A5101862232- Image Retrieval and Classification Techniques
- Advanced Image and Video Retrieval Techniques
- Service-Oriented Architecture and Web Services
- Heavy metals in environment
- Plant Parasitism and Resistance
- Recommender Systems and Techniques
- Mycorrhizal Fungi and Plant Interactions
- Neuroscience and Neural Engineering
- Video Analysis and Summarization
- Cognitive Computing and Networks
- Ferroelectric and Negative Capacitance Devices
- Software Testing and Debugging Techniques
- Data Mining Algorithms and Applications
- Software Reliability and Analysis Research
- Semantic Web and Ontologies
- Microbial Natural Products and Biosynthesis
- Wikis in Education and Collaboration
- Advanced Decision-Making Techniques
- Remote Sensing and LiDAR Applications
- Adversarial Robustness in Machine Learning
- QR Code Applications and Technologies
- Biosensors and Analytical Detection
- Advanced Optical Sensing Technologies
- Mobile Agent-Based Network Management
- Generative Adversarial Networks and Image Synthesis
Collaborative Innovation Center of Advanced Microstructures
2025
Nanjing University
2019-2025
Wuhan University
2008-2021
State Key Laboratory of Information Engineering in Surveying Mapping and Remote Sensing
2020
University of Michigan
2011
AT&T (United States)
2011
Long Island University
2009
Jiangsu Police Officer College
2007
We propose a novel framework for multimodal video indexing and retrieval using shrinkage optimized directed information assessment (SODA) as similarity measure.The (DI) is variant of the classical mutual which attempts to capture direction flow that videos naturally possess.It applied directly empirical probability distributions both audio-visual features over successive frames.We utilize RASTA-PLP audio feature representation SIFT visual representation.We compute joint density functions in...
Traditional online map tiles, widely used on the Internet such as Google Map and Baidu Map, are rendered from vector data. Timely updating tiles data, of which generating is time-consuming, a difficult mission. It shortcut to generate in time remote sensing images, can be acquired timely without However, this mission challenging or even impossible. Inspired by image-to-image translation (img2img) techniques based generative adversarial networks (GAN), we proposed semi-supervised Generation...
Abstract Learning rate scheduling (LRS) is a critical factor influencing the performance of neural networks by accelerating convergence learning algorithms and enhancing generalization capabilities. The escalating computational demands in artificial intelligence (AI) necessitate advanced hardware solutions capable supporting network training with LRS. This not only requires linear symmetric analog programming capabilities but also precise adjustment channel conductance to achieve tunable...
Screen-camera communication using dynamic barcode streaming has emerged as a convenient and secure method for short-range, impromptu device-to-device communication. Conventional systems adopt rule-based approach to recognize the color stream in receiver, which is empirical, inflexible, lacks self-adaptiveness. In this paper, we propose novel solution framework recognition basing on machine learning techniques. By including number of training frames into build classification model, proposed...
Bug triage is a process where bugs are assigned to developers.In large open source projects such as Mozilla and Eclipse, bug time-consuming because numerous submitted everyday.To improve triage, many studies have proposed automatic approaches recommend proper developers for resolving bugs.These based on machine learning algorithms, which treat like text classification.Although they effective, the accuracy of them can be further improved.Our goal propose method not only has good performance...
Considering the success of generative adversarial networks (GANs) for image-to-image translation, researchers have attempted to translate remote sensing images (RSIs) maps (rs2map) through GAN cartography. However, these studies involved limited scales, which hinders multi-scale map creation. By extending their method, RSIs can be trivially translated (multi-scale rs2map translation) scale-wise models trained certain scales (parallel strategy). this strategy has two theoretical limitations....
Abstract A new ring‐fused streptovaricin analogue, named ansavaricin J, was unprecedently isolated from the culture of genetically modified strains ΔstvP5 which derived Streptomyces spectabilis CCTCC M2017417. Its structure elucidated via comprehensive spectroscopic analyses, including 1D‐ and 2D‐NMR tests, HR‐ESI‐MS data analysis. Notably, J E represent only two reported examples heterocyclic streptovaricins thus far, however, it showed insignificant antibacterial activities against...
Knowledge service such as the subject cognitive mapping plays an important role in effective online learning, skill training and resource recommendation. However, challenges of on automatic construction maintenance its objectivity seriously restrict popularity. Accordingly, this paper proposes association semantic link enabled technique to automatically construct mapping. The proposed employs information resources mining method build main tasks involve extraction key terms, representation...
Along with the development of earth observation technology, large amounts geospatial information are accessible. There also a lot data and services which shared on Internet. However they vary in formats stored at various organizations leading to problems discovery, interoperability usability. The Open Geospatial Consortium (OGC) has developed standard service called catalogue order overcome this problem. goal is support wide range users discovering relevant geographic from heterogeneous...
Recently, induced by incorporating the ubiquitous data collected AI application, there are more and demand for transferring to cloud servers due restriction of computing resources devices. To transfer image efficiently, neural network (NN) codec can be a wise choice, which result in higher ratio compression with similar quality compared conventional methods. However, when NN is employed communication system, it likely that disturbed channel distortions. In this paper, we innovatively propose...