• International Journal of Naval Architecture and Ocean Engineering
• /
• 제13권1호
• /
• pp.321-339
• /
• 2021

Response Surface Method (RSM) has been widely used for flammable cloud size prediction as it can reduce computational intensity for further Explosion Risk Analysis (ERA) especially during the early design phase of offshore platforms. However, RSM encounters the overfitting problem under very limited simulations. In order to overcome the disadvantage of RSM, Bayesian Regularization Artificial Neural (BRANN)-based model has been recently developed and its robustness and efficiency have been widely verified. However, for ERA during the early design phase, there seems to be room to further reduce the computational intensity while ensuring the model's acceptable accuracy. This study aims to develop an integrated method, namely the combination of Center Composite Design (CCD) method with Bayesian Regularization Artificial Neural Network (BRANN), for flammable cloud size prediction. A case study with constant and transient leakages is conducted to illustrate the feasibility and advantage of this hybrid method. Additionally, the performance of CCD-BRANN is compared with that of RSM. It is concluded that the newly developed hybrid method is more robust and computational efficient for ERAs during early design phase.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권11호
• /
• pp.4105-4121
• /
• 2021

Recently, malware classification based on Deep Neural Networks (DNN) has gained significant attention due to the rise in popularity of artificial intelligence (AI). DNN-based malware classifiers are a novel solution to combat never-before-seen malware families because this approach is able to classify malwares based on structural characteristics rather than requiring particular signatures like traditional malware classifiers. However, these DNN-based classifiers have been found to lack robustness against malwares that are carefully crafted to evade detection. These specially crafted pieces of malware are referred to as adversarial examples. We consider a clever adversary who has a thorough knowledge of DNN-based malware classifiers and will exploit it to generate a crafty malware to fool DNN-based classifiers. In this paper, we propose a DNN-based malware classifier that becomes resilient to these kinds of attacks by exploiting Generative Adversarial Network (GAN) based data augmentation. The experimental results show that the proposed scheme classifies malware, including AEs, with a false positive rate (FPR) of 3.0% and a balanced accuracy of 70.16%. These are respective 26.1% and 18.5% enhancements when compared to a traditional DNN-based classifier that does not exploit GAN.

• Smart Structures and Systems
• /
• 제28권6호
• /
• pp.811-825
• /
• 2021

High-voltage isolating switches play a paramount role in ensuring the safety of power supply systems. However, their exposure to outdoor environmental conditions may cause serious physical defects, which may result in great risk to power supply systems and society. Image processing-based methods have been used for anomaly detection. However, their accuracy is affected by numerous uncertainties due to manually extracted features, which makes the anomaly detection of isolating switches still challenging. In this paper, a vision-based anomaly detection method for isolating switches, which uses the rotational angle of the switch system for more accurate and direct anomaly detection with the help of deep learning (DL) and image processing methods (Single Shot Multibox Detector (SSD), improved frame differencing method, and Hough transform), is proposed. The SSD is a deep learning method for object classification and localization. In addition, an improved frame differencing method is introduced for better feature extraction and a hough transform method is adopted for rotational angle calculation. A number of experiments are conducted for anomaly detection of single and multiple switches using video frames. The results of the experiments demonstrate that the SSD outperforms the You-Only-Look-Once network. The effectiveness and robustness of the proposed method have been proven under various conditions, such as different illumination and camera locations using 96 videos from the experiments.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권8호
• /
• pp.2627-2647
• /
• 2022

The easy scalability and low cost of range-free localization algorithms have led to their wide attention and application in node localization of wireless sensor networks. However, the existing range-free localization algorithms still have problems, such as large cumulative errors and poor localization performance. To solve these problems, an incremental strategy-based residual regression network is proposed for node localization in wireless sensor networks. The algorithm predicts the coordinates of the nodes to be solved by building a deep learning model and fine-tunes the prediction results by regression based on the intersection of the communication range between the predicted and real coordinates and the loss function, which improves the localization performance of the algorithm. Moreover, a correction scheme is proposed to correct the augmented data in the incremental strategy, which reduces the cumulative error generated during the algorithm localization. The analysis through simulation experiments demonstrates that our proposed algorithm has strong robustness and has obvious advantages in localization performance compared with other algorithms.

