• 한국군사과학기술학회지
• /
• 제25권3호
• /
• pp.300-310
• /
• 2022

A vast amount of labeled data is required for deep neural network training. A typical strategy to improve the performance of a neural network given a training data set is to use data augmentation technique. The goal of this work is to offer a novel image augmentation method for improving object detection accuracy. An object in an image is removed, and a similar object from the training data set is placed in its area. An in-painting algorithm fills the space that is eliminated but not filled by a similar object. Our technique shows at most 2.32 percent improvements on mAP in our testing on a military vehicle dataset using the YOLOv4 object detector.

• International Journal of Computer Science & Network Security
• /
• 제23권6호
• /
• pp.140-146
• /
• 2023

A detailed survey, applications and challenges of video encoding-decoding systems is discussed in this paper. A novel architecture has also been set aside for future work in the same direction. The literature reviews span the years 1960 to the present, highlighting the benchmark methods proposed by notable academics in the field of video compression. The timeline used to illustrate the review is divided into three sections. Classical methods, conventional heuristic methods, and current deep learning algorithms are all used for video compression in these categories. The milestone contributions are discussed for each category. The methods are summarized in various tables, along with their benefits and drawbacks. The summary also includes some comments regarding specific approaches. Existing studies' shortcomings are thoroughly described, allowing potential researchers to plot a course for future research. Finally, a closing note is made, as well as future work in the same direction.

• 대한치과교정학회지
• /
• 제51권2호
• /
• pp.77-85
• /
• 2021

Objective: To evaluate the accuracy of a multi-stage convolutional neural network (CNN) model-based automated identification system for posteroanterior (PA) cephalometric landmarks. Methods: The multi-stage CNN model was implemented with a personal computer. A total of 430 PA-cephalograms synthesized from cone-beam computed tomography scans (CBCT-PA) were selected as samples. Twenty-three landmarks used for Tweemac analysis were manually identified on all CBCT-PA images by a single examiner. Intra-examiner reproducibility was confirmed by repeating the identification on 85 randomly selected images, which were subsequently set as test data, with a two-week interval before training. For initial learning stage of the multi-stage CNN model, the data from 345 of 430 CBCT-PA images were used, after which the multi-stage CNN model was tested with previous 85 images. The first manual identification on these 85 images was set as a truth ground. The mean radial error (MRE) and successful detection rate (SDR) were calculated to evaluate the errors in manual identification and artificial intelligence (AI) prediction. Results: The AI showed an average MRE of 2.23 ± 2.02 mm with an SDR of 60.88% for errors of 2 mm or lower. However, in a comparison of the repetitive task, the AI predicted landmarks at the same position, while the MRE for the repeated manual identification was 1.31 ± 0.94 mm. Conclusions: Automated identification for CBCT-synthesized PA cephalometric landmarks did not sufficiently achieve the clinically favorable error range of less than 2 mm. However, AI landmark identification on PA cephalograms showed better consistency than manual identification.

• 한국정보통신학회논문지
• /
• 제25권11호
• /
• pp.1477-1485
• /
• 2021

최근 이미지 초해상도는 정수배율만 가능한 모델에만 집중적으로 연구되고 있다. 하지만 관심 객체 인식, 디스플레이 화질 개선 등 실제 초해상도 기술의 대표 적용 분야에서는 소수 배율을 포함하는 임의배율 확대 필요성이 대두되고 있다. 본 논문에서는 기존 정수배율 모델의 가중치를 활용하여 임의배율을 실행할 수 있는 모델을 제안한다. 이 모델은 정수배율에 의해 우수한 성능을 가진 초해상도 결과를 DCT 스펙트럼 도메인으로 변환하여 임의배율을 위한 공간을 확장한다. DCT 스펙트럼 도메인에 의한 확장으로 인해 발생하는 이미지의 고주파 정보 손실 문제를 줄이기 위해 고주파 스펙트럼 정보를 적절히 복원할 수 있는 모델인 고주파 집중 네트워크를 제안한다. 제안된 네트워크는 고주파 정보를 제대로 생성하기 위해서 RGB 채널간의 상관관계를 학습하는 레이어인 channel attention을 활용하고, 잔차 학습 구조를 통해 모델을 깊게 만들어 성능을 향상시켰다.

