Lingxuan Meng

Lingxuan Meng

Auto-generated profile, based on publicly available information

Lingxuan Meng

School of Resources and Environment, Center for Information Geoscience, University of Electronic Science and Technology of China, Chengdu, China

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
28 March 2022 in IEEE Transactions on Pattern Analysis and Machine Intelligence
Verified
Journal Article
Closed access
JWJinwang Wang
Jinwang Wang
LMLingxuan Meng
Lingxuan Meng
WLWeijia Li
Weijia Li
Wen YangWen Yang
Wen Yang
LYLei Yu
Lei Yu
Gui-Song XiaGui-Song Xia
Gui-Song Xia
Extracting building footprints from aerial images is essential for precise urban mapping with photogrammetric computer vision technologies. Existing approaches mainly assume that the roof and footprint of a building are well overlapped, which may not hold in off-nadir aerial images as there is often a big offset between them. In this paper, we propose an offset vector learning scheme, which turns the building footprint extraction problem in off-nadir images into an instance-level joint prediction problem of the building roof and its corresponding “ roof to footprint ” offset vector. Thus the footprint can be estimated by translating the predicted roof mask according to the predicted offset vector. We further propose a simple but effective feature-level offset augmentation module, which can significantly refine the offset vector prediction by introducing little extra cost. Moreover, a new dataset, Buildings in Off-Nadir Aerial Images (BONAI), is created and released in this paper. It contains 268,958 building instances across 3,300 aerial images with fully annotated instance-level roof, footprint, and corresponding offset vector for each building. Experiments on the BONAI dataset demonstrate that our method achieves the state-of-the-art, outperforming other competitors by 3.37 to 7.39 points in F1-score. The codes, datasets, and trained models are available at https://github.com/jwwangchn/BONAI.git .

18 March 2024 in IEEE Transactions on Geoscience and Remote Sensing
Verified
Journal Article
Closed access
WLWeijia Li
Weijia Li
ZHZhenghao Hu
Zhenghao Hu
LMLingxuan Meng
Lingxuan Meng
JWJinwang Wang
Jinwang Wang
Juepeng ZhengJuepeng Zheng
Juepeng Zheng
DLDahua Lin
Dahua Lin
3D building reconstruction from monocular remote sensing imagery is an important research problem that has been extensively studied for several decades. Although monocular remote sensing imagery is a more economic data source compared with the LiDAR data and multi-view imagery, its limited information results in great challenges and restricts the performance of existing monocular reconstruction methods. Moreover, the expensive cost and the limited quantity of 3D annotations also restrict the application scenes of existing methods, which are mostly based on fully-supervised learning. In our previous work, we have proposed MTBR-Net, a monocular building reconstruction method that consists of a fully-supervised multi-task network and a post-processing module for optimizing the reconstruction results. In this work, we further propose WS-MTBR-Net, a weakly-supervised building reconstruction network that uses fewer 3D annotations and achieves better performance in an end-to-end manner. Specifically, our WS-MTBR-Net fully leverages the relation between different components of a 3D building instance and the property of off-nadir images to improve the footprint segmentation boundary, based on six modified tasks and a new network structure with an improved feature warping module to support weakly-supervised learning. We also design a new training strategy via a hybrid loss function that enables utilizing the training samples with different annotation levels, i.e., complete 3D annotations, 2D footprint annotations, and image-level angle annotations. Results on BONAI Shanghai and Xi’an test datasets demonstrate that our method achieves competitive performance when using 50% fewer 3D-annotated samples, and improves the footprint segmentation F1-score by around 4% compared with current state-of-the-art.

1 August 2018
Verified
Conference Paper
Closed access
YZYingjun Zhang
Yingjun Zhang
JZJi Zhou
Ji Zhou
LMLingxuan Meng
Lingxuan Meng
LMLi Mingsong
Li Mingsong
LDLirong Ding
Lirong Ding
JMJin Ma
Jin Ma
Crops are greatly affected by the temperature of farmland surface during their growing period. It is feasible to investigate the growth status of crops based on temperature information. For serving the research of crop growth status, the component temperature (e.g. temperature of vegetation and temperature of soil) are in need to be obtained. In this study, an unmanned aerial vehicle (UAV) temperature measurement system with a thermal infrared (TIR) imager and a charge-coupled device (CCD) camera is assembled and applied to measure the brightness temperatures of farmland surface. The target areas were photographed by the UAV temperature measurement system according to a pre-set route, and obtain TIR and visible images. The component temperatures are obtained from the TIR image as following processes: (1) When shaded components are negligible at noon, two components, i.e. vegetation and soil, are divided by the OTSU algorithm; and (2) When shaded components cannot be ignored in the morning and afternoon, various components, i.e. vegetation, soil and concrete, the TIR image is divided into soil, vegetation and concrete by the corresponding classified visible images; Then, each of the components is divided into light and shaded components by the OTSU algorithm; thus, four components are obtained, including sunlit vegetation, shaded vegetation, sunlit soil, and shaded soil. The derived component temperatures can serve as inputs to agricultural and water resource models.