Chunle Wang

Dr. Chunle Wang


Research Focus & Metrics

Academic Career Length

13Years (2014-Present)

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

Representative Publications

View All Publications
1 November 2023 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Verified
Journal Article
Open access
SXSiyu Xie
Siyu Xie
MZMei Zhou
Mei Zhou
CWChunle Wang
Chunle Wang
SHShisheng Huang
Shisheng Huang
Detecting and recognizing objects are crucial steps in interpreting remote sensing images. At present, deep learning methods are predominantly employed for detecting objects in remote sensing images, necessitating a significant number of floating-point computations. However, low computing power and small storage in computing devices are hard to afford the large model parameter quantity and high computing complexity. To address these constraints, this article presents a lightweight detection model called CSPPartial-YOLO. This model introduces the partial hybrid dilated convolution (PHDC) Block module that combines hybrid dilated convolutions and partial convolutions to increase the receptive field at a low computational cost. By using the PHDC Block within the model design framework of cross-stage partial connection, we construct CSPPartialStage that reduces computational burden without compromising accuracy. Coordinate attention module is also employed in CSPPartialStage to aggregate position information and improve the detection of small objects with complex distributions in remote sensing images. A backbone and neck are developed with CSPPartialStage, and the rotation head of the PPYOLOE-R model adapts to objects of multiple orientations in remote sensing images. Empirical experiments using the dataset for object deTection in aerial images (DOTA) dataset and a large-scale small object detection dAtaset (SODA-A) dataset indicate that our method is faster and resource efficient than the baseline model (PPYOLOE-R), while achieving higher accuracy. Furthermore, comparisons with current state-of-the-art YOLO series detectors show our proposed model's competitiveness in terms of speed and accuracy. Moreover, compared to mainstream lightweight networks, our model exhibits better hardware adaptability, with lower inference latency and higher detection accuracy.

29 April 2020 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Verified
Journal Article
Open access
YWYu Wang
Yu Wang
WYWeidong Yu
Weidong Yu
CWChunle Wang
Chunle Wang
XLXiuqing Liu
Xiuqing Liu
In this paper, a modified four-component decomposition method with refined volume scattering models is proposed for polarimetric synthetic aperture radar (SAR) image processing. In the new decomposition method, after the orientation angle compensation (OAC), the orientation angle is placed in the probability density functions. General forms of the volume scattering models and branch conditions can be obtained. Similar to the four-component scattering power decomposition with extended volume scattering model (S4R) proposed by Sato et al., refined volume scattering models can be used for various land covers based on the criteria. Since the orientation angles are contained in the refined volume scattering models, the oriented buildings can be discriminated from the vegetation areas and the overestimation problem of volume scattering (OVS) is substantially overcome. In this paper, the performance of the proposed method is evaluated by the spaceborne C-band Gaofen-3 data and airborne L-band E-SAR data. Several approaches are employed as a comparison of the proposed methods. Experimental results show that, compared with the existing decomposition methods, the proposed method can effectively represent the scattering characteristics of the ambiguous regions and the double-bounce scattering contributions over the urban areas can be substantially enhanced.

5 March 2024 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Verified
Journal Article
Open access
YWYu Wang
Yu Wang
WYWeidong Yu
Weidong Yu
RWRuishi Wang
Ruishi Wang
LWLi Wang
Li Wang
DGDaqing Ge
Daqing Ge
LZLi Zhang
Li Zhang
Urban area extraction using polarimetric synthetic aperture radar (PolSAR) data is an important application for urban planning and disaster assessment. Due to the variability in urban structures, the problem of mistaking built-up areas for vegetation remains challenging. Then, a new urban extraction method is proposed using eigenvalues and optimal roll-invariant features. First, similar to the entropy/anisotropy plane, a 2-D RVI/PA plane is adopted to construct the extractor of surface scattering areas. Then, the optimal ratio of the correlation coefficient is proposed as an extractor to depict the scattering characteristics of buildings with various orientation angles, while restraining the scattering characteristics of other land covers. In addition, reliable hidden features are selected to refine the extraction results. Finally, the spatial information extraction and classification methods are introduced to automatically determine the thresholds of parameters above, based on which the urban area extraction map can be obtained. C-band Gaofen-3 SAR data collected in Runan, Henan Province, China and San Francisco, USA, and L-band E-SAR data collected in Oberpfaffenhofen, Germany are used to validate the performance of the proposed method. Experimental results demonstrate that the method can distinguish between urban and natural areas with high accuracy and has good visual effects in extracting building areas.

