Aili Wang

Soochow University

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

Haibin Wu

Harbin University of Science and Technology

Yuji Iwahori

Chubu University

Chengyang Liu

Jiangsu University

Yushi Chen

Harbin Institute of Technology

Xue Dong

Shanghai Jiao Tong University

Yonghong Zhang

Soochow University

Kaiyang Xu

Harbin University of Science and Technology

Huaming Zhou

Harbin University of Science and Technology

Hongyuan Wu

China Southern Power Grid (China)

Baotian Dong

Beijing Jiaotong University

Cooperative Institutions

Chinese Academy of Sciences

Harbin University of Science and Technology

Soochow University

University of Chinese Academy of Sciences

Shanghai Institute of Materia Medica

Sun Yat-sen University

Shandong University

Capital Medical University

Peking University

Nanjing University of Chinese Medicine

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

Noise-Disruption-Inspired Neural Architecture Search with Spatial–Spectral Attention for Hyperspectral Image Classification

Remote Sensing (2024)

Aili Wang Kang Zhang Haibin Wu Shiyu Dai Yuji Iwahori

In view of the complexity and diversity of hyperspectral images (HSIs), the classification task has been a major challenge in the field of remote sensing image processing. Hyperspectral classification (HSIC) methods based on neural architecture search (NAS) is a current attractive frontier that not only automatically searches for neural network architectures best suited to the characteristics of HSI data, but also avoids the possible limitations of manual design of neural networks when dealing with new classification tasks. However, the existing NAS-based HSIC methods have the following limitations: (1) the search space lacks efficient convolution operators that can fully extract discriminative spatial–spectral features, and (2) NAS based on traditional differentiable architecture search (DARTS) has performance collapse caused by unfair competition. To overcome these limitations, we proposed a neural architecture search method with receptive field spatial–spectral attention (RFSS-NAS), which is specifically designed to automatically search the optimal architecture for HSIC. Considering the core needs of the model in extracting more discriminative spatial–spectral features, we designed a novel and efficient attention search space. The core component of this innovative space is the receptive field spatial–spectral attention convolution operator, which is capable of precisely focusing on the critical information in the image, thus greatly enhancing the quality of feature extraction. Meanwhile, for the purpose of solving the unfair competition issue in the traditional differentiable architecture search (DARTS) strategy, we skillfully introduce the Noisy-DARTS strategy. The strategy ensures the fairness and efficiency of the search process and effectively avoids the risk of performance crash. In addition, to further improve the robustness of the model and ability to recognize difficult-to-classify samples, we proposed a fusion loss function by combining the advantages of the label smoothing loss and the polynomial expansion perspective loss function, which not only smooths the label distribution and reduces the risk of overfitting, but also effectively handles those difficult-to-classify samples, thus improving the overall classification accuracy. Experiments on three public datasets fully validate the superior performance of RFSS-NAS.

10.3390/rs16173123

Cite

Citations (1)

The Structure and Development of Regional Ecological Water Conservancy Economic System

Journal of Coastal Research (2020)

Chuanchang Gao Aili Wang Xinyuan Guo Shunsheng Wang

Gao, C.; Wang, A.; Guo, X., and Wang, S., 2020. The structure and development of regional ecological water conservancy economic system. In: Yang, Y.; Mi, C.; Zhao, L., and Lam, S. (eds.), Global Topics and New Trends in Coastal Research: Port, Coastal and Ocean Engineering. Journal of Coastal Research, Special Issue No. 103, pp. 65–69. Coconut Creek (Florida), ISSN 0749-0208.In order to the eco-economic theory, ecological water conservancy research, and regional water resources rational coordination theory, the paper studied development and change rules of regional ecological water conservancy economic system using the method of system analysis, combining with the research status of the regional ecological water conservancy economic system in China. Base on analyzing the basic concept of regional ecological water conservancy economic system, the components of regional ecological water conservancy economic system were discussed. The research subject of this paper is to provide a theoretical basis and train of thought for the establishment of new development mode of regional economy and society, which will play a certain role in promoting the protection of natural environment and the development of economy and society.

Port (circuit theory)

10.2112/si103-014.1

Cite

Citations (4)

Hyperspectral image classification based on convolutional neural network and random forest

Remote Sensing Letters (2019)

Aili Wang Ying Wang Yushi Chen

Deep learning-based methods, especially deep convolutional neural network (CNN), have proven their powerfulness in hyperspectral image (HSI) classification. On the other hand, ensemble learning is a useful method for classification task. In this letter, in order to further improve the classification accuracy, the combination of CNN and random forest (RF) is proposed for HSI classification. The well-designed CNN is used as individual classifier to extract the discriminant features of HSI and RF randomly selects the extracted features and training samples to formulate a multiple classifier system. Furthermore, the learned weights of CNN are adopted to initialize other individual CNN. Experimental results with two hyperspectral data sets indicate that the proposed method provides competitive classification results compared with state-of-the-art methods.