• 지질공학
• /
• 제32권2호
• /
• pp.221-239
• /
• 2022

플랜트, 토목 및 건축 사업에서 말뚝 설계 시 어려움을 겪는 주된 요인은 지반 특성의 불확실성이다. 특히 표준관입시험을 통해 구한 N치가 설계 시 주요 입력값이나 짧은 입찰기간과 광범위한 구역에서 다수의 현장시험을 실시하는 것은 실제적으로 어려운 상황이다. 본 연구에서는 인공지능(AI)을 가지고 회귀분석을 적용하여 N치를 예측하는 연구를 수행하였으며, 최소한의 시추자료를 학습시킨 후 표준관입시험을 실시하지 못한 곳에서 N치를 예측하는데 그 목적이 있다. AI의 학습 성능을 높이기 위해서는 빅 데이터가 중요하며, 금회 연구 시 부족한 시추자료를 빅 데이터화 하는데 '원형증강법'을 적용하여 시추반경 2 m까지 가상 N치를 생성시키는 작업을 선행하였다. AI 모델 중 인공신경망, 의사결정 나무, 오토 머신러닝을 각각 적용하였으며 이 중 최적의 모델을 선택하였다. 최적의 모델을 선택하는 방법은 세 가지의 예측된 AI 모델 중 최소 오차값을 가지는 것이다. 이를 위해 폴란드, 인도네시아, 말레이시아에서 수행한 6개 프로젝트를 대상으로 표준관입시험의 실측N치와 AI의 예측N치를 비교하여 타당성 여부를 연구하였고, 연구 결과 AI 예측값에 대한 신뢰도가 높은 것으로 분석되었다. AI 예측값을 가지고 미시추 구간에서 지반특성을 파악 할 수 있었으며 3차원 N치 분포도를 사용하면 최적의 구조물 배치가 가능함을 확인하였다.

• 전자통신동향분석
• /
• 제33권5호
• /
• pp.121-134
• /
• 2018

The Von Neumann based architecture of the modern computer has dominated the computing industry for the past 50 years, sparking the digital revolution and propelling us into today's information age. Recent research focus and market trends have shown significant effort toward the advancement and application of artificial intelligence technologies. Although artificial intelligence has been studied for decades since the Turing machine was first introduced, the field has recently emerged into the spotlight thanks to remarkable milestones such as AlexNet-CNN and Alpha-Go, whose neural-network based deep learning methods have achieved a ground-breaking performance superior to existing recognition, classification, and decision algorithms. Unprecedented results in a wide variety of applications (drones, autonomous driving, robots, stock markets, computer vision, voice, and so on) have signaled the beginning of a golden age for artificial intelligence after 40 years of relative dormancy. Algorithmic research continues to progress at a breath-taking pace as evidenced by the rate of new neural networks being announced. However, traditional Von Neumann based architectures have proven to be inadequate in terms of computation power, and inherently inefficient in their processing of vastly parallel computations, which is a characteristic of deep neural networks. Consequently, global conglomerates such as Intel, Huawei, and Google, as well as large domestic corporations and fabless companies are developing dedicated semiconductor chips customized for artificial intelligence computations. The AI Processor Research Laboratory at ETRI is focusing on the research and development of super low-power AI processor chips. In this article, we present the current trends in computation platform, parallel processing, AI processor, and super-threaded AI processor research being conducted at ETRI.

