Hua Wei

Hua Wei

Auto-generated profile, based on publicly available information

Hua Wei

Shandong University

Research Focus & Metrics

Academic Career Length

15Years (2012-Present)

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

Representative Publications

View All Publications
7 June 2021 in IEEE Access
Verified
Journal Article
Open access
BWBo Wang
Bo Wang
Ming ZhuMing Zhu
Ming Zhu
YLYing Lu
Ying Lu
JWJiarong Wang
Jiarong Wang
WGWen Gao
Wen Gao
HWHua Wei
Hua Wei
This paper aims to apply real-time light-weight high-precision 3D detection for autonomous driving. We propose LIDAR-based 3D object detection based on foreground segmentation using a fully sparse convolutional network (FS 2 3D). We design a sparse convolutional backbone network and a sparse convolutional detection head to efficiently use the computing and memory resources and accelerate the inference. Instead of using the anchor-based method, we convert the detection problem into a foreground segmentation problem on a bird's-eye view. The sparse convolutional detection head predicts the objectness and bounding box on each active point on the sparse feature map. We design a new oriented bounding box coding method and corresponding loss functions. We predict the endpoints of two mutually perpendicular lines that pass through the foreground active points and indirectly predict the objects' oriented bounding box from these four endpoints. We use the indirectly calculated object center, size, and orientation as inputs of loss functions in the training step. Experiments on the KITTI dataset show that the sparse backbone network we designed is 2.2 times faster and 18.4 times fewer FLOPs than the dense backbone network. The average improvement of the loss functions based on the bounding box code is 1.1% and 0.8% on the BEV and 3D detection, respectively, compared to no addition of these losses. Moreover, FS 2 3D outperforms the state-of-the-art LIDAR-based method in speed and precision for both cars and cyclists.

3 June 2020 in IEEE Access
Verified
Journal Article
Open access
HWHua Wei
Hua Wei
MZMing Zhu
Ming Zhu
BWBo Wang
Bo Wang
JWJiarong Wang
Jiarong Wang
DSDeyao Sun
Deyao Sun
The learning of discriminative features is the key for fine-grained image recognition. To better extract effective features and improve the accuracy of fine-grained image recognition, we propose a two-level progressive attention convolutional network (TPA-CNN) for fine-grained image recognition. The model includes a multi-channel attention-fusion (MCAF) module and a cross-layer element-attention (CEA) module. The MCAF module is used to find distinctive feature map channels which significantly responds to specific regions. Inspired by Hierarchical Bilinear Pooling model, The CEA module is further assign weight values to feature map elements. From the perspective of attention visualization map, MCAF module can focused on one or more positive regions, CEA module further locates the most discriminative regions by interaction between the feature map elements. The model can dynamically search the discriminative region of the image, not limited to the boost or crop a selected region. Compared to previous models basing on attention mechanism, the model can extract non-correlated part features which spread over object foreground areas, further improving the recognition accuracy. Experimental results on CUB-200-2011, FGVC-Aircraft, and Stanford Cars datasets demonstrate that the proposed TPA-CNN achieves competitive performance.

15 November 2012 in Journal of Seismology
Verified
Journal Article
Closed access
PDF file
WHWei Hua
Wei Hua
ZCZhangli Chen
Zhangli Chen
SZSihua Zheng
Sihua Zheng
CYChunqing Yan
Chunqing Yan

26 March 2024 in IEEE Geoscience and Remote Sensing Letters
Verified
Journal Article
Closed access
HWHua Wei
Hua Wei
KGKun Gao
Kun Gao
JWJing Wang
Jing Wang
QTQiuyan Tang
Qiuyan Tang
XTXiongxin Tang
Xiongxin Tang
FXFanjiang Xu
Fanjiang Xu
Due to atmospheric turbulence, optical system limitations, satellite platform jitter, and other reasons, remote sensing images inevitably undergo different degrees of degradation. Employing the deep learning method to improve the on-orbit image quality faces many challenges such as lack of data, limited computing resources, network architecture design, and so on. Among these factors, establishing a physics-guided dataset during the image restoration stage and avoiding unforeseen effects such as ringing pose a significant challenge for remote sensing image restoration. This letter proposes an optical imaging degradation simulation model and Transformer-based algorithm to improve remote sensing image quality. First, we model the degradation result from phase to image of optical remote sensing imaging using Zernike Polynomials, thus, a large-scale paired dataset is constructed. Then, a multi-level feature fusion transformer is introduced to mitigate the defect during restoration. The proposed algorithm incorporates a multi-level feature fusion module to fuse feature information from multi-scales effectively. Additionally, a multi-level space and frequency loss function is introduced to enhance the learning of high-frequency information to ensure that the edge suppresses noise amplification and ringing effects during recovery. Finally, experimental results on synthetic data show that our method improved by 25.4% and 22.3% with the blurred images on the PSNR index and SSIM index. Visual results on the GaoFen-1/2A PMS images have enhanced clarity and suppressed artifacts such as ringing which demonstrate the effectiveness and capability of our proposed method.

8 January 2026 in Expert Systems with Applications
Verified
Journal Article
Closed access
MXMuhao Xu
Muhao Xu
ZNZihan Nie
Zihan Nie
BFBaochen Fu
Baochen Fu
ZCZhuangzhuang Chen
Zhuangzhuang Chen
ZLZijian Li
Zijian Li
WSWeiye Song
Weiye Song

23 July 2025 in IEEE Sensors Journal
Verified
Journal Article
Closed access
ZLZijian Li
Zijian Li
HWHua Wei
Hua Wei
YLYanshuo Li
Yanshuo Li
XHXiang He
Xiang He
MXMuhao Xu
Muhao Xu
WSWeiye Song
Weiye Song
Denoising is a commonly used technique in medical image preprocessing, particularly in the processing of optical coherence tomography (OCT) images. Due to the characteristics of OCT systems and their operational principles, the acquired images inevitably contain a superposition of various types of noise, including speckle noise, structural noise, and random noise, which consequently reduces the accuracy of subsequent segmentation tasks. Moreover, medical image processing places a high emphasis on image details, making it challenging to remove noise without causing excessive smoothing of the images. In addition, the difficulty in obtaining ground truth for medical images limits the application of many supervised learning methods. To address these issues, this article proposes an unsupervised masked region-based denoising diffusion probabilistic model algorithm. This algorithm employs pixel-level masking, which not only effectively enhances the removal of random noise but also enables the model to focus more on adjacent regions, thereby avoiding excessive alteration of the original image’s details and edges. Furthermore, the method leverages information complementarity among data within the same group, effectively preventing information loss caused by masking. Experimental results demonstrate that compared to the baseline, the proposed algorithm increases the CNR to 3.77 with minimal reduction in the structural similarity index measure (SSIM). In addition, further segmentation experiments confirm the effectiveness of the denoising algorithm, achieving a 5.5% improvement in segmentation accuracy. This study presents the MR-DDPM algorithm, which improves the denoising quality of OCT images and consequently enhances segmentation accuracy, without the need for labeled data. This contribution is valuable to the advancement of medical image processing algorithms.