Hierarchical Information Fusion Research Articles

Metal surface crack depth laser acoustic emission detection method based on multivariate feature adaptive extraction and cross-modal interaction fusion

Surface crack depth detection of metal structures is of great significance to ensure the safe of equipment. However, the commonly used detection methods currently suffer from sensitivity and imaging dimension limitations, which make it difficult to ensure accuracy and feature extraction. To this end, a laser acoustic emission crack depth detection method that combines modal decomposition and deep learning is proposed. The new method is more sensitive to elastic wave changes that detect the depth of microscopic cracks and shows higher accuracy than traditional detection methods. At the same time, frequency-adaptive feature mode decomposition of multivariate signals and color multimodal feature fusion are used to solve the difficulty of time–frequency domain feature mining in current research. The crack depth detection network CDDNet is constructed using an attention-based fusion backbone, a novel cross-modal interaction fusion, and a hierarchical information fusion strategy. Its transfer learning achieves a crack detection accuracy level of 0.05 mm and outperforms other models in terms of accuracy, reaching 93.6 %. This paper shows that the proposed transfer learning laser acoustic emission technology and supporting data processing methods can effectively establish a connection between the experimental and simulation datasets, solving the current challenges of crack depth detection in terms of micro-precision and data processing.

Application of Wireless Sensor Network Data Fusion Technology in Mechanical Fault Diagnosis

In order to solve the problem that a large number of vibration signals cannot be transmitted in real time in the application of wireless sensor networks (WSNs) in mechanical fault diagnosis, a mechanical fault diagnosis method based on multilevel and hierarchical information fusion of WSNs was proposed. In this method, the cluster tree network structure is used to expand the coverage of network monitoring, and WSNs information fusion is divided into three levels: data-level fusion, feature-level fusion, and decision-level fusion. The terminal node performs data-level fusion on the original vibration information to extract feature information; the cluster-head node performs feature-level fusion on the feature information to obtain pattern recognition results; and the gateway node performs decision-level fusion on the recognition results to evaluate the running status of mechanical equipment. The results show that the slight damage fault of the bearing inner ring can be accurately diagnosed by decision-level fusion based on four groups of probability distribution functions. According to the statistics of 30 test results, the fault recognition rate is 83.3%. The method can be applied to mechanical fault diagnosis effectively.

Mind the Gap: Cross-Lingual Information Retrieval with Hierarchical Knowledge Enhancement

Cross-Lingual Information Retrieval (CLIR) aims to rank the documents written in a language different from the user’s query. The intrinsic gap between different languages is an essential challenge for CLIR. In this paper, we introduce the multilingual knowledge graph (KG) to the CLIR task due to the sufficient information of entities in multiple languages. It is regarded as a “silver bullet” to simultaneously perform explicit alignment between queries and documents and also broaden the representations of queries. And we propose a model named CLIR with HIerarchical Knowledge Enhancement (HIKE) for our task. The proposed model encodes the textual information in queries, documents and the KG with multilingual BERT, and incorporates the KG information in the query-document matching process with a hierarchical information fusion mechanism. Particularly, HIKE first integrates the entities and their neighborhood in KG into query representations with a knowledge-level fusion, then combines the knowledge from both source and target languages to further mitigate the linguistic gap with a language-level fusion. Finally, experimental results demonstrate that HIKE achieves substantial improvements over state-of-the-art competitors.

Feature pyramid-based graph convolutional neural network for graph classification

As an important task of Graph Neural Networks (GNN), graph classification has received increasing attention, as it can be widely used in numerous fields, such as protein prediction and community prediction. The GNN models for graph classification usually require aggregating the structure and feature information of an input graph into a hidden representation vector. Such a technique can be generally referred as graph pooling, which intends to maximize the removal of noise while minimize the loss of important features. Therefore, there exists a trade-off between the noise reduction and information maximization that needs to be balanced for better performance. However, the existing pooling-based GNNs for graph classification depends on the manual setting of the graph-pooling degree, making it difficult to balance the trade-off. To this end, we propose a Feature Pyramid-based Graph Convolutional Neural network for Graph Classification (FPGCN-GC), which constructs multi-scale hierarchical information fusion to reduce information loss, and achieves an adaptive feature fusion through a learnable weighted residual connection and self-attention mechanism. The superior performance of the proposed method is verified on multiple graph classification datasets, thus illustrating the effectiveness and superiority of FPGCN-GC. In addition, we visualize FPGCN-GC by T-SNE to further analyze the underlying reason for its effectiveness.

Research on Simulation Experiment of Underwater Cluster Multi-source Information Fusion

Intelligent cluster information fusion technology is one of the key technologies to solve systematic cooperative operations in underwater manned and unmanned cooperative operations clusters. In this study, aiming at the multi-node hybrid multi-level hierarchical information fusion architecture and global optimization association algorithm based on cluster formation detection, the test scheme and evaluation method of multi-level information fusion are proposed, and the cluster information fusion simulation test system including sonar system, simulation deduction and evaluation system and multi-source information fusion system is constructed. For the first time, the system comprehensively verified the ability of multi-platform from orientation to positioning in the grouping of fusion algorithm, and proved the correctness and effectiveness of the fusion algorithm, which could greatly improve the accuracy of orientation and distance, and formed a highly credible simulation test evaluation ability. The system is extensible at the same time, and can also be applied to other simulation experiments.

