Dr. Lu Liu Profile

Dr. Lu Liu

at National Univ. of Defense Technology

SPIE Involvement:

Author

Publications (3)

This will count as one of your downloads.

You will have access to both the presentation and article (if available).

DOWNLOAD NOW

This content is available for download via your institution's subscription. To access this item, please sign in to your personal account.

Email or Username Forgot your username?

Password Forgot your password?

Show

Keep me signed in

No SPIE account? Create an account

Proceedings Article | 31 January 2020 Paper

Known classes classification and unknown classes targets detection of hyperspectral images based on convolutional neural networks

Lixiong Zhang, Yuanxi Pen, Tian Jian, Yu Liu, Kecheng Gong, Lu Liu, Liyuan Zhao, Longlong Zhang

Proceedings Volume 11427, 114272Y (2020) https://doi.org/10.1117/12.2552740

Read Abstract +

Traditional hyperspectral feature extraction methods focus on spectral features and neglect spatial features,its extraction method is set in advance and is not suitable for all hyperspectral images. Faced with these problems, we propose a three-dimensional convolutional network for hyperspectral classification, which consists of a convolutional layer,2 downsampling layers, 2 identification layers, a flatten layer, and 4 fully connected layers. The proposed network employs three-dimensional convolution operation to extract spectral-spatial features from hyperspectral images,there are two reasons for this, the first reason is three-dimensional convolution can automatically learn a large number of mappings between input and output.The second reason is three-dimensional convolution can effectively extract spectral-spatial features and improve network classification performance. In order to extract high-level features and prevent network performance degradation, the proposed network adopts residual connections.More importantly, the OpenMax algorithm is employed to detect hyperspectral unknown targets. In addition to the probability that the output belongs to a known class, the OpenmMx adds the probability that the predicted input belongs to unknown classes, as a result,the deep convolutional network can respond to inputs of unknown classes.experiments based on typical hyperspectral data show that the proposed network perform accurately in the known classes classification and the openmax algorithm is suitable for unknown targets detection of hyperspectral images.

Proceedings Article | 31 January 2020 Paper

Interleaved group convolution network for hyperspectral image classification

Mingrui Su, Yu Liu, Lu Liu, Yuanxi Peng, Tian Jiang

Proceedings Volume 11427, 114270B (2020) https://doi.org/10.1117/12.2549150

KEYWORDS: Convolution, Image classification, Hyperspectral imaging, Network architectures, Sensors, Image sensors, Neural networks

Read Abstract +

Researches have shown that using convolution neural network (CNN) on spatial-spectral domain can improve the performance of hyperspectral image (HSI) classification in recently years. However, due to the existence of spectral redundancy and the high dimensional kernels used in 3D-CNN, the HSI classification models are often heavy with a huge number of parameters and high computation complexity. Motivated by the lightweight model, this paper introduced a modular convolution structure named three-dimensional interleaved group convolution (3D-IGC). This structure contains two successive group convolutions with a channel shuffle operation between them. First group convolution extracts feature on spatial-spectral domain. Then the channel shuffle enables cross-group information interchange. After this, the second group convolution perform the point-wise convolution. We proved that an IGC is wider than a normal convolution in most cases by inferred formula. The empirical results demonstrate that the increment of width in 3D-IGC model is beneficial to HSI classification with the computation complexity preserved, especially when the model has fewer parameters. Compared with the normal convolution, the 3D-IGC can largely reduce the redundancy of convolution filters in channel domain, which greatly decreases the number of parameters and the computation cost without losing classification accuracy. We also considered the effects of the 3D-IGC on deep neural networks, therefore we used the 3D-IGC to modify the residual unit and get a lightweight model compared with ResNets.

Proceedings Article | 31 January 2020 Paper

Deep residual prototype learning network for hyperspectral image classification

Yu Liu, Mingrui Su, Lu Liu, Chunchao Li, Yuanxi Peng, Jing Hou, Tian Jiang

Proceedings Volume 11427, 1142705 (2020) https://doi.org/10.1117/12.2548524

KEYWORDS: Hyperspectral imaging, Visualization, Image classification, 3D visualizations, Image visualization, Pattern recognition, Remote sensing

Read Abstract +

Deep residual networks (ResNets) can learn deep feature representation from hyperspectral images (HSIs), and therefore have been widely used for HSI classification. Despite their high accuracies, there still exist a lot of challenging cases, such as open world recognition, limited-sample learning and visualization of learned classification features, which cannot be well addressed. Most of the challenges in HSI classification can be attributed to the dependence on softmax based loss function and classifiers, which cause the lack of robustness for deep learning models and the hardness to visualize the learned classification features. To improve the robustness and achieve the visualization of learned classification features, we propose a novel learning framework called Residual Prototype Learning Network, a combination of residual network and prototypes learning mechanism. Under the framework, a prototype learning based loss function is proposed to enhance intra-class compactness and the inter-class separation of these feature representations; in addition, a prototype learning based classifier is simultaneously proposed to achieve the 2D or 3D visualization of the classification features. The effectiveness of our proposed learning framework is evaluated on several publicly available HSI benchmarks, and the experimental results show that our approach achieve better results than traditional softmax based ResNets.

My Library

You currently do not have any folders to save your paper to! Create a new folder below.

Folder Name

Folder Description

View contact details

UPDATE YOUR PROFILE

Is this your profile? Update it now.

Sign into your SPIE.org account

Don’t have a profile and want one?

Create an account on SPIE.org

Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks. You are receiving this notice because your organization may not have SPIE eBooks access.*

*Shibboleth/Open Athens users─please sign in to access your institution's subscriptions.

To obtain this item, you may purchase the complete book in print or electronic format on SPIE.org.

ORGANIZATIONAL
Sign in with credentials provided by your organization.

Organizational Username

Organizational Password

Show/Hide Password

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members:

Non-members: ADD TO CART

Keywords/Phrases

Search In:

Publication Years