A5100368243- Advanced Neural Network Applications
- Advanced Image and Video Retrieval Techniques
- Thermodynamic properties of mixtures
- Phase Equilibria and Thermodynamics
- Chemical Thermodynamics and Molecular Structure
- Image Retrieval and Classification Techniques
- Medical Image Segmentation Techniques
- Quantum Computing Algorithms and Architecture
- Remote-Sensing Image Classification
- Ocean Acidification Effects and Responses
- Quantum Information and Cryptography
- Fluid Dynamics and Turbulent Flows
- Echinoderm biology and ecology
- Visual Attention and Saliency Detection
- Optical Network Technologies
- Photonic Crystal and Fiber Optics
- Smart Grid Energy Management
- Domain Adaptation and Few-Shot Learning
- Video Surveillance and Tracking Methods
- Radiomics and Machine Learning in Medical Imaging
- Water Systems and Optimization
- Advanced Fiber Laser Technologies
- Coral and Marine Ecosystems Studies
- Advanced Photocatalysis Techniques
- Quantum-Dot Cellular Automata
Chengdu University
2021-2024
Ministry of Agriculture and Rural Affairs
2021-2024
Beijing University of Technology
2014-2024
NARI Group (China)
2015-2024
Dalian Ocean University
2020-2023
Shandong Institute of Automation
2023
Ministry of Education of the People's Republic of China
2023
Tokyo University of Technology
2023
Chongqing University
2021-2023
Chongqing Jiaotong University
2023
This paper introduces a convolutional neural network (CNN) semantic re-ranking system to enhance the performance of sketch-based image retrieval (SBIR). Distinguished from existing approaches, proposed can leverage category information brought by CNNs support effective similarity measurement between images. To achieve classification query sketches and high-quality initial results, one CNN model is trained for sketches, another that natural Through training dual models, both images captured...
Traditional surveillance systems for observing water levels are often complex, costly, and time-consuming. In this paper, we developed a low-cost unmanned system consisting of remote measuring stations monitoring center. The uses map-based Web service, as well video cameras, level analyzers, wireless communication routers necessary to display real-time measurements rivers reservoirs on platform. With the aid router, information is transmitted server connected Internet via cellular network....
Safety is a core requirement of the transformer substation where dangerous due to high voltage. It requires detect pedestrians efficiently based on surveillance video near ensure safety pedestrian and device. This paper presents new method detection for Gaussian Mixture Model (GMM) YOLO. We use GMM model background preliminarily, at same time, YOLO, kind convolution neural network (CNN), also applied detection. Through combining two results with different weights, outputs better result. Our...
To improve the performance of sketch-based image retrieval (SBIR) methods, most existing SBIR methods develop brand new methods. In fact, a re-ranking approach, which can refine results is also beneficial. Inspired by this, in this paper, an approach based on multi-clustering proposed. order to make invisible users and adaptive different types datasets, we made it unsupervised method using blind feedback. Distinguished from uses semantic information three images: edge maps, object images...
Strongylocentrotus intermedius is one of the most economically valuable sea urchin species in China and has experienced mass mortality owing to outbreaks bacterial diseases such as black mouth disease. This caused serious economic losses farming industry. To investigate immune response mechanism S. with different tube feet colors Vibrio harveyi infection, we examined feet-colored under V. challenge compared their transcriptome microRNA (miRNA) profiles using RNA-Seq. We obtained 1813...
In this paper, we propose a novel hierarchical generative model, named author-genre-topic model (AGTM), to perform satellite image annotation. Different from the existing author-topic in which each author and topic are associated with multinomial distributions over topics words, AGTM, genre, author, authors, topics, respectively. The bias of distribution authors respect can be rectified by incorporating genres authors. Therefore, classification accuracy documents is improved when information...
ADVERTISEMENT RETURN TO ISSUEPREVArticleNEXTExcess volumes for binary mixtures of n-butyl methyl ether with n-alkanes at 298.15 KLuo Wang, George C. Benson, and Benjamin Y. LuCite this: J. Chem. Eng. Data 1990, 35, 3, 242–244Publication Date (Print):July 1, 1990Publication History Published online1 May 2002Published inissue 1 July 1990https://pubs.acs.org/doi/10.1021/je00061a006https://doi.org/10.1021/je00061a006research-articleACS PublicationsRequest reuse permissionsArticle...
Abstract This article presents a novel color edge detection algorithm based on quaternion convolution. A set of multiscale masks are presented, and the filtered results multiplied at adjacent scales using dot product to enhance structures while diluting noise. Experimental show that proposed method is efficient robust for images. © 2010 Wiley Periodicals, Inc. Int J Imaging Syst Technol, 20, 354–358,
This paper focuses on weakly supervised image understanding, in which the semantic labels are available only at image-level, without specific object or scene location an image. Existing algorithms implicitly assume that image-level error-free, might be too restrictive. In practice, obtained from pretrained predictors easily contaminated. To solve this problem, we propose a novel algorithm for segmentation when noisy during training. More specifically, space is constructed first by encoding...