• 한국멀티미디어학회논문지
• /
• 제23권12호
• /
• pp.1540-1551
• /
• 2020

Aiming at the problem that the existing human behavior recognition algorithm cannot fully utilize the multi-level spatio-temporal information of the network, a human behavior recognition algorithm based on a dense three-dimensional residual network is proposed. First, the proposed algorithm uses a dense block of three-dimensional residuals as the basic module of the network. The module extracts the hierarchical features of human behavior through densely connected convolutional layers; Secondly, the local feature aggregation adaptive method is used to learn the local dense features of human behavior; Then, the residual connection module is applied to promote the flow of feature information and reduced the difficulty of training; Finally, the multi-layer local feature extraction of the network is realized by cascading multiple three-dimensional residual dense blocks, and use the global feature aggregation adaptive method to learn the features of all network layers to realize human behavior recognition. A large number of experimental results on benchmark datasets KTH show that the recognition rate (top-l accuracy) of the proposed algorithm reaches 93.52%. Compared with the three-dimensional convolutional neural network (C3D) algorithm, it has improved by 3.93 percentage points. The proposed algorithm framework has good robustness and transfer learning ability, and can effectively handle a variety of video behavior recognition tasks.

• International Journal of Computer Science & Network Security
• /
• 제23권7호
• /
• pp.39-48
• /
• 2023

Breast cancer screening makes extensive utilization of mammography. Even so, there has been a lot of debate with regards to this application's starting age as well as screening interval. The deep learning technique of transfer learning is employed for transferring the knowledge learnt from the source tasks to the target tasks. For the resolution of real-world problems, deep neural networks have demonstrated superior performance in comparison with the standard machine learning algorithms. The architecture of the deep neural networks has to be defined by taking into account the problem domain knowledge. Normally, this technique will consume a lot of time as well as computational resources. This work evaluated the efficacy of the deep learning neural network like Visual Geometry Group Network (VGG Net) Residual Network (Res Net), as well as inception network for classifying the mammograms. This work proposed optimization of ResNet with Teaching Learning Based Optimization (TLBO) algorithm's in order to predict breast cancers by means of mammogram images. The proposed TLBO-ResNet, an optimized ResNet with faster convergence ability when compared with other evolutionary methods for mammogram classification.

• Structural Engineering and Mechanics
• /
• 제81권2호
• /
• pp.195-203
• /
• 2022

This paper presents the quantification and location damage detection of plane and space truss structures in a two-phase method to reduce the computations efforts significantly. In the first phase, a proposed damage indicator based on the residual force vector concept is used to get the suspected damaged members. In the second phase, using damage quantification as a variable, a teaching-learning based optimization algorithm (TLBO) is used to obtain the damage quantification value of the suspected members obtained in the first phase. TLBO is a relatively modern algorithm that has proved distinguished in solving optimization problems. For more verification of TLBO effeciency, the classical particle swarm optimization (PSO) is used in the second phase to make a comparison between TLBO and PSO algorithms. As it is clear, the first phase reduces the search space in the second phase, leading to considerable reduction in computations efforts. The method is applied on three examples, including plane and space trusses. Results have proved the capability of the proposed method to precisely detect the quantification and location of damage easily with low computational efforts, and the efficiency of TLBO in comparison to the classical PSO.

• Nuclear Engineering and Technology
• /
• 제56권4호
• /
• pp.1347-1356
• /
• 2024

Measuring the residual stress in the components in nuclear power plants is crucial to their safety evaluation. The instrumented indentation technique is a minimally invasive approach that can be conveniently used to determine the residual stress in structural materials in service. Because the indentation behavior of a structure with residual stresses is closely related to the elastic-plastic behavior of the indented material, an accurate understanding of the elastic-plastic behavior of the material is essential for evaluation of the residual stresses in the structures. However, due to the analytical problems associated with solving the elastic-plastic behavior, empirical equations with limited applicability have been used. In the present study, the impact of the non-equibiaxial residual stress state on indentation behavior was investigated using finite element analysis. In addition, a new nonequibiaxial residual-stress prediction methodology is proposed using a convolutional neural network, and the performance was validated. A more accurate residual-stress measurement will be possible by applying the proposed residual-stress prediction methodology in the future.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권12호
• /
• pp.3991-4007
• /
• 2022

In the public healthcare, a computational system that can automatically and efficiently detect and classify falls from a video sequence has significant potential. With the advancement of deep learning, which can extract temporal and spatial information, has become more widespread. However, traditional 3D CNNs that usually adopt shallow networks cannot obtain higher recognition accuracy than deeper networks. Additionally, some experiences of neural network show that the problem of gradient explosions occurs with increasing the network layers. As a result, an enhanced three-dimensional ResNet-based method for fall detection (3D-ERes-FD) is proposed to directly extract spatio-temporal features to address these issues. In our method, a 50-layer 3D residual network is used to deepen the network for improving fall recognition accuracy. Furthermore, enhanced residual units with four convolutional layers are developed to efficiently reduce the number of parameters and increase the depth of the network. According to the experimental results, the proposed method outperformed several state-of-the-art methods.