Health management and pattern analysis of daily living activities of people with dementia using in-home sensors and machine learning techniques.

The number of people diagnosed with dementia is expected to rise in the coming years. Given that there is currently no definite cure for dementia and the cost of care for this condition soars dramatically, slowing the decline and maintaining independent living are important goals for supporting people with dementia. This paper discusses a study that is called Technology Integrated Health Management (TIHM). TIHM is a technology assisted monitoring system that uses Internet of Things (IoT) enabled solutions for continuous monitoring of people with dementia in their own homes. We have developed machine learning algorithms to analyse the correlation between environmental data collected by IoT technologies in TIHM in order to monitor and facilitate the physical well-being of people with dementia. The algorithms are developed with different temporal granularity to process the data for long-term and short-term analysis. We extract higher-level activity patterns which are then used to detect any change in patients’ routines. We have also developed a hierarchical information fusion approach for detecting agitation, irritability and aggression. We have conducted evaluations using sensory data collected from homes of people with dementia. The proposed techniques are able to recognise agitation and unusual patterns with an accuracy of up to 80%.

Hierarchical information fusion for decision making in craniofacial superimposition

Craniofacial superimposition is one of the most important skeleton-based identification methods. The process studies the possible correspondence between a found skull and a candidate (missing person) through the superimposition of the former over a variable number of images of the face of the latter. Within craniofacial superimposition we identified three different stages, namely: (1) image acquisition-processing and landmark location; (2) skull-face overlay; and (3) decision making. While we have already proposed and validated an automatic skull-face overlay technique in previous works, the final identification stage, decision making, is still performed manually by the expert. This consists of the determination of the degree of support for the assertion that the skull and the ante-mortem image belong to the same person. This decision is made through the analysis of several criteria assessing the skull-face anatomical correspondence based on the resulting skull-face overlay. In this contribution, we present a hierarchical framework for information fusion to support the anthropologist expert in the decision making stage. The main goal is the automation of this stage based on the use of several skull-face anatomical criteria combined at different levels by means of fuzzy aggregation functions. We have implemented two different experiments for our framework. The first aims to obtain the most suitable aggregation functions for the system and the second validates the proposed framework as an identification system. We tested the framework with a dataset of 33 positive and 411 negative identification instances. The present proposal is the first automatic craniofacial superimposition decision support system evaluated in an objective and statistically meaningful way.

Hierarchical Information Fusion for Global Displacement Estimation in Microsensor Motion Capture

This paper presents a novel hierarchical information fusion algorithm to obtain human global displacement for different gait patterns, including walking, running, and hopping based on seven body-worn inertial and magnetic measurement units. In the first-level sensor fusion, the orientation for each segment is achieved by a complementary Kalman filter (CKF) which compensates for the orientation error of the inertial navigation system solution through its error state vector. For each foot segment, the displacement is also estimated by the CKF, and zero velocity update is included for the drift reduction in foot displacement estimation. Based on the segment orientations and left/right foot locations, two global displacement estimates can be acquired from left/right lower limb separately using a linked biomechanical model. In the second-level geometric fusion, another Kalman filter is deployed to compensate for the difference between the two estimates from the sensor fusion and get more accurate overall global displacement estimation. The updated global displacement will be transmitted to left/right foot based on the human lower biomechanical model to restrict the drifts in both feet displacements. The experimental results have shown that our proposed method can accurately estimate human locomotion for the three different gait patterns with regard to the optical motion tracker.

Application of the fault diagnosis strategy based on hierarchical information fusion in motors fault diagnosis

This paper has analyzed merits and demerits of both neural network technique and of the information fusion methods based on the D-S (dempster-shafer evidence) Theory as well as their complementarity, proposed the hierarchical information fusion fault diagnosis strategy by combining the neural network technique and the fused decision diagnosis based on D-S Theory, and established a corresponding functional model. Thus, we can not only solve a series of problems caused by rapid growth in size and complexity of neural network structure with diagnosis parameters increasing, but also can provide effective method for basic probability assignment in D-S Theory. The application of the strategy to diagnosing faults of motor bearings has proved that this method is of fairly high accuracy and reliability in fault diagnosis.

家庭生活様態の知的計測システム 低次階層情報融合機構の構成

家庭生活様態の知的計測システム 低次階層情報融合機構の構成

家庭生活様態の知的計測システム 高次階層情報融合機構の知識と推論

家庭生活様態の知的計測システム 高次階層情報融合機構の知識と推論

家庭生活様態の知的計測システム 低次階層の情報融合機構

家庭生活様態の知的計測システム 低次階層の情報融合機構