Contextual image classification

10.1080/2150704x.2019.1649736

Cite

Citations (65)

Association between light rare earth elements in maternal plasma and the risk of spontaneous preterm birth: a nested case-control study from the Beijing birth cohort study

Environmental Health (2023)

Junxi Chen Aili Wang Hang An Weiling Han Junhua Huang

Parental exposure to rare earth elements (REEs) could increase the risk of premature rupture of membranes, a major cause of spontaneous preterm birth (SPB). In addition, different subtypes of SPB, such as spontaneous preterm labor (SPL) and preterm premature rupture of membranes (PPROM), may have different susceptibility to environmental exposure. Therefore, we investigated the potential associations between REE exposure in different trimesters and SPB and its subtypes.A nested case-control study was performed. We included 244 women with SPB as cases and 244 women with full-term delivery as controls. The plasma concentrations of light REEs were measured in the first and third trimesters. Logistic regression was used to analyze the associations between single REE levels and SPB, and Bayesian kernel machine regression (BKMR) was used to analyze the mixed-exposure effect.Exposure to light REEs was associated with SPB and its subtypes only in the third trimester. Specifically, the intermediate- and highest-tertile concentration groups of La and the highest-tertile concentration group of Sm were associated with an increased risk of SPL, with adjusted odds ratios (AORs) of 2.00 (95% CIs: 1.07-3.75), 1.87 (95% CIs: 1.01-3.44), and 1.82 (95% CIs: 1.00-3.30), respectively. The highest-tertile concentration group of Pr was associated with an increased risk of PPROM, with an AOR of 1.69 (95% CIs: 1.00-2.85). Similar results were also found in BKMR models.La and Sm levels in plasma may be associated with the risk of SPL, and Pr levels in plasma may be associated with the risk of PPROM.

Nested case-control study

10.1186/s12940-023-01027-1

Cite

Citations (5)

LiDAR Data Classification Using Morphological Profiles and Convolutional Neural Networks

IEEE Geoscience and Remote Sensing Letters (2018)

Aili Wang Xin He Pedram Ghamisi Yushi Chen

In recent years, deep learning-based methods, especially convolutional neural networks (CNNs), have shown their capabilities in remote sensing data processing. The efficacy of light detection and ranging (LiDAR) has been already proven in a wide variety of research areas. Most of the existing methods do not extract the informative features from LiDAR-derived rasterized digital surface models (LiDAR-DSM) data in a deep manner. In order to utilize the advantages of deep models for the classification of LiDAR-derived features, deep CNN is proposed here to hierarchically extract the robust and discriminant features of the input data. Moreover, morphological profiles and multiattribute profiles (MAPs) are investigated to enrich the inputs of the CNN and further to improve the ultimate classification performance. Furthermore, a new activation function, sigmoid-weighted linear units (SiLUs), is introduced. The proposed frameworks are tested on two LiDAR-DSMs (i.e., Bayview Park and Houston data sets). The MAP-CNNs with SiLU outperform original CNNs by 6.62% and 6.88% in terms of overall accuracy on Bayview Park and Houston data sets, respectively, when the number of training samples of each class is 40.

Ranging

Sigmoid function

Contextual image classification

10.1109/lgrs.2018.2810276

Cite

Citations (31)

Mean Shift Pedestrian Tracking Algorithm Based on Multi-Feature Probability Distribution

Huanan Ligong Daxue Xuebao. Ziran Kexue Ban (2016)

Aili Wang Baotian Dong Hongyuan Wu

Feature (linguistics)

Mean-shift

Tracking (education)

Source

Cite

Citations (1)

Efficient Convolutional Neural Architecture Search for LiDAR DSM Classification

IEEE Transactions on Geoscience and Remote Sensing (2022)

Aili Wang Xue Dong Haibin Wu Yanfeng Gu

Light detection and ranging (LiDAR) data provide rich elevation information, so it plays an irreplaceable role in ground object classification. Recently, convolutional neural networks (CNNs) have shown excellent performance in LiDAR digital surface models (DSMs) classification. However, the architecture of CNN model relies heavily on manual design, so it has great limitations. In addition, different sensors capture LiDAR datasets with different properties, so the model should be designed to suit for different datasets, which further increases the workload of architecture design. Therefore, this article proposes a method of automatic design of LiDAR DSM classification model. First, attention mechanism is introduced into search space to improve the feature extraction capability of the model. Then, a gradient-based search strategy is used to obtain the optimal architecture from this search space. Second, a learning rate adjustment strategy is proposed to reduce the time spent in the search stage and evaluation stage to improve the classification accuracy of the model. Finally, a regularization scheme is introduced to enhance the robustness of the model and avoid overfitting. Experimental results on three public LiDAR datasets (Bayview Park, Recology, and Houston) obtained from different sensors show that the proposed neural architecture search method achieves the impressive classification performance compared to several state-of-the-art classification methods and improves the classification accuracy under the condition of limited training samples.

