom and it can accomplish various tasks. Applysing the reinforcement learning to the hyper-redundant system is very attractive because it is possible to acquire various behaviors for various tasks automatically. In this paper we present a new reinforcement learning algorithm "Q-learning with propagation of motion". The algorithm is designed for the multi-agent systems that have strong connections. The proposed algorithm needs only one small Q-table even for a large scale system. So using the proposed algorithm, it is possible for the hyper-redundant system to learn the effective behavior. In this algorithm, only one leader agent learns the own behavior using its local information and the motion of the leader is propagated to another agents with time delay. The reward of the leader agent is given by using the whole system information. And the effective behavior of the leader is learned and the effective behavior of the system is acquired. We apply the proposed algorithm to a snake-like hyper-redundant robot. The necessary condition of the system to be Markov decision process is discussed. And the computer simulation of learning the locomotion is demonstrated. From the simulation results we find that the task of the locomotion of the robot to the desired point is learned and the winding motion is acquired. We can conclude that our proposed system and our analysis of the condition, that the system is Markov decision process, is valid.

• Journal of Internet Computing and Services
• /
• v.1 no.1
• /
• pp.63-72
• /
• 2000

Data mining means a process of nontrivial extraction of hidden knowledge or potentially useful information from data in large databases. Data mining algorithm is a multi-disciplinary field of research; machine learning, statistics, and computer science all make a contribution. Different classification schemes can be used to categorize data mining methods based on the kinds of tasks to be implemented and the kinds of application classes to be utilized, and classification has been identified as an important task in the emerging field of data mining. Since classification is the basic element of human's way of thinking, it is a well-studied problem in a wide varietyof application. In this paper, we propose a classifier system based on genetic algorithm with robust property, and the proposed system is evaluated by applying it to nDmC problem related to classification task in data mining.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.16 no.5
• /
• pp.209-214
• /
• 2021

Panoptic segmentation, which is now widely used in computer vision such as medical image analysis, and autonomous driving, helps understanding an image with holistic view. It identifies each pixel by assigning a unique class ID, and an instance ID. Specifically, it can classify 'thing' from 'stuff', and provide pixel-wise results of semantic prediction and object detection. As a result, it can solve both semantic segmentation and instance segmentation tasks through a unified single model, producing two different contexts for two segmentation tasks. Semantic segmentation task focuses on how to obtain multi-scale features from large receptive field, without losing low-level features. On the other hand, instance segmentation task focuses on how to separate 'thing' from 'stuff' and how to produce the representation of detected objects. With the advances of both segmentation techniques, several panoptic segmentation models have been proposed. Many researchers try to solve discrepancy problems between results of two segmentation branches that can be caused on the boundary of the object. In this survey paper, we will introduce the concept of panoptic segmentation, categorize the existing method into two representative methods and explain how it is operated on two methods: top-down method and bottom-up method. Then, we will analyze the performance of various methods with experimental results.

• Journal of Intelligence and Information Systems
• /
• v.29 no.1
• /
• pp.203-214
• /
• 2023

The Korea Customs Service is efficiently handling business with an electronic customs system that can effectively handle one-stop business. This is the case and a more effective method is needed. Import and export require HS Code (Harmonized System Code) for classification and tax rate application for all goods, and item classification that classifies the HS Code is a highly difficult task that requires specialized knowledge and experience and is an important part of customs clearance procedures. Therefore, this study uses various types of data information such as product name, product description, and product image in the item classification request form to learn and develop a deep learning model to reflect information well based on Multimodal representation learning. It is expected to reduce the burden of customs duties by classifying and recommending HS Codes and help with customs procedures by promptly classifying items.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.474-476
• /
• 2021

Glioblastoma is the most common brain malignancies arising from glial cells. Early diagnosis and treatment plan establishment are important, and cancer is diagnosed mainly through T1CE imaging through injection of a contrast agent. However, the risk of injection of gadolinium-based contrast agents is increasing recently. Region segmentation that marks cancer regions in medical images plays a key role in CAD systems, and deep neural network models for synthesizing new images are also being studied. In this study, we propose a model that simultaneously learns the generation of T1CE images and segmentation of cancer regions. The performance of the proposed model is evaluated using similarity measurements including mean square error and peak signal-to-noise ratio, and shows average result values of 21 and 39 dB.

• Journal of the Korean Society of Earth Science Education
• /
• v.5 no.3
• /
• pp.235-244
• /
• 2012

This study examined the learning tendency & social characteristics and the science camp participation attitude of the global science talented and the science gifted c. The survey was carried out on children who are taking part in Global Bridge project group and in Science Education Institute for the gifted S National University of education. The results of this study were as follows. First, the science gifted children was more superior to the global science talented, when it comes to task commitment and fluency, creativity, enthusiasm for learning on the learning tendency. Second, the science gifted children have much more sociality than the global science talented in the sense of social characteristics. The global science talented showed lack of interpersonal relationship & confidence for human relationship. Third, both parties were positive in terms of attitude which participating science camp. It was proved that science camp made a positive affect on both groups in several senses such as improving awareness & attitude of science activity and enhancing sociality. As a result, unlike ordinary program for the science gifted children, one for the global science talented in global bridge project is highly demanded that it should be considered the characteristics of the multi-cultural students. Moreover, it might be considered that educational circumstance would be needed, under which it is able to stimulates students' scientific curiosity throughout launching science hands-on program, such as systemized science camp etc.

• Smart Structures and Systems
• /
• v.6 no.3
• /
• pp.335-347
• /
• 2010

Recent advances in hardware and instrumentation technology have allowed the possibility of deploying very large sensor arrays on structures. Exploiting the huge amount of data that can result in order to perform vibration-based structural health monitoring (SHM) is not a trivial task and requires research into a number of specific problems. In terms of pressing problems of interest, this paper discusses: the design and optimisation of appropriate sensor networks, efficient data reduction techniques, efficient and automated feature extraction methods, reliable methods to deal with environmental and operational variability, efficient training of machine learning techniques and multi-scale approaches for dealing with very local damage. The paper is a result of the ESF-S3T Eurocores project "Smart Sensing For Structural Health Monitoring" (S3HM) in which a consortium of academic partners from across Europe are attempting to address issues in the design of automated vibration-based SHM systems for structures.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.4 no.3
• /
• pp.327-336
• /
• 2004

In distributed autonomous mobile robot system, each robot (predator or prey) must behave by itself according to its states and environments, and if necessary, must cooperate with other robots in order to carry out a given task. Therefore it is essential that each robot have both learning and evolution ability to adapt to dynamic environment. This paper proposes a pursuing system utilizing the artificial life concept where virtual robots emulate social behaviors of animals and insects and realize their group behaviors. Each robot contains sensors to perceive other robots in several directions and decides its behavior based on the information obtained by the sensors. In this paper, a neural network is used for behavior decision controller. The input of the neural network is decided by the existence of other robots and the distance to the other robots. The output determines the directions in which the robot moves. The connection weight values of this neural network are encoded as genes, and the fitness individuals are determined using a genetic algorithm. Here, the fitness values imply how much group behaviors fit adequately to the goal and can express group behaviors. The validity of the system is verified through simulation. Besides, in this paper, we could have observed the robots' emergent behaviors during simulation.