← back to search

Xianbin Wen

Ministry of Education of the People's Republic of China · CN
🔎 Find collaborators in Media Technology · Computer Vision and Pattern Recognition →
Search 5.9M scientists by topic, h-index, country & funding — free.
Area of research
Media Technology · Computer Vision and Pattern Recognition
Research interest
Research topics from publications: A Remote-Sensing Scene-Image Classification Method Based on Deep Multiple-Instance Learning with a Residual Dense Attention ConvNet; Deep Tensor Capsule Network; Deep Multi-Instance Learning with Induced Self-Attention for Medical Image Classification; A lightweight and stochastic depth residual attention network for remote sensing scene classification; A multi-scale dense residual correlation network for remote sensing scene classification. Representative work: The spatial distribution of remote-sensing scene images is highly complex in character, so how to extract local key semantic information and discriminative features is the key to making it possible to classify accurately. However, most of the existing convolutional neural network (CNN) models tend to have global feature representations and lose the shallow features. In addition, when the network is too deep, gradient disappearance and overfitting tend to occur. To solve these problems, a lightweight, multi-instance CNN model for remote sensing scene classification is proposed in this paper: MILRDA. In the instance extraction and classifier part, more discriminative features are extracted by Capsule network is a promising model in computer vision. It has achieved excellent results on simple datasets such as MNIST, but the performance deteriorates as data becomes complicated. In order to address this issue, we propose a deep capsule network in this paper. To deepen the capsule network, we present a new tensor capsule based routing algorithm and the corresponding convolution operation. Compared to vector capsules, tensor capsules can capture more instance-level information. Together, the relevant convolution operation is beneficial for reducing the amount of parameters in the routing process. Furthermore, we propose a dropout mechanism for vectors and tensors in order to alleviate
h-index
citations
0
works
0
NIH funding
primary concept
email

Recent publications

A multi-scale dense residual correlation network for remote sensing scene classification
Scientific Reports 2024cited by 14position: lastdoi
A lightweight and stochastic depth residual attention network for remote sensing scene classification
IET Image Processing 2023cited by 16position: lastdoi
A Remote-Sensing Scene-Image Classification Method Based on Deep Multiple-Instance Learning with a Residual Dense Attention ConvNet
Remote Sensing 2022cited by 29position: lastdoi
Deep Tensor Capsule Network
IEEE Access 2020cited by 26position: lastdoi
Deep Multi-Instance Learning with Induced Self-Attention for Medical Image Classification
2020cited by 21position: lastdoi

Grants

No grants ingested yet.

Frequent collaborators

Haixia Xu · Ministry of Education of the People's Republic of China5 papers (2020–2024)Liming Yuan · Ministry of Education of the People's Republic of China3 papers (2020–2023)Xinyu Wang · South China Agricultural University2 papers (2022–2023)Liming Yuan · Ministry of Education of the People's Republic of China2 papers (2020–2024)Furong Shi · Ministry of Education of the People's Republic of China1 papers (2024–2024)Wei Dai · Tianjin University of Technology1 papers (2024–2024)Wei Dai · Ministry of Education of the People's Republic of China1 papers (2022–2022)Xinyu Wang · Ministry of Education of the People's Republic of China1 papers (2024–2024)Zhenliang Li · Ministry of Education of the People's Republic of China1 papers (2020–2020)Rui Cheng · Ministry of Education of the People's Republic of China1 papers (2020–2020)Kun Sun · Ministry of Education of the People's Republic of China1 papers (2020–2020)
Looking for a research collaborator?
Search millions of scientists by field, institution, impact, and funding status — see their work, find their email, and reach out directly.
Find collaborators in Media Technology · Computer Vision and Pattern Recognition →