• 제목/요약/키워드: Learning Processing

검색결과 3,605건 처리시간 0.029초

A Study on Image Labeling Technique for Deep-Learning-Based Multinational Tanks Detection Model

  • Kim, Taehoon;Lim, Dongkyun
    • International Journal of Internet, Broadcasting and Communication
    • /
    • 제14권4호
    • /
    • pp.58-63
    • /
    • 2022
  • Recently, the improvement of computational processing ability due to the rapid development of computing technology has greatly advanced the field of artificial intelligence, and research to apply it in various domains is active. In particular, in the national defense field, attention is paid to intelligent recognition among machine learning techniques, and efforts are being made to develop object identification and monitoring systems using artificial intelligence. To this end, various image processing technologies and object identification algorithms are applied to create a model that can identify friendly and enemy weapon systems and personnel in real-time. In this paper, we conducted image processing and object identification focused on tanks among various weapon systems. We initially conducted processing the tanks' image using a convolutional neural network, a deep learning technique. The feature map was examined and the important characteristics of the tanks crucial for learning were derived. Then, using YOLOv5 Network, a CNN-based object detection network, a model trained by labeling the entire tank and a model trained by labeling only the turret of the tank were created and the results were compared. The model and labeling technique we proposed in this paper can more accurately identify the type of tank and contribute to the intelligent recognition system to be developed in the future.

Cascaded-Hop For DeepFake Videos Detection

  • Zhang, Dengyong;Wu, Pengjie;Li, Feng;Zhu, Wenjie;Sheng, Victor S.
    • KSII Transactions on Internet and Information Systems (TIIS)
    • /
    • 제16권5호
    • /
    • pp.1671-1686
    • /
    • 2022
  • Face manipulation tools represented by Deepfake have threatened the security of people's biological identity information. Particularly, manipulation tools with deep learning technology have brought great challenges to Deepfake detection. There are many solutions for Deepfake detection based on traditional machine learning and advanced deep learning. However, those solutions of detectors almost have problems of poor performance when evaluated on different quality datasets. In this paper, for the sake of making high-quality Deepfake datasets, we provide a preprocessing method based on the image pixel matrix feature to eliminate similar images and the residual channel attention network (RCAN) to resize the scale of images. Significantly, we also describe a Deepfake detector named Cascaded-Hop which is based on the PixelHop++ system and the successive subspace learning (SSL) model. By feeding the preprocessed datasets, Cascaded-Hop achieves a good classification result on different manipulation types and multiple quality datasets. According to the experiment on FaceForensics++ and Celeb-DF, the AUC (area under curve) results of our proposed methods are comparable to the state-of-the-art models.

작업 종속 및 위치기반 선형학습효과를 갖는 2-에이전트 단일기계 스케줄링 (Two-Agent Single-Machine Scheduling with Linear Job-Dependent Position-Based Learning Effects)

  • 최진영
    • 산업경영시스템학회지
    • /
    • 제38권3호
    • /
    • pp.169-180
    • /
    • 2015
  • Recently, scheduling problems with position-dependent processing times have received considerable attention in the literature, where the processing times of jobs are dependent on the processing sequences. However, they did not consider cases in which each processed job has different learning or aging ratios. This means that the actual processing time for a job can be determined not only by the processing sequence, but also by the learning/aging ratio, which can reflect the degree of processing difficulties in subsequent jobs. Motivated by these remarks, in this paper, we consider a two-agent single-machine scheduling problem with linear job-dependent position-based learning effects, where two agents compete to use a common single machine and each job has a different learning ratio. Specifically, we take into account two different objective functions for two agents: one agent minimizes the total weighted completion time, and the other restricts the makespan to less than an upper bound. After formally defining the problem by developing a mixed integer non-linear programming formulation, we devise a branch-and-bound (B&B) algorithm to give optimal solutions by developing four dominance properties based on a pairwise interchange comparison and four properties regarding the feasibility of a considered sequence. We suggest a lower bound to speed up the search procedure in the B&B algorithm by fathoming any non-prominent nodes. As this problem is at least NP-hard, we suggest efficient genetic algorithms using different methods to generate the initial population and two crossover operations. Computational results show that the proposed algorithms are efficient to obtain near-optimal solutions.