• 한국컴퓨터정보학회논문지
• /
• 제25권6호
• /
• pp.35-47
• /
• 2020

박스 오피스 예측은 영화 이해관계자들에게 중요하다. 따라서 정확한 박스 오피스 예측과 이에 영향을 미치는 주요 변수를 선별하는 것이 필요하다. 본 논문은 영화의 박스 오피스 예측 정확도 향상을 위해 다변량 시계열 데이터 분류와 주요 변수 선택 방법을 제안한다. 연구 방법으로 한국 영화 일별 데이터를 KOBIS와 NAVER에서 수집하였고, 랜덤 포레스트(Random Forest) 방법으로 주요 변수를 선별하였으며, 딥러닝(Deep Learning)으로 다변량 시계열을 예측하였다. 한국의 스크린 쿼터제(Screen Quota) 기준, 딥러닝을 이용하여 영화 개봉 73일째 흥행 예측 정확도를 주요 변수와 전체 변수로 비교하고 통계적으로 유의한지 검정하였다. 딥러닝 모델은 다층 퍼셉트론(Multi-Layer Perceptron), 완전 합성곱 신경망(Fully Convolutional Neural Networks), 잔차 네트워크(Residual Network)로 실험하였다. 결과적으로 주요 변수를 잔차 네트워크에 사용했을 때 예측 정확도가 약 93%로 가장 높았다.

• 대한조선학회논문집
• /
• 제57권6호
• /
• pp.312-321
• /
• 2020

In this study, machine learning techniques were applied to predict the residual resistance coefficient (Cr) of low-speed full ships. The used machine learning methods are Ridge regression, support vector regression, random forest, neural network and their ensemble model. 19 hull form variables were used as input variables for machine learning methods. The hull form variables and Cr data obtained from 139 hull forms of KRISO database were used in analysis. 80 % of the total data were used as training models and the rest as validation. Some non-linear models showed the overfitted results and the ensemble model showed better results than others.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권6호
• /
• pp.1743-1758
• /
• 2023

Automatic modulation classification is a critical algorithm for non-cooperative communication systems. This paper addresses the challenging problem of closed-set and open-set signal modulation classification in complex channels. We propose a novel approach that incorporates a self-learning filter and center-loss in Deep Residual Shrinking Networks (DRSN) for closed-set modulation classification, and the Opendistance method for open-set modulation classification. Our approach achieves better performance than existing methods in both closed-set and open-set recognition. In closed-set recognition, the self-learning filter and center-loss combination improves recognition performance, with a maximum accuracy of over 92.18%. In open-set recognition, the use of a self-learning filter and center-loss provide an effective feature vector for open-set recognition, and the Opendistance method outperforms SoftMax and OpenMax in F1 scores and mean average accuracy under high openness. Overall, our proposed approach demonstrates promising results for automatic modulation classification, providing better performance in non-cooperative communication systems.

• 전기전자학회논문지
• /
• 제27권3호
• /
• pp.340-344
• /
• 2023

본 논문에서는 날씨 변화에 따른 실외 LED 전광판의 시인성 확보를 위한 딥러닝 구조 개발에 관한 연구를 제안한다. 제안하는 기법은 영상장치를 이용한 딥러닝을 사용하여 날씨 변화에 따른 LED 휘도를 자동 조절함으로써 실외 LED 전광판의 시인성을 확보한다. 날씨 변화에 따른 LED 휘도를 자동 조절하기 위하여, 먼저 평면화된 배경 부분 이미지 데이터에 대한 전처리 과정을 거친 후, 합성곱 네트워크를 이용하여 학습시켜 날씨에 대한 분류를 진행할 수 있는 딥러닝 모델을 만들어낸다. 적용된 딥러닝 네트워크는 Residual learning 함수를 사용하여 입력값과 출력값의 차이를 줄임으로써 초기의 입력값의 특징을 가지고 가면서 학습하도록 유도한다. 다음에 날씨를 인식하여 날씨 변화에 따라 실외 LED 전광판의 휘도를 조절하는 제어기를 사용하여 주변 환경이 밝아지면 휘도가 높아지도록 변경하여 선명하게 보이도록 한다. 또한, 주변 환경이 어두워지면 빛의 산란에 의해 시인성이 떨어지기 때문에 전광판의 휘도가 내려가도록 하여 선명하게 보이도록 한다. 본 논문에서 제안하는 방법을 적용하여 LED 전광판의 날씨 변화에 따른 휘도 측정의 공인 측정 실험 결과는, 날씨 변화에 따라 실외 LED 전광판의 시인성이 확보됨을 확인하였다.

• 사물인터넷융복합논문지
• /
• 제9권1호
• /
• pp.131-138
• /
• 2023

본 논문은 RAPGAN(Relativistic Average Patch GAN)과 RRDB(Residual in Residual Dense Block)을 이용한 Image-to-Image 변환의 성능 개선에 관한 것이다. 본 논문은 Image-to-Image 변환의 일종인 기존의 pix2pix의 결점을 보완하기 위해 세 가지 측면의 기술적 개선을 통한 성능 향상을 도모함에 그 목적이 있다. 첫째, 기존의 pix2pix 생성자와 달리 입력 이미지를 인코딩하는 부분에서 RRDB를 이용함으로써 더욱 더 깊은 학습을 가능하게 한다. 둘째, RAPGAN 기반의 손실함수를 사용해 원본 이미지가 생성된 이미지에 비해 얼마나 진짜 같은지를 예측하기 때문에 이 두 이미지가 모두 적대적 생성 학습에 영향을 미치게 된다. 마지막으로, 생성자를 사전학습시켜 판별자가 조기에 학습되는 것을 억제하도록 조치한다. 제안된 방법에 따르면, FID 측면에서 기존의 pix2pix보다 평균 13% 이상의 우수한 이미지를 생성할 수 있었다.