Ensuring the precise grading of discontinuities is imperative to guarantee the quality of castings and enhance profitability in casting production. Recent grading methods leveraging computer vision are advanced by per...
详细信息
Ensuring the precise grading of discontinuities is imperative to guarantee the quality of castings and enhance profitability in casting production. Recent grading methods leveraging computer vision are advanced by performing a single-label image classification or regression, which loses the intrinsically ordinal relationship. Motivated by this observation, we propose a label smoothing technology for ordinal variables to convert the level of each defect instance into a discrete probability distribution, aiming to model the noise label and ordinal relationship. Furthermore, we design a convolutional neural network framework based on multi-task learning. This framework, by simultaneously learning the level label distribution and regressing the level directly, outperforms a single-task network in terms of overall performance. Finally, we construct a casting gas porosity defect grading dataset. Experimental results on this dataset highlight the significant advantages of our proposed method compared to traditional single-label image classification or regression algorithms.
Achieving pixel-level crack segmentation in complex scenarios is a major challenge, as current methods have difficulty effectively integrating both local features and irregular pixel dependencies. In this paper, we in...
详细信息
Trade policy differences among different countries are important factors affecting international trade cooperation. In this paper, we build an evolutionary game model of international trade in which complex networks p...
详细信息
Trade policy differences among different countries are important factors affecting international trade cooperation. In this paper, we build an evolutionary game model of international trade in which complex networks portray game relationships and trade policy differences are game strategies of players. Compared with the fully coupled game relationship and two-strategies game, the game relationship dynamic adjustment, trade policy differences and stochastic game payoffs in this paper are more in line with the real international trade context. We use mathematical analyses and computer simulation methods to reveal the impact of trade policy differences on the game payoffs of countries and the overall international trade networks. The results show that the reduce of the tolerance "t" of fair-trade is difficult to eliminate the opportunism trade policy and lead to a decline in the game payoffs of all countries and the global trading network. Dynamic adjustment of trade relations can not only improve the game payoffs of all countries and the trade network, but also offset the negative impact of the decline of the tolerance "t". In terms of long-term, only the win-win trade policy can maximize the payoffs of the global trade network and achieve a balanced distribution of trade benefits.
In the digital age, where data security is paramount, managing the security of library repositories poses significant challenges due to the increasing threat of unauthorized access and data breaches. Face recognition ...
详细信息
This study constructs or strengthens the network security of the emergency command system from all aspects based on different requirements such as secure physical environment, secure communication network, secure area...
详细信息
Effective tool wear monitoring is of great importance for machining process. With existing deep learning-based methods, the end-to-end model is often combined with sensor data to predict the state of the tool wear. Th...
详细信息
Effective tool wear monitoring is of great importance for machining process. With existing deep learning-based methods, the end-to-end model is often combined with sensor data to predict the state of the tool wear. These methods lack an effective mechanism to identify and highlight the important tool wear information in different sensor monitoring data, which limits the monitoring accuracy of the model. Therefore, a new deep prediction framework, called a multi-input parallel convolutional attention network (MPCAN), is proposed in this paper and used for intelligent tool wear monitoring. A multi-input parallel convolutional (MPC) network structure is developed to extract multi-scale features from monitoring data of various sensors. Efficient channel attention (ECA) is used to assign different attention weights to the tool wear information. A bidirectional long short-term memory (Bi-LSTM) is used to obtain the time series features related to tool wear. Finally, the predicted tool wear value is output through the fully connected network. An experimental analysis has been undertaken to illustrate, the effectiveness and superiority of the proposed method in improving the accuracy of the tool wear.
Single-carrier frequency domain contention (S-FDC) is an efficient wireless contention mechanism based on orthogonal frequency-division multiplexing (OFDM). In each round of S-FDC, each node randomly selects and signa...
详细信息
An essential diagnostic technique for brain tumors, Magnetic Resonance Imaging (MRI) enables early detection and improved patient outcomes. However, manual interpretation of MRI scans can be time-consuming and subject...
详细信息
In response to the current problems of low data transmission efficiency, high latency, and insufficient stability in computer communications, this paper studies a comprehensive optimized data transmission algorithm, w...
详细信息
In the present day, as the number of connected devices increases exponentially, an organization faces challenges in monitoring, managing and detecting vulnerable devices. Moreover, by the introduction of Bring Your Ow...
详细信息
ISBN:
(纸本)9783031800191;9783031800207
In the present day, as the number of connected devices increases exponentially, an organization faces challenges in monitoring, managing and detecting vulnerable devices. Moreover, by the introduction of Bring Your Own Device (BYOD) schemes in an organization, the device management and vulnerability assessment has become one of the key security requirements. The zero trust system uses both the user and device identity to provide the access to the requested resources. Hence, it is important to understand the security posture of the device before providing the requested resource access. 'InTrust' is an asset monitoring, analysis and vulnerability assessment system for zero trust network access system. InTrust automatically keeps track of the list of assets, operating system, services & their vulnerabilities and brings out the security posture. These inputs can be used to derive the dynamic trust score for a resource access in the Zero Trust network. As InTrust uses a non-intrusive and privacy-aware approach for data collection, it can cater to all enterprise and e-governance networks where security and transparency are critical to the organization.
暂无评论