A5087821333- Image Retrieval and Classification Techniques
- Advanced Image and Video Retrieval Techniques
- Video Analysis and Summarization
- Electrical Contact Performance and Analysis
- Tribology and Wear Analysis
- Coding theory and cryptography
- Retinal Imaging and Analysis
- Digital Imaging for Blood Diseases
- Remote-Sensing Image Classification
- Cooperative Communication and Network Coding
- Evaluation Methods in Various Fields
- Metal and Thin Film Mechanics
- Text and Document Classification Technologies
- Image Processing Techniques and Applications
- Mechanical stress and fatigue analysis
- Welding Techniques and Residual Stresses
- Advanced materials and composites
- Smart Agriculture and AI
- Lubricants and Their Additives
- graph theory and CDMA systems
- Hydrogen embrittlement and corrosion behaviors in metals
- Aluminum Alloys Composites Properties
- Data Management and Algorithms
- Corrosion Behavior and Inhibition
Southwest Petroleum University
2020-2025
Central South University
2010
Multi-instance learning (MIL) is more general and challenging than traditional supervised in that labels are given at the bag level. The popular feature mapping approaches convert each into an instance new space. However, most of them hardly maintain distinguishability bags, MIL model does not support self-reinforcement. In this article, we propose multi-instance ensemble with discriminative bags (ELDB) algorithm two techniques. selection technique obtains a set (dBagSet) according to parts....
Real-world applications always contain complex data objects with multiple semantics. The machine learning tasks generalized from these can be formulated as multi-instance multi-label (MIML) problems. Each object is represented by a bag of instances and associated labels simultaneously. Extensive studies on the MIML have been conducted during past few years. However, many them do not consider local label correlations, which critical in scenario. To address this problem, we propose based...
A bimetallic composite pipe butt joint was prepared by automatic gas tungsten arc welding (GTAW). The microstructure and pitting properties of the were investigated. results show that is columnar crystal, dendrite, equiaxial crystal. Mo, Nb, Cr segregate in form Laves NbC precipitates interdendritic. heat-affected zone (HAZ) becomes preferred region for due to recrystallization behavior large continuous Laves, this demonstrated both immersion corrosion electrochemical experiments. To better...
Abstract The composite material sintered by copper and graphite belongs to 'pseudo-alloy', which reduces its self-lubricating performance a certain extent. Because WS 2 have similar characteristics, the synergistic lubrication mechanism of copper-based graphite-WS (Cu-Gr-WS ) has become hot research topic in recent years. In this paper, friction contact surface characteristics Cu-Gr-WS composites under different loads were studied. results show that coefficient first decreases then increases...
Let m, k be positive integers with m=2k and k≥3. C(u, ν) is a class of cyclic codes length 2m-1 whose parity-check polynomial mu(x)mν(x), where mu(x) mν(x) are the minimal polynomials α-u α-ν over GF(2). For case $(u,\nu)=(1,\frac{1}{3}(2^m-1))$, weight distributions binary was determined in 2017. This paper determines for (u, ν)=(3, 2k-1+1). The application these secret sharing also considered.