• Journal of Information Processing Systems
• /
• 제17권2호
• /
• pp.411-425
• /
• 2021

In this paper, a vehicle recognition algorithm based on deep convolutional neural network and compression dictionary is proposed. Firstly, the network structure of fine vehicle recognition based on convolutional neural network is introduced. Then, a vehicle recognition system based on multi-scale pyramid convolutional neural network is constructed. The contribution of different networks to the recognition results is adjusted by the adaptive fusion method that adjusts the network according to the recognition accuracy of a single network. The proportion of output in the network output of the entire multiscale network. Then, the compressed dictionary learning and the data dimension reduction are carried out using the effective block structure method combined with very sparse random projection matrix, which solves the computational complexity caused by high-dimensional features and shortens the dictionary learning time. Finally, the sparse representation classification method is used to realize vehicle type recognition. The experimental results show that the detection effect of the proposed algorithm is stable in sunny, cloudy and rainy weather, and it has strong adaptability to typical application scenarios such as occlusion and blurring, with an average recognition rate of more than 95%.

• KNOM Review
• /
• 제24권2호
• /
• pp.24-34
• /
• 2021

엣지 컴퓨팅을 사용하는 서비스 공급업체는 높은 수준의 서비스를 제공한다. 이에 따라 다양하고 중요한 정보들이 단말 장치에 저장되면서 탐지하기 더욱 어려운 최신 사이버 공격의 핵심 목표가 됐다. 보안을 위해 침입 탐지시스템과 같은 보안 시스템이 자주 활용되지만, 기존의 침입 탐지 시스템은 탐지 정확도가 낮은 문제점이 존재한다. 따라서 본 논문에서는 엣지 컴퓨팅에서 단말 장치의 더욱 정확한 침입 탐지를 위한 기계 학습 모델을 제안한다. 제안하는 모델은 희소성 제약을 사용하여 입력 데이터의 중요한 특징 벡터들을 추출하는 stacked sparse autoencoder (SSAE)와 convolutional neural network (CNN)를 결합한 하이브리드 모델이다. 최적의 모델을 찾기 위해 SSAE의 희소성 계수를 조절하면서 모델의 성능을 비교 및 분석했다. 그 결과 희소성 계수가 일 때 96.9%로 가장 높은 정확도를 보여주었다. 따라서 모델이 중요한 특징들만 학습할 경우 더 높은 성능을 얻을 수 있었다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권8호
• /
• pp.3873-3887
• /
• 2018

In this paper, we propose a novel method to recover images corrupted by impulse noise. The proposed method uses two stages: noise detection and filtering. In the first stage, we use pixel values, rank-ordered logarithmic difference values, and median values to train a neural-network-based impulse noise detector. After training, we apply the network to detect noisy pixels in images. In the next stage, we use group-based weighted couple sparse representation to filter the noisy pixels. During this second stage, conventional methods generally use only clean pixels to recover corrupted pixels, which can yield unsuccessful dictionary learning if the noise density is high and the number of useful clean pixels is inadequate. Therefore, we use reconstructed pixels to balance the deficiency. Experimental results show that the proposed noise detector has better performance than the conventional noise detectors. Also, with the information of noisy pixel location, the proposed impulse-noise removal method performs better than the conventional methods, through the recovered images resulting in better quality.

• 디지털산업정보학회논문지
• /
• 제10권4호
• /
• pp.11-18
• /
• 2014

Recently, ICT technology has been evolving towards an M2M (Machine to Machine) environment that allows communication between machine and machine from the communication between person and person, and now the IoT (Internet of Things) technology that connects all things without human intervention is receiving great attention. In such a network environment, the communication network between object and object as well as between person and person, and person and object is available which leads to the sharing of information between all objects, which is the essential technical element for us to move forward to the information service society of the era of future ubiquitous computing. On this paper, the protocol related to DTMNs in a Sparse M2M environment was applied and the improved routing protocol was applied by using the azimuth and density of the moving node in order to support a more efficient network environment to deliver the message between nodes in an M2M environment. This paper intends to verify the continuity of the study related to efficient routing protocols to provide an efficient network environment in the IoT and IoE (Internet of Everything) environment which is as of recently in the spotlight.

• 한국군사과학기술학회지
• /
• 제2권1호
• /
• pp.159-169
• /
• 1999

최근 전산망의 트래픽을 제어하여 해킹방지를 위해 방화벽을 구축한다. 방화벽의 보안 서비스는 인증, 접근통제, 기밀성, 무결성 그리고 감사기록 이다. 사용자는 방화벽에 인증을 위하여 토큰을 사용한다. 토큰은 작은 배터리를 내장하므로 전력 용량이 한정된다. 본 논문은 TCP/IP를 이용하는 전산망의 해킹방지를 위한 경제적인 방화벽 토큰 설계 방법을 제안한다. 공개키 암호 시스템의 주요 연산이며, 토큰 전력 소모의 대부분을 차지하는 지수연산에 Sparse 소수를 이용한 고속 처리 방법을 제안한다. 제안한 방법은 지수연산에서 모듈러 연산 량을 감소시킴으로 토큰의 배터리 용량 또는 CPU 가격을 낮출 수 있다.

• 전자공학회논문지S
• /
• 제35S권3호
• /
• pp.1-8
• /
• 1998

In this paper, we propose a Neural Call Admission Control (CAC) method using a Sparse Distributed Memory(SDM). CAC is a key technology of TM network traffic control. It should be adaptable to the rapid and various changes of the ATM network environment. conventional approach to the ATM CAC requires network analysis in all cases. So, the optimal implementation is said to be very difficult. Therefore, neural approach have recently been employed. However, it does not mett the adaptability requirements. because it requires additional learning data tables and learning phase during CAC operation. We have proposed a neural network CAC method based on SDM that is more actural than conventioal approach to apply it to CAC. We compared it with previous neural network CAC method. It provides CAC with good adaptability to manage changes. Experimenatal results show that it has rapid adaptability and stability without additional learning table or learning phase.

