Di Zhang

Dr. Di Zhang

orcid

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
29 March 2018 in IOP Conference Series: Materials Science and Engineering
Verified
Conference Paper
Open access
PDF file
DZDi Zhang
Di Zhang
JLJia Liu
Jia Liu
WHWang Heng
Wang Heng
KRKaijun Ren
Kaijun Ren
JSJunqiang Song
Junqiang Song
Ship recognition is the backbone of marine surveillance systems. Recent deep learning methods, e.g. Convolutional Neural Networks (CNNs), have shown high performance for optical images. Learning CNNs, however, requires a number of annotated samples to estimate numerous model parameters, which prevents its application to Synthetic Aperture Radar (SAR) images due to the limited annotated training samples. Transfer learning has been a promising technique for applications with limited data. To this end, a novel SAR ship recognition method based on CNNs with transfer learning has been developed. In this work, we firstly start with a CNNs model that has been trained in advance on Moving and Stationary Target Acquisition and Recognition (MSTAR) database. Next, based on the knowledge gained from this image recognition task, we fine-tune the CNNs on a new task to recognize three types of ships in the OpenSARShip database. The experimental results show that our proposed approach can obviously increase the recognition rate comparing with the result of merely applying CNNs. In addition, compared to existing methods, the proposed method proves to be very competitive and can learn discriminative features directly from training data instead of requiring pre-specification or pre-selection manually.

12 December 2025 in Expert Systems with Applications
Verified
Journal Article
Closed access
DLDawei Li
Dawei Li
JWJingnan Wang
Jingnan Wang
KDKefeng Deng
Kefeng Deng
DZDi Zhang
Di Zhang
CZChengwu Zhao
Chengwu Zhao
JSJunqiang Song
Junqiang Song

16 December 2024 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Verified
Journal Article
Open access
WWWensheng Wang
Wensheng Wang
CLCongjian Li
Congjian Li
DZDi Zhang
Di Zhang
HZHuihui Zhou
Huihui Zhou
MXMingli Xie
Mingli Xie
KFKun Fu
Kun Fu
RGBT tracking technology often struggles in non-cooperative challenges such as illumination variation, scale variation, fast motion, occlusion, and thermal crossover. In this work, we propose a novel RGBT tracking model, called challenge-based Feature complementary Fusion Network (FcFNet) especially aiming at non-cooperative scenarios. In particular, we firstly decouple the mixed challenging attributes and adaptively combine features at each layer using the Challenge Branch Fusion Module (CBFM), forming more discriminative target representations. Subsequently, we employ two non-shared convolution kernels in each layer and one shared convolution kernels to respectively extract individual and common features of different modalities. By applying the Dynamic Convolution Fusion Module (