• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권12호
• /
• pp.4439-4455
• /
• 2021

Compared with two-stage object detection algorithms, one-stage algorithms provide a better trade-off between real-time performance and accuracy. However, these methods treat the intermediate features equally, which lacks the flexibility to emphasize meaningful information for classification and location. Besides, they ignore the interaction of contextual information from different scales, which is important for medium and small objects detection. To tackle these problems, we propose an image pyramid network based on dual attention mechanism (DAIPNet), which builds an image pyramid to enrich the spatial information while emphasizing multi-scale informative features based on dual attention mechanisms for one-stage object detection. Our framework utilizes a pre-trained backbone as standard detection network, where the designed image pyramid network (IPN) is used as auxiliary network to provide complementary information. Here, the dual attention mechanism is composed of the adaptive feature fusion module (AFFM) and the progressive attention fusion module (PAFM). AFFM is designed to automatically pay attention to the feature maps with different importance from the backbone and auxiliary network, while PAFM is utilized to adaptively learn the channel attentive information in the context transfer process. Furthermore, in the IPN, we build an image pyramid to extract scale-wise features from downsampled images of different scales, where the features are further fused at different states to enrich scale-wise information and learn more comprehensive feature representations. Experimental results are shown on MS COCO dataset. Our proposed detector with a 300 × 300 input achieves superior performance of 32.6% mAP on the MS COCO test-dev compared with state-of-the-art methods.

• Journal of electromagnetic engineering and science
• /
• 제18권3호
• /
• pp.206-211
• /
• 2018

In this paper, we proposed a two-level feature vector fusion technique to improve the performance of target classification. The proposed method combines feature vectors of the early-time region and late-time region in the first-level fusion. In the second-level fusion, we combine the monostatic and bistatic features obtained in the first level. The radar cross section (RCS) of the 3D full-scale model is obtained using the electromagnetic analysis tool FEKO, and then, the feature vector of the target is extracted from it. The feature vector based on the waveform structure is used as the feature vector of the early-time region, while the resonance frequency extracted using the evolutionary programming-based CLEAN algorithm is used as the feature vector of the late-time region. The study results show that the two-level fusion method is better than the one-level fusion method.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권4호
• /
• pp.1234-1257
• /
• 2023

In this paper, we propose a framework for multi-focus image fusion called PATN. In our approach, by aggregating deep features extracted based on the U-type Transformer mechanism and shallow features extracted using the PSA module, we make PATN feed both long-range image texture information and focus on local detail information of the image. Meanwhile, the edge-preserving information value of the fused image is enhanced using a dense residual block containing the Sobel gradient operator, and three loss functions are introduced to retain more source image texture information. PATN is compared with 17 more advanced MFIF methods on three datasets to verify the effectiveness and robustness of PATN.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권9호
• /
• pp.2927-2941
• /
• 2022

The primary task of machine fault diagnosis is to judge whether the current state is normal or damaged, so it is a typical binary classification problem with mutual exclusion. Mutually exclusive events and out-of-domain detection have one thing in common: there are two types of data and no intersection. We proposed a fusion model method to improve the accuracy of machine fault diagnosis, which is based on the mutual exclusivity of events and the commonality of out-of-distribution detection, and finally generalized to all binary classification problems. It is reported that the performance of a convolutional neural network (CNN) will decrease as the recognition type increases, so the variational auto-encoder (VAE) is used as the primary model. Two VAE models are used to train the machine's normal and fault sound data. Two reconstruction probabilities will be obtained during the test. The smaller value is transformed into a correction value of another value according to the mutually exclusive characteristics. Finally, the classification result is obtained according to the fusion algorithm. Filtering normal data features from fault data features is proposed, which shields the interference and makes the fault features more prominent. We confirm that good performance improvements have been achieved in the machine fault detection data set, and the results are better than most mainstream models.

• Journal of Information Processing Systems
• /
• 제20권4호
• /
• pp.535-549
• /
• 2024

Facial expressions (FEs) serve as fundamental components for human emotion assessment and human-computer interaction. Traditional convolutional neural networks tend to overlook valuable information during the FE feature extraction, resulting in suboptimal recognition rates. To address this problem, we propose a deep learning framework that incorporates hierarchical feature fusion, contextual data, and an attention mechanism for precise FE recognition. In our approach, we leveraged an enhanced VGGNet16 as the backbone network and introduced an improved group convolutional channel attention (GCCA) module in each block to emphasize the crucial expression features. A partial decoder was added at the end of the backbone network to facilitate the fusion of multilevel features for a comprehensive feature map. A reverse attention mechanism guides the model to refine details layer-by-layer while introducing contextual information and extracting richer expression features. To enhance feature distinguishability, we employed islanding loss in combination with softmax loss, creating a joint loss function. Using two open datasets, our experimental results demonstrated the effectiveness of our framework. Our framework achieved an average accuracy rate of 74.08% on the FER2013 dataset and 98.66% on the CK+ dataset, outperforming advanced methods in both recognition accuracy and stability.