• 대한임베디드공학회논문지
• /
• 제16권5호
• /
• pp.179-186
• /
• 2021

For stereo matching based on deep learning, the design of network structure is crucial to the calculation of matching cost, and the time-consuming problem of convolutional neural network in image processing also needs to be solved urgently. In this paper, a method of stereo matching using sparse loss volume in parallax dimension is proposed. A sparse 3D loss volume is constructed by using a wide step length translation of the right view feature map, which reduces the video memory and computing resources required by the 3D convolution module by several times. In order to improve the accuracy of the algorithm, the nonlinear up-sampling of the matching loss in the parallax dimension is carried out by using the method of multi-category output, and the training model is combined with two kinds of loss functions. Compared with the benchmark algorithm, the proposed algorithm not only improves the accuracy but also shortens the running time by about 30%.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권6호
• /
• pp.2806-2825
• /
• 2018

Unsupervised learning has shown good performance on image, video and audio classification tasks, and much progress has been made so far. It studies how systems can learn to represent particular input patterns in a way that reflects the statistical structure of the overall collection of input patterns. Many promising deep learning systems are commonly trained by the greedy layerwise unsupervised learning manner. The performance of these deep learning architectures benefits from the unsupervised learning ability to disentangling the abstractions and picking out the useful features. However, the existing unsupervised learning algorithms are often difficult to train partly because of the requirement of extensive hyperparameters. The tuning of these hyperparameters is a laborious task that requires expert knowledge, rules of thumb or extensive search. In this paper, we propose a simple and effective unsupervised feature learning algorithm for image classification, which exploits an explicit optimizing way for population and lifetime sparsity. Firstly, a sparse target matrix is built by the competitive rules. Then, the sparse features are optimized by means of minimizing the Euclidean norm ($L_2$) error between the sparse target and the competitive layer outputs. Finally, a classifier is trained using the obtained sparse features. Experimental results show that the proposed method achieves good performance for image classification, and provides discriminative features that generalize well.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제8권4호
• /
• pp.1223-1236
• /
• 2014

With the extension of wireless technology, vehicular ad hoc networks provide important services for the dissemination of general data and emergency warnings. However, since, the vehicle topology frequently changes from a dense to a sparse network depending on the speed of the moving vehicles and the time of day, vehicular ad hoc networks require a protocol that can facilitate the efficient and reliable dissemination of emergency messages in a highly mobile environment under dense or intermittent vehicular connectivity. Therefore, this paper proposes a new vehicular broadcast protocol, called BL-CAST, that can operate effectively in both dense and sparse network scenarios. As a low overhead multi-hop broadcast protocol, BL-CAST does not rely on the periodic exchange of beacons for updating location information. Instead, the location information of a vehicle is included in a broadcast message to identify the last rebroadcasting vehicle in an intermittently connected network. Simulation results show that BL-CAST outperforms the DV-CAST protocol in terms of the end-to-end delay, message delivery ratio and network overhead.

• 한국방사선학회논문지
• /
• 제16권3호
• /
• pp.295-302
• /
• 2022

희박뷰 전산화단층촬영(computed tomography; CT) 영상화 기술은 피폭 방사선량을 감소시킬 수 있을 뿐만 아니라 획득한 투영상의 균일성을 유지하고 잡음을 감소시킬 수 있는 장점이 있다. 하지만 재구성 영상 내 인공물 발생으로 인하여 화질 및 피사체 구조가 왜곡되는 단점이 있다. 본 연구에서는 희박뷰 CT 영상의 인공물 감소를 위해 wavelet 변환과 잔차 학습(residual learning)을 적용한 콘볼루션 신경망(convolutional neural network; CNN) 기반 영상화 모델을 개발하고, 개발한 모델을 통한 희박뷰 CT 영상의 인공물 감소 정도를 정량적으로 분석하였다. CNN은 wavelet 변환 층, 콘볼루션 층 및 역 wavelet 변환 층으로 구성하였으며, 희박뷰 CT 영상과 잔차 영상을 각각 입출력 영상으로 설정하여 영상화 모델 학습을 진행하였다. 영상화 모델 학습을 위해 평균제곱오차(mean squared error; MSE)를 손실함수로, Adam 함수를 최적화 함수로 사용하였다. 학습된 모델을 통해 입력 희박뷰 CT 영상에 대한 예측 잔차 영상을 획득하고, 두 영상간의 감산을 통해 최종 결과 영상을 획득하였다. 또한 최종 결과 영상에 대한 시각적 특성, 최대신호대잡음비(peak signal-to- noise ratio; PSNR) 및 구조적유사성지수(structural similarity; SSIM)를 측정하였다. 연구결과 본 연구에서 개발한 영상화 모델을 통해 희박뷰 CT 영상의 인공물이 효과적으로 제거되며, 공간분해능이 향상되는 결과를 확인하였다. 또한 wavelet 변환과 잔차 학습을 미적용한 영상화 모델에 비해 본 연구에서 개발한 영상화 모델은 결과 영상의 PSNR 및 SSIM을 각각 8.18% 및 19.71% 향상시킬 수 있음을 확인하였다. 따라서 본 연구에서 개발한 영상화 모델을 이용하여 희박뷰 CT 영상의 인공물 제거는 물론 공간분해능 향상 및 정량적 정확도 향상 효과를 획득할 수 있다.