信息工程最新文献

Background

Face image animation generates a synthetic human face video that harmoniously integrates the identity derived from the source image and facial motion obtained from the driving video. This technology could be beneficial in multiple medical fields, such as diagnosis and privacy protection. Previous studies on face animation often relied on a single source image to generate an output video. With a significant pose difference between the source image and the driving frame, the quality of the generated video is likely to be suboptimal because the source image may not provide sufficient features for the warped feature map.

Methods

In this study, we propose a novel face-animation scheme based on multiple sources and perspective alignment to address these issues. We first introduce a multiple-source sampling and selection module to screen the optimal source image set from the provided driving video. We then propose an inter-frame interpolation and alignment module to further eliminate the misalignment between the selected source image and the driving frame.

Conclusions

The proposed method exhibits superior performance in terms of objective metrics and visual quality in large-angle animation scenes compared to other state-of-the-art face animation methods. It indicates the effectiveness of the proposed method in addressing the distortion issues in large-angle animation.

Deep learning has been extensively applied to medical image segmentation, resulting in significant advancements in the field of deep neural networks for medical image segmentation since the notable success of U-Net in 2015. However, the application of deep learning models to ocular medical image segmentation poses unique challenges, especially compared to other body parts, due to the complexity, small size, and blurriness of such images, coupled with the scarcity of data. This article aims to provide a comprehensive review of medical image segmentation from two perspectives: the development of deep network structures and the application of segmentation in ocular imaging. Initially, the article introduces an overview of medical imaging, data processing, and performance evaluation metrics. Subsequently, it analyzes recent developments in U-Net-based network structures. Finally, for the segmentation of ocular medical images, the application of deep learning is reviewed and categorized by the type of ocular tissue.

Understanding the interplay between investor sentiment and cryptocurrency returns has become a critical area of research. Indeed, this study aims to uncover the role of Google investor sentiment on cryptocurrency returns (including Bitcoin, Litecoin, Ethereum, and Tether), especially during the 2017–18 bubble (January 01, 2017, to December 31, 2018) and the COVID-19 pandemic (January 01, 2020, to March 15, 2022). To achieve this, we use two techniques: quantile causality and wavelet coherence. First, the quantile causality test revealed that investors’ optimistic sentiments have notably higher cryptocurrency returns, whereas pessimistic sentiments have significantly opposite effects. Moreover, the wavelet coherence analysis shows that co-movement between investor sentiment and Tether cannot be considered significant. This result supports the role of Tether as a stablecoin in portfolio diversification strategies. In fact, the findings will help investors improve the accuracy of cryptocurrency return forecasts in times of stressful events and pave the way for enhanced decision-making utility.

The implementation of blockchain technology in integrated IoT networks for constructing scalable Intelligent Transportation Systems (ITSs) in India has the potential to revolutionize the way we approach transportation. By leveraging the power of IoT and blockchain, we can create a highly secure, transparent, and efficient system that can transform the way we move people and goods. India, one of the world’s most populous countries, has a highly congested and inefficient transportation system that often leads to delays, accidents, and waste of time and resources. The integration of IoT and blockchain can help address these issues by enabling real-time monitoring, tracking, and optimization of traffic flows, thereby reducing congestion, improving safety, and increasing the overall efficiency of the transportation system. This paper explores the potential of blockchain technology in the context of integrated IoT networks for constructing scalable ITS systems in India. The methodology followed is to develop a proof-of-concept blockchain-based application for ITS, implement the blockchain solution into the existing ITS infrastructure, and ensure proper integration and compatibility with other systems. Conduct thorough research and maintenance to ensure the reliability and sustainability of such blockchain-based systems. This research discusses the various benefits and challenges of this approach and the various applications of this technology in the transportation sector, including the green sustainability concept. The results find various ways in which such implementations of blockchain and IoT-Machine Learning (IoT-ML) can revolutionize transportation systems.

To address the challenges of low credibility, difficult data sharing, and regulatory supervision issues involving electronic evidence storage in the judicial preservation process, this paper proposes a blockchain-based judicial evidence preservation scheme. The scheme utilizes the characteristics of blockchain’s immutability to achieve credible forensics of electronic evidence on the chain and employs the decentralized storage of the interplanetary file system for secure and efficient off-chain storage. Simultaneously, it resolves the problem of declining throughput due to limited block capacity. Additionally, it leverages smart contract technology to encompass major aspects of the judicial process, including user case registration, authority management, judicial evidence uploading and downloading, case data sharing, partial disclosure of case information, and regulatory review. Simulation experiments demonstrate that the scheme significantly improves throughput and stability. Performance tests indicate that the transfer speed of the interplanetary file system can meet the data-sharing needs of judicial organizations.

Background

Deep convolutional neural networks have garnered considerable attention in numerous machine learning applications, particularly in visual recognition tasks such as image and video analyses. There is a growing interest in applying this technology to diverse applications in medical image analysis. Automated three-dimensional Breast Ultrasound is a vital tool for detecting breast cancer, and computer-assisted diagnosis software, developed based on deep learning, can effectively assist radiologists in diagnosis. However, the network model is prone to overfitting during training, owing to challenges such as insufficient training data. This study attempts to solve the problem caused by small datasets and improve model detection performance.

Methods

We propose a breast cancer detection framework based on deep learning (a transfer learning method based on cross-organ cancer detection) and a contrastive learning method based on breast imaging reporting and data systems (BI-RADS).

Results

When using cross organ transfer learning and BIRADS based contrastive learning, the average sensitivity of the model increased by a maximum of 16.05%.

Conclusion

Our experiments have demonstrated that the parameters and experiences of cross-organ cancer detection can be mutually referenced, and contrastive learning method based on BI-RADS can improve the detection performance of the model.

Engine data management is of great significance for ensuring data security and sharing, as well as facilitating multi-party collaborative learning. Traditional data management approaches often involve decentralized data storage that is vulnerable to tampering, making it challenging to conduct multi-party collaborative learning under privacy protection conditions and fully leverage the value of data. Moreover, data with compromised integrity can lead to incorrect results if used for model training. Therefore, this paper aims to break down data sharing barriers and fully utilize decentralized data for multi-party collaborative learning under privacy protection conditions. We propose a trustworthy engine data management method based on blockchain technology to ensure data immutability and non-repudiation. To address the issue of limited data samples for some users resulting in poor model performance, we introduce swarm learning techniques based on centralized machine learning and design a trustworthy data management method for swarm learning, achieving trustworthy regulation of the entire process. We conduct research on engine models under swarm learning based on the NASA open dataset, effectively organizing decentralized data samples for collaborative training while ensuring data privacy and fully leveraging the value of data.