Overfitting

Robustness

10.1109/tgrs.2022.3171520

Cite

Citations (6)

Association of Biomarkers of Inflammation with Dyslipidemia and Its Components among Mongolians in China

PLoS ONE (2014)

Lingyan Tang Hao Peng Tian Xu Aili Wang Guiyan Wang

Objective This study aims to examine the association between inflammatory biomarkers and dyslipidemia and its components among Mongolians in China. Methods Data were obtained from 2544 Mongolians via standard questionnaires and blood samples in Inner Mongolia, China. High sensitivity C-reactive protein (hsCRP), soluble intercellular adhesion molecule-1 (sICAM-1) and soluble E-selectin (sE-selectin) as well as blood lipids were examined. Results Individuals with dyslipidemia had higher levels of hsCRP, sICAM-1 and sE-selectin than those without dyslipidemia (all P values<0.001). Compared to the lowest quartile of inflammatory biomarkers, individuals with the highest quartile were more likely to have dyslipidemia (odds ratio, 95% confidence interval: 3.215, 2.551–4.116 for hsCRP; 1.575, 1.253–1.980 for sICAM-1; 1.495, 1.193–1.873 for sE-selectin). Moreover, hsCRP was associated with all the components of dyslipidemia, whereas, sICAM-1 was not related to high density lipoprotein cholesterol (HDL-c) or triglycerides (TAG). Additionally, sE-selectin was just associated with TAG. Conclusion Our study indicated that elevated plasma levels of hsCRP, sICAM-1 and sE-selectin were positively and significantly associated with increased risk of dyslipidemia among Mongolians. However, the associations were not identical for different inflammatory biomarkers with the components of dyslipidemia.

Dyslipidemia

Quartile

10.1371/journal.pone.0089023

Cite

Citations (36)

A Novel Dual-Encoder Model for Hyperspectral and LiDAR Joint Classification via Contrastive Learning

Remote Sensing (2023)

Haibin Wu Shiyu Dai Chengyang Liu Aili Wang Yuji Iwahori

Deep-learning-based multi-sensor hyperspectral image classification algorithms can automatically acquire the advanced features of multiple sensor images, enabling the classification model to better characterize the data and improve the classification accuracy. However, the currently available classification methods for feature representation in multi-sensor remote sensing data in their respective domains do not focus on the existence of bottlenecks in heterogeneous feature fusion due to different sensors. This problem directly limits the final collaborative classification performance. In this paper, to address the bottleneck problem of joint classification due to the difference in heterogeneous features, we innovatively combine self-supervised comparative learning while designing a robust and discriminative feature extraction network for multi-sensor data, using spectral–spatial information from hyperspectral images (HSIs) and elevation information from LiDAR. The advantages of multi-sensor data are realized. The dual encoders of the hyperspectral encoder by the ConvNeXt network (ConvNeXt-HSI) and the LiDAR encoder by Octave Convolution (OctaveConv-LiDAR) are also used. The adequate feature representation of spectral–spatial features and depth information obtained from different sensors is performed for the joint classification of hyperspectral images and LiDAR data. The multi-sensor joint classification performance of both HSI and LiDAR sensors is greatly improved. Finally, on the Houston2013 dataset and the Trento dataset, we demonstrate through a series of experiments that the dual-encoder model for hyperspectral and LiDAR joint classification via contrastive learning achieves state-of-the-art classification performance.

Feature (linguistics)

Feature Learning

Discriminative model

Autoencoder

10.3390/rs15040924

Cite

Citations (6)

A Corin Variant Identified in Hypertensive Patients That Alters Cytoplasmic Tail and Reduces Cell Surface Expression and Activity

Scientific Reports (2014)

Yue Zhang Hui Li Jianping Zhou Aili Wang Junhua Yang

Corin is a membrane-bound protease that regulates blood pressure by activating the natriuretic peptides. CORIN variants have been associated with hypertension and heart disease in African Americans. In this study, we conducted targeted exome sequencing and identified an insertion variant, c.102_103insA, in exon 1 of the CORIN gene. Analysis of two independent cohorts showed that the variant was preferentially present in hypertensive patients (38/795 or 4.78% vs. 4/632 or 0.63% in normal individuals, p = 4.14E-6). The insertion shifted the reading frame, resulting in a corin variant with a truncated cytoplasmic tail. In cell-based studies, the corin variant exhibited poor trafficking in the Golgi, reduced cell surface expression and zymogen activation and low natriuretic peptide processing activity. Compared with normal individuals with the wild-type allele, individuals with the variant allele had lower levels of plasma corin [0.59 ± 0.07 ng/mL (n = 25) vs. 0.91 ± 0.02 ng/mL (n = 215), p<0.001] and higher levels of plasma N-terminal pro-atrial natriuretic peptide (NT-pro-ANP) [2.39 ± 3.6 nmol/L (n = 21) vs. 0.87 ± 0.6 nmol/L (n = 48), p = 0.005]. These results indicate that the variant altered corin structure and impaired the natriuretic peptide processing activity in vivo. The results highlight corin defects as an important underlying mechanism in hypertension.

Atrial natriuretic peptide

10.1038/srep07378

Cite

Citations (26)