Liuyang Luo
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Liuyang Luo has a total of 10 co-authors. Here are the top co-authors by number of shared publications:
Bao Zhang
1 shared publicationChenglin Zhao
1 shared publicationChunlong Li
1 shared publicationChunyu Pu
State Key Laboratory of Coal Mine Disaster Dynamics and Control, Chongqing University, Chongqing 400044, China and Key Laboratory of Optoelectronic Technology and Systems of the Education Ministry of China, Chongqing University, Chongqing 400044, China
1 shared publication- Hong Huang
Key Laboratory of Optoelectronic Technology and Systems of the Education Ministry of China, Chongqing University, Chongqing, China
1 shared publication - Xingqi Zou
Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China
1 shared publication
Latest Publications
- 11 January 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Letters
- Vol. 19 (1545598X) , 1-5
Self-Supervised Convolutional Neural Network via Spectral Attention Module for Hyperspectral Image Classification
Hyperspectral image (HSI) classification is a hot topic in the field of remote sensing, and convolutional neural networks (CNNs) have shown good classification performance for their capabilities of feature extraction. However, traditional CNN-based methods require a lot of labeled data during their training process, while the acquisition of labeled samples is complicated and time-consuming. In addition, a key issue for HSI classification is how to effectively explore the correlation within the spectral dimension and emphasize important spectral bands. In this letter, an end-to-end framework named spectral attention-based self-supervised CNN (SASCNN) is put forward for HSI classification. At first, the SASCNN takes raw 3-D cubes as input data, and a spectral attention module (SAM) is used to adaptively optimize channel-wise characteristics by adjusting the importance among continuous spectral bands. Then, by flexibly adding multilayer concatenation to integrate shallow and abstract features, the designed encoder-decoder part can be used to learn discriminative features and reproduce the inputs in a self-supervised manner. Experiments over the Heihe and Houston datasets demonstrate the effectiveness of the proposed self-supervised learning method.
- 11 January 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Letters
- Vol. 19 (1545598X) , 1-5
Self-Supervised Convolutional Neural Network via Spectral Attention Module for Hyperspectral Image Classification
Hyperspectral image (HSI) classification is a hot topic in the field of remote sensing, and convolutional neural networks (CNNs) have shown good classification performance for their capabilities of feature extraction. However, traditional CNN-based methods require a lot of labeled data during their training process, while the acquisition of labeled samples is complicated and time-consuming. In addition, a key issue for HSI classification is how to effectively explore the correlation within the spectral dimension and emphasize important spectral bands. In this letter, an end-to-end framework named spectral attention-based self-supervised CNN (SASCNN) is put forward for HSI classification. At first, the SASCNN takes raw 3-D cubes as input data, and a spectral attention module (SAM) is used to adaptively optimize channel-wise characteristics by adjusting the importance among continuous spectral bands. Then, by flexibly adding multilayer concatenation to integrate shallow and abstract features, the designed encoder-decoder part can be used to learn discriminative features and reproduce the inputs in a self-supervised manner. Experiments over the Heihe and Houston datasets demonstrate the effectiveness of the proposed self-supervised learning method.
- 30 July 2019
- Published by IOP Publishing in Semiconductor Science and Technology
- Vol. 34 (9) , 095004
An effective process to remove etch damage prior to selective epitaxial growth in 3D NAND flash memory
An effective post etch treatment (PET) process was proposed to eliminate etch damage in the channel hole, of which the depth and the aspect ratio is beyond 3 μ m and 30:1 respectively, prior to selective epitaxial growth (SEG) in three dimensions (3D) NAND flash memory. In this work, it is demonstrated that the damaged layer both at the channel hole bottom and sidewall induced by capacitively coupled plasma (CCP) was effectively eliminated using low energy plasma of CL 2 + NF 3 /CH 2 F 2 in PET, and then obtaining an excellent surface condition in channel hole and fabricating a void-free SEG epitaxial layer.
- 30 July 2019
- Published by IOP Publishing in Semiconductor Science and Technology
- Vol. 34 (9) , 095004
An effective process to remove etch damage prior to selective epitaxial growth in 3D NAND flash memory
An effective post etch treatment (PET) process was proposed to eliminate etch damage in the channel hole, of which the depth and the aspect ratio is beyond 3 μ m and 30:1 respectively, prior to selective epitaxial growth (SEG) in three dimensions (3D) NAND flash memory. In this work, it is demonstrated that the damaged layer both at the channel hole bottom and sidewall induced by capacitively coupled plasma (CCP) was effectively eliminated using low energy plasma of CL 2 + NF 3 /CH 2 F 2 in PET, and then obtaining an excellent surface condition in channel hole and fabricating a void-free SEG epitaxial layer.