20 August 2022 in Remote Sensing
Verified
Journal Article
Open access
PDF file
HLHongliang Lu
Hongliang Lu
JSJili Sun
Jili Sun
JWJili Wang
Jili Wang
CWChunle Wang
Chunle Wang
Synthetic aperture radar (SAR) tomography (TomoSAR) has been widely used in the three-dimensional (3D) reconstruction of urban areas using the multi-baseline (MB) SAR data. For urban scenarios, the MB SAR data are often acquired by repeat-pass using the spaceborne SAR system. Such a data stack generally has long time baselines, which result in different atmospheric disturbances of the data acquired by different tracks. These factors can lead to the presence of phase errors (PEs). PEs are multiplicative noise for observation data, which can cause diffusion and defocus in TomoSAR imaging and seriously affect the extraction of target 3D information. In this paper, we combine the methods of the block-building network (BBN) and phase gradient autofocus (PGA) to propose a novel phase compensation method called BBN-PGA. The BBN-PGA method can effectively and efficiently compensate for PEs of the MB SAR data over a wide area and improve the accuracy of 3D reconstruction of urban areas. The applicability of this proposed BBN-PGA method is proved by using simulated data and the spaceborne MB SAR data acquired by the TerraSAR-X satellite over an area in Barcelona, Spain.

28 May 2020 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Verified
Journal Article
Open access
YWYu Wang
Yu Wang
WYWeidong Yu
Weidong Yu
XLXiuqing Liu
Xiuqing Liu
CWChunle Wang
Chunle Wang
Arjan KuijperArjan Kuijper
Arjan Kuijper
SGStefan Guthe
Stefan Guthe
The overestimation of volume scattering is an essential shortcoming of the model-based polarimetric synthetic aperture radar (PolSAR) target decomposition method. It is likely to affect the measurement accuracy and result in mixed ambiguity of scattering mechanism. In this paper, an extended adaptive four-component decomposition method (ExAG4UThs) is proposed. First, the orientation angle compensation (OAC) is applied to the coherency matrix and artificial areas are extracted as the basis for selecting the decomposition method. Second, for the decomposition of artificial areas, one of the two complex unitary transformation matrices of the coherency matrix is selected according to the wave anisotropy (Aw). In addition, the branch condition that is used as a criterion for the hierarchical implementation decomposition is the ratio of the correlation coefficient (Rcc). Finally, the selected unitary transformation matrix and discriminative threshold are used to determine the structure of the selected volume scattering models, which are more effectively to adapt to various scattering mechanisms. In this paper, the performance of the proposed method is evaluated on GaoFen-3 full PolSAR data sets for various time periods and regions. The experimental results demonstrate that the proposed method can effectively represent the scattering characteristics of the ambiguous regions and the oriented building areas can be well discriminated as dihedral or odd-bounce structures.

24 July 2020 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Verified
Journal Article
Open access
YWYu Wang
Yu Wang
WYWeidong Yu
Weidong Yu
XLXiuqing Liu
Xiuqing Liu
CWChunle Wang
Chunle Wang
In this paper, an extended seven-component decomposition method using refined volume scattering models and new configurations of mixed dipoles (M7SD) is proposed to interpret the scattering mechanisms of various terrain types. In the M7SD, refined volume scattering models are proposed to reasonably characterize the scattering mechanisms of various land covers with different orientation angles, and the redefined configurations of the mixed dipoles, which have better explanations for the scattering mechanisms, are introduced. Although the diameter of the dipole used in the experiment is very small, the superposition of two dipoles in the same plane will cause one of the dipoles to have a distance from the plane. According to the experimental results, since the distance between one of the dipoles and plane cannot be ignored, it will cause an error between the scattering matrix obtained by the experiment and the scattering matrix obtained under ideal conditions. In this case, in the M7SD, two dipoles are placed in two planes, respectively, with a distance of half wavelength. The performance of the proposed method is demonstrated and evaluated with the spaceborne C-band GaoFen-3, Radarsat-2 and airborne L-band ESAR fully polarimetric data over different study sites. The experimental results demonstrate that the use of M7SD can reasonably interpret the scattering mechanisms of various terrain types (especially the building areas with large orientation angles), overcome the overestimation of the volume scattering and reduce the percentage of negative scattering power pixels.