• Journal of Welding and Joining
• /
• 제14권3호
• /
• pp.93-104
• /
• 1996

This study evaluated relative hot cracking susceptibility of commercial aluminum alloy welds, and then suggested possible mechanisms operated in the weld fusion zone and in the heat affected zone based on the observed cracking morphologies, fractography and microstructural features. The fusion zone solidification cracking was found to be mainly due to a microsegregation of Cu, Si, and Mg in grain boundaries, while liquation cracking in the HAZ was by the incipient melting of the segregated grain boundaries and the consitutional liquation of large aging precipitates and intermetallic compounds in the partially melted zone adjacent to the fusion line which experienced a rapid thermal excursion during welding.

• International Journal of Computer Science & Network Security
• /
• 제23권4호
• /
• pp.39-47
• /
• 2023

This study compares various transformation techniques for multifocus image fusion. Multi-focus image fusion is a procedure of merging multiple images captured at unalike focus distances to produce a single composite image with improved sharpness and clarity. In this research, the purpose is to compare different popular frequency domain approaches for multi-focus image fusion, such as Discrete Wavelet Transforms (DWT), Stationary Wavelet Transforms (SWT), DCT-based Laplacian Pyramid (DCT-LP), Discrete Cosine Harmonic Wavelet Transform (DC-HWT), and Dual-Tree Complex Wavelet Transform (DT-CWT). The objective is to increase the understanding of these transformation techniques and how they can be utilized in conjunction with one another. The analysis will evaluate the 10 most crucial parameters and highlight the unique features of each method. The results will help determine which transformation technique is the best for multi-focus image fusion applications. Based on the visual and statistical analysis, it is suggested that the DCT-LP is the most appropriate technique, but the results also provide valuable insights into choosing the right approach.

• Journal of Information Processing Systems
• /
• 제17권3호
• /
• pp.556-570
• /
• 2021

The existing video expression recognition methods mainly focus on the spatial feature extraction of video expression images, but tend to ignore the dynamic features of video sequences. To solve this problem, a multi-mode convolution neural network method is proposed to effectively improve the performance of facial expression recognition in video. Firstly, OpenFace 2.0 is used to detect face images in video, and two deep convolution neural networks are used to extract spatiotemporal expression features. Furthermore, spatial convolution neural network is used to extract the spatial information features of each static expression image, and the dynamic information feature is extracted from the optical flow information of multiple expression images based on temporal convolution neural network. Then, the spatiotemporal features learned by the two deep convolution neural networks are fused by multiplication. Finally, the fused features are input into support vector machine to realize the facial expression classification. Experimental results show that the recognition accuracy of the proposed method can reach 64.57% and 60.89%, respectively on RML and Baum-ls datasets. It is better than that of other contrast methods.

• 대한원격탐사학회지
• /
• 제21권2호
• /
• pp.145-162
• /
• 2005

SAR 자료의 분류에서 토지 피복 구분 분류 정확도의 향상을 위해 이 논문은 다중 시기 SAR 자료를 이용한 분류에서의 특징 추출과 정보 융합 방법론을 제시하였다. 다중 시기 SAR 센서의 산란 특성을 고려하여 평균 후방 산란계수, 시간적 변이도와 긴밀도를 특징으로서 추출하였다. 이렇게 추출된 특징의 효율적인 응합을 위해 Dempster-Shafer theory of evidence(D-S 이론)와 퍼지 논리를 적용하였다. 특히 D-S 이론의 적용시 특징 기반 mass function 할당을 제안하였고, 퍼지 논리의 적용시 다양한 퍼지 결합 연산자의 결과를 비교하였다. 다중 시기 Radarsat-1 자료에의 적용 결과, 추출된 특징들은 서로 상호 보완적인 정보를 제공할 수 있으며 수계, 논과 도심지를 효율적으로 구분할 수 있었다. 그러나 산림과 밭은 구분이 애매한 경우가 나타났다. 정보 융합 방법론 측면에서, D-S 이론과 퍼지 Max와 Algebraic Sum 연산자를 제외한 다른 퍼지 연산자는 서로 유사한 분류 정확도를 나타내었다.

• 한국공작기계학회논문집
• /
• 제17권1호
• /
• pp.77-85
• /
• 2008

The objective of the study is to construct a sensor fusion system for tool-condition monitoring (TCM) that will lead to a more efficient and economical drill usage. Drill-wear monitoring has an important attribute in the automatic machining processes as it can help preventing the damage of tools and workpieces, and optimizing the drill usage. In this study, we present the architectures of a multi-layer feed-forward neural network with Levenberg-Marquardt training algorithm based on sensor fusion for the monitoring of drill-wear condition. The input features to the neural networks were extracted from AE, vibration and current signals using the wavelet packet transform (WPT) analysis. Training and testing were performed at a moderate range of cutting conditions in the dry drilling of steel plates. The results show good performance in drill- wear monitoring by the proposed method of sensor fusion and neural network analysis.