• International Journal of Computer Science & Network Security
• /
• 제23권11호
• /
• pp.183-189
• /
• 2023

A stroke is a medical disease where a blood vessel in the brain ruptures, causes damage to the brain. If the flow of blood and different nutrients to the brain is intermittent, symptoms may occur. Stroke is other reason for loss of life and widespread disorder. The prevalence of stroke is high in growing countries, with ischemic stroke being the high usual category. Many of the forewarning signs of stroke can be recognized the seriousness of a stroke can be reduced. Most of the earlier stroke detections and prediction models uses image examination tools like CT (Computed Tomography) scan or MRI (Magnetic Resonance Imaging) which are costly and difficult to use for actual-time recognition. Machine learning (ML) is a part of artificial intelligence (AI) that makes software applications to gain the exact accuracy to predict the end results not having to be directly involved to get the work done. In recent times ML algorithms have gained lot of attention due to their accurate results in medical fields. Hence in this work, Stroke disease identification system by using Machine Learning algorithm is presented. The ML algorithm used in this work is Artificial Neural Network (ANN). The result analysis of presented ML algorithm is compared with different ML algorithms. The performance of the presented approach is compared to find the better algorithm for stroke identification.

• 한국컴퓨터정보학회논문지
• /
• 제28권12호
• /
• pp.57-66
• /
• 2023

본 연구에서는 딥러닝 기법과 정서적 AI를 적용하여 사용자의 감정 상태를 추정하고 이를 추천 과정에 반영할 수 있는 추천 시스템에 대한 새로운 연구 프레임워크를 제안한다. 이를 위해 분노, 혐오, 공포, 행복, 슬픔, 놀람, 중립의 7가지 감정을 각각 분류하는 감정분류모델을 구축하고, 이 결과를 추천 과정에 반영할 수 있는 모형을 제안한다. 그러나 일반적인 감정 분류 데이터에서는 각 레이블 간 분포 비율의 차이가 크기 때문에 일반화된 분류 결과를 기대하기 어려울 수 있다. 본 연구에서는 감정 이미지 데이터에서 혐오감 등의 감정 개수가 부족한 경우가 많으므로 데이터 증강을 이용한다. 마지막으로, 이미지 증강을 통해 데이터 기반의 감정 예측 모델을 추천시스템에 반영하는 방법을 제안한다.

• International Journal of Computer Science & Network Security
• /
• 제23권11호
• /
• pp.73-76
• /
• 2023

We introduce DCNN and DRAE appraoches for compression of medical videos, in order to decrease file size and storage requirements, there is an increasing need for medical video compression nowadays. Using a lossy compression technique, a higher compression ratio can be attained, but information will be lost and possible diagnostic mistakes may follow. The requirement to store medical video in lossless format results from this. The aim of utilizing a lossless compression tool is to maximize compression because the traditional lossless compression technique yields a poor compression ratio. The temporal and spatial redundancy seen in video sequences can be successfully utilized by the proposed DCNN and DRAE encoding. This paper describes the lossless encoding mode and shows how a compression ratio greater than 2 (2:1) can be achieved.

• 전력전자학회:학술대회논문집
• /
• 전력전자학회 2000년도 전력전자학술대회 논문집
• /
• pp.620-623
• /
• 2000

A very simple control approach using neural network for the robust position control of a Permanent Magnet Synchronous Motor(PMSM) is presented The linear quadratic controller plus feedforward neural network is employed to obtain the robust PMSM system approximately linearized using field-orientation method for an AC servo. The neural network is trained in on-line phases and this neural network is composed by a fedforward recall and error back-propagation training. Since the total number of nodes are only eight this system can be easily realized by the general microprocessor. During the normal operation the input-output response is sampled and the weighting value is trained multi-times by error back-propagation method at each sample period to accommodate the possible variations in the parameters or load torque. And the state space analysis is performed to obtain the state feedback gains systematically. IN addition the robustness is also obtained without affecting overall system response. This method is realized by a floating-point Digital Singal Processor DS1102 Board (TMS320C31) The basic DSP software is used to write C program which is compiled by using ANSI-C style function prototypes.