대용량 악성코드의 특징 추출 가속화를 위한 분산 처리 시스템 설계 및 구현 (Distributed Processing System Design and Implementation for Feature Extraction from Large-Scale Malicious Code)

  • 이현종;어성율;황두성
    • 정보처리학회논문지:컴퓨터 및 통신 시스템
    • /
    • 제8권2호
    • /
    • pp.35-40
    • /
    • 2019
  • 기존 악성코드 탐지는 다형성 또는 난독화 기법이 적용된 변종 악성코드 탐지에 취약하다. 기계학습 알고리즘은 악성코드에 내재된 패턴을 학습시켜 유사 행위 탐지가 가능해 기존 탐지 방법을 대체할 수 있다. 시간에 따라 변화하는 악성코드 패턴을 학습시키기 위해 지속적으로 데이터를 수집해야한다. 그러나 대용량 악성코드 파일의 저장 및 처리 과정은 높은 공간과 시간 복잡도가 수반된다. 이 논문에서는 공간 복잡도를 완화하고 처리 시간을 가속화하기 위해 HDFS 기반 분산 처리 시스템을 설계한다. 분산 처리 시스템을 이용해 2-gram 특징과 필터링 기준에 따른 API 특징 2개, APICFG 특징을 추출하고 앙상블 학습 모델의 일반화 성능을 비교했다. 실험 결과로 특징 추출의 시간 복잡도는 컴퓨터 한 대의 처리 시간과 비교했을 때 약 3.75배 속도가 개선되었으며, 공간 복잡도는 약 5배의 효율성을 보였다. 특징 별 분류 성능을 비교했을 때 2-gram 특징이 가장 우수했으나 훈련 데이터 차원이 높아 학습 시간이 오래 소요되었다.

Deep Learning in Genomic and Medical Image Data Analysis: Challenges and Approaches

  • Yu, Ning;Yu, Zeng;Gu, Feng;Li, Tianrui;Tian, Xinmin;Pan, Yi
    • Journal of Information Processing Systems
    • /
    • 제13권2호
    • /
    • pp.204-214
    • /
    • 2017
  • Artificial intelligence, especially deep learning technology, is penetrating the majority of research areas, including the field of bioinformatics. However, deep learning has some limitations, such as the complexity of parameter tuning, architecture design, and so forth. In this study, we analyze these issues and challenges in regards to its applications in bioinformatics, particularly genomic analysis and medical image analytics, and give the corresponding approaches and solutions. Although these solutions are mostly rule of thumb, they can effectively handle the issues connected to training learning machines. As such, we explore the tendency of deep learning technology by examining several directions, such as automation, scalability, individuality, mobility, integration, and intelligence warehousing.

Rate Adaptation with Q-Learning in CSMA/CA Wireless Networks

  • Cho, Soohyun
    • Journal of Information Processing Systems
    • /
    • 제16권5호
    • /
    • pp.1048-1063
    • /
    • 2020
  • In this study, we propose a reinforcement learning agent to control the data transmission rates of nodes in carrier sensing multiple access with collision avoidance (CSMA/CA)-based wireless networks. We design a reinforcement learning (RL) agent, based on Q-learning. The agent learns the environment using the timeout events of packets, which are locally available in data sending nodes. The agent selects actions to control the data transmission rates of nodes that adjust the modulation and coding scheme (MCS) levels of the data packets to utilize the available bandwidth in dynamically changing channel conditions effectively. We use the ns3-gym framework to simulate RL and investigate the effects of the parameters of Q-learning on the performance of the RL agent. The simulation results indicate that the proposed RL agent adequately adjusts the MCS levels according to the changes in the network, and achieves a high throughput comparable to those of the existing data transmission rate adaptation schemes such as Minstrel.