• Structural Engineering and Mechanics
• /
• 제87권3호
• /
• pp.231-242
• /
• 2023

A novel combinatorial type-2 adaptive neuro-fuzzy inference system (T2-ANFIS) and robust proportional integral derivative (PID) control framework for intelligent vibration mitigation of uncertain structural system is introduced. The fuzzy logic controllers (FLCs), are designed independently of the mathematical model of the system. The type-1 FLCs, have a limited ability to reduce the effect of uncertainty, due to their fuzzy sets with a crisp degree of membership. In real applications, the consequent part of the fuzzy rules is uncertain. The type-2 FLCs, are robust to the fuzzy rules and the process parameters due to the fuzzy degree of membership functions and footprint of uncertainty (FOU). The adaptivity of the proposed method is provided with the optimum tuning of the parameters using the neural network training algorithms. In our approach, the PID control force is obtained using the generalized type-2 neuro-fuzzy in such a way that the stability and robustness of the controller are guaranteed. The robust performance and stability of the presented framework are demonstrated in a numerical study for an eleven-story seismically-excited building structure combined with an active tuned mass damper (ATMD). The results indicate that the introduced type-2 neuro-fuzzy PID control scheme is effective to attenuate plant states in the presence of the structured and unstructured uncertainties, compared to the conventional, type-1 FLC, type-2 FLC, and type-1 neuro-fuzzy PID controllers.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2022년도 학술발표회
• /
• pp.143-143
• /
• 2022

Precipitation plays an essential role in water resources management and disaster prevention. Therefore, the understanding related to spatiotemporal characteristics of rainfall is necessary. Nowadays, highly accurate precipitation is mainly obtained from gauge observation systems. However, the density of gauge stations is a sparse and uneven distribution in mountainous areas. With the proliferation of technology, satellite-based precipitation sources are becoming increasingly common and can provide rainfall information in regions with complex topography. Nevertheless, satellite-based data is that it still remains uncertain. To overcome the above limitation, this study aims to take the strengthens of machine learning to generate a new reanalysis of precipitation data by fusion of multiple satellite precipitation products (SPPs) with gauge observation data. Several machine learning algorithms (i.e., Random Forest, Support Vector Regression, and Artificial Neural Network) have been adopted. To investigate the robustness of the new reanalysis product, observed data were collected to evaluate the accuracy of the products through Kling-Gupta efficiency (KGE), probability of detection (POD), false alarm rate (FAR), and critical success index (CSI). As a result, the new precipitation generated through the machine learning model showed higher accuracy than original satellite rainfall products, and its spatiotemporal variability was better reflected than others. Thus, reanalysis of satellite precipitation product based on machine learning can be useful source input data for hydrological simulations in ungauged river basins.

• International Journal of Advanced Culture Technology
• /
• 제11권4호
• /
• pp.393-405
• /
• 2023

Accurate prediction of wind speed and power is vital for enhancing the efficiency of wind energy systems. Numerous solutions have been implemented to date, demonstrating their potential to improve forecasting. Among these, deep learning is perceived as a revolutionary approach in the field. However, despite their effectiveness, the noise present in the collected data remains a significant challenge. This noise has the potential to diminish the performance of these algorithms, leading to inaccurate predictions. In response to this, this study explores a novel feature engineering approach. This approach involves altering the data input shape in both Convolutional Neural Network-Long Short-Term Memory (CNN-LSTM) and Autoregressive models for various forecasting horizons. The results reveal substantial enhancements in model resilience against noise resulting from step increases in data. The approach could achieve an impressive 83% accuracy in predicting unseen data up to the 24th steps. Furthermore, this method consistently provides high accuracy for short, mid, and long-term forecasts, outperforming the performance of individual models. These findings pave the way for further research on noise reduction strategies at different forecasting horizons through shape-wise feature engineering.

• Current Optics and Photonics
• /
• 제8권3호
• /
• pp.270-281
• /
• 2024

Information-hiding technology is introduced into an optical ghost imaging encryption scheme, which can greatly improve the security of the encryption scheme. However, in the current mainstream research on camouflage ghost imaging encryption, information hiding techniques such as digital watermarking can only hide 1/4 resolution information of a cover image, and most secret images are simple binary images. In this paper, we propose an equal-resolution image-hiding encryption scheme based on deep learning and computational ghost imaging. With the equal-resolution image steganography network based on deep learning (ERIS-Net), we can realize the hiding and extraction of equal-resolution natural images and increase the amount of encrypted information from 25% to 100% when transmitting the same size of secret data. To the best of our knowledge, this paper combines image steganography based on deep learning with optical ghost imaging encryption method for the first time. With deep learning experiments and simulation, the feasibility, security, robustness, and high encryption capacity of this scheme are verified, and a new idea for optical ghost imaging encryption is proposed.