Content Modeling Based on Social Network Community Activity

  • Kim, Kyung-Rog;Moon, Nammee
    • Journal of Information Processing Systems
    • /
    • 제10권2호
    • /
    • pp.271-282
    • /
    • 2014
  • The advancement of knowledge society has enabled the social network community (SNC) to be perceived as another space for learning where individuals produce, share, and apply content in self-directed ways. The content generated within social networks provides information of value for the participants in real time. Thus, this study proposes the social network community activity-based content model (SoACo Model), which takes SNC-based activities and embodies them within learning objects. The SoACo Model consists of content objects, aggregation levels, and information models. Content objects are composed of relationship-building elements, including real-time, changeable activities such as making friends, and participation-activity elements such as "Liking" specific content. Aggregation levels apply one of three granularity levels considering the reusability of elements: activity assets, real-time, changeable learning objects, and content. The SoACo Model is meaningful because it transforms SNC-based activities into learning objects for learning and teaching activities and applies to learning management systems since they organize activities -- such as tweets from Twitter -- depending on the teacher's intention.

귀납법칙 학습과 개체위주 학습의 결합방법 (A Combined Method of Rule Induction Learning and Instance-Based Learning)

  • 이창환
    • 한국정보처리학회논문지
    • /
    • 제4권9호
    • /
    • pp.2299-2308
    • /
    • 1997
  • 대부분의 기계학습 방법들은 특정한 방법을 중심으로 연구되어 왔다. 하지만 두 가지 이상의 기계학습방법을 효과적으로 통합할 수 있는 방법에 대한 요구가 증가하며, 이에 따라 본 논문은 귀납법칙 (rule induction) 방법과 개체위주 학습방법 (instance-based learning)을 통합하는 시스템의 개발을 제시한다. 귀납법칙 단계에서는 엔트로피 함수의 일종인 Hellinger 변량을 사용하여 귀납법칙을 자동 생성하는 방법을 보이고, 개체위주 학습방법에서는 기존의 알고리즘의 단점을 보완한 새로운 개체위주 학습방법을 제시한다. 개발된 시스템은 여러 종류의 데이터에 의해 실험되었으며 다른 기계학습 방법과 비교되었다.

  • PDF

물리치료학에서의 문제중심학습(Problem Based Learning) (Problem Based Learning in Physical Therapy)

  • 이경희;김철용;김성학
    • 대한물리치료과학회지
    • /
    • 제9권4호
    • /
    • pp.141-153
    • /
    • 2002
  • Problem based learning(PBL) is one of the learning strategies from the constructivism. It is a learning centered students. The tutors are facillitators as activators, helpers and cooperators not organizer in the classrooms. PBL makes that students learn creativity, independence, reasoning skits, communication and collaboration for problem solving. As the PBL process, students get the problems that are in real situation, discussed with others for brain storming, self directed study and revisited to the situation. They think critically and apply to the real situation. When students are to be physical therapists, they are easy to adopt their job and efficient to manage well. But inspite of a lot of advantages to them, there are much conflict to use as the learning strategies. Students perceived one of best learning method that they have experienced, but there are stress, burden, anxiety, timeless to prepare, lack of information and so on. PBL is effective to learning health oriented subjects, problem solving, even a lot preparation and processing for learning. It is reduced the differences between theories in colleges and practices in the fields. In processing of PBL, students get more many skills than the conventional learning. As trying many times to the classrooms, we can fixed to PBL with mistakes and conflict for better the development of the teaching and learning.

  • PDF

랜덤 변환에 대한 컨볼루션 뉴럴 네트워크를 이용한 특징 추출 (Feature Extraction Using Convolutional Neural Networks for Random Translation)

  • 진태석
    • 한국산업융합학회 논문집
    • /
    • 제23권3호
    • /
    • pp.515-521
    • /
    • 2020
  • Deep learning methods have been effectively used to provide great improvement in various research fields such as machine learning, image processing and computer vision. One of the most frequently used deep learning methods in image processing is the convolutional neural networks. Compared to the traditional artificial neural networks, convolutional neural networks do not use the predefined kernels, but instead they learn data specific kernels. This property makes them to be used as feature extractors as well. In this study, we compared the quality of CNN features for traditional texture feature extraction methods. Experimental results demonstrate the superiority of the CNN features. Additionally, the recognition process and result of a pioneering CNN on MNIST database are presented.