• 한국차세대컴퓨팅학회논문지
• /
• 제15권5호
• /
• pp.64-74
• /
• 2019

비강체 정합은 임상적 필요성은 높으나 계산 복잡도가 높고, 정합의 정확성 및 강건성을 확보하기 어려운 분야이다. 본 논문은 비지도 학습 환경에서 3차원 뇌 자기공명 영상 데이터에 딥러닝 네트워크를 이용한 비강체 정합 기법을 제안한다. 서로 다른 환자의 두 영상을 입력받아 네트워크를 통하여 두 영상 간의 특징 벡터를 생성하고, 변위 벡터장을 만들어 기준 영상에 맞추어 다른 쪽 영상을 변형시킨다. 네트워크는 U-Net 형태를 기반으로 설계하여 정합 시 두 영상의 전역적, 지역적인 차이를 모두 고려한 특징 벡터를 만들 수 있고, 손실함수에 균일화 항을 추가하여 3차원 선형보간법 적용 후에 실제 뇌의 움직임과 유사한 변형 결과를 얻을 수 있다. 본 방법은 비지도 학습을 통해 임의의 두 영상만을 입력으로 받아 단일 패스 변형으로 비강체 정합을 수행한다. 이는 반복적인 최적화 과정을 거치는 비학습 기반의 정합 방법들보다 빠르게 수행할 수 있다. 실험은 50명의 뇌를 촬영한 3차원 자기공명 영상을 가지고 수행하였고, 정합 전·후의 Dice Similarity Coefficient 측정 결과 평균 0.690으로 정합 전과 비교하여 약 16% 정도의 유사도 향상을 확인하였다. 또한, 비학습 기반 방법과 비교하여 유사한 성능을 보여주면서 약 10,000배 정도의 속도 향상을 보여주었다. 제안 기법은 다양한 종류의 의료 영상 데이터의 비강체 정합에 활용이 가능하다.

• 대한전자공학회논문지SD
• /
• 제42권8호
• /
• pp.35-42
• /
• 2005

본 논문에서는 전력소모 감소 및 강건성 (robustness) 향상을 위한 새로운 구조의 플립-플롭을 제안한다. 가변 샘플링 윈도우 플립-플롭(Variable sampling window flip-flop, VSWFF)은 입력 데이터에 따라 샘플링 윈도우의 폭을 변화시켜 강인한 데이터-래치 동작을 제공할 뿐 아니라 더욱 짧은 hold time을 갖는다. 또한, 이 플립-플롭은 입력 스위칭 행위(input switching activity)가 큰 경우에 기존의 저전력 플립-플롭보다 내부 전력소모를 감소시킬 수 있다. 클럭 진폭 감쇄형 가변 샘플링 윈도우 플립-플롭(Clock swing-reduced variable sampling window flip-flop, CSR-VSWFF)은 작은 스윙 폭의 클럭을 사용함으로써 클럭분배망(clock distribution network)의 전력소모를 감소시킬 수 있다. 기존의 클럭 진폭 감쇄형 플립-플롭(Reduced clock swing flip-flop, RCSFF)과 달리, 제안된 플립-플롭은 공급전압만으로 동작하므로 고전압의 발생 및 분배로 인한 설계 상의 비용증가를 제거한다. 시뮬레이션 결과, 기존의 플립-플롭과 비교하여 더욱 좁은 샘플링 윈도우에서도 불변의 지연값(latency) 을 유지하고 전력-지연 곱(power-delay product, PDP)이 개선됨을 확인하였다. 제안된 플립-플롭의 성능을 평가하기 위하여 $0.3\mu m$ CMOS 공정기술을 이용하여 테스트 칩을 설계하였으며, 실험 결과, VSWFF는 입력 스위칭 행위가 최대일 때 전력소모가 감소하며 CSR-YSWFF를 이용하여 설계된 동기 카운터는 부가 고전압의 사용 없이 전력소모가 감소됨을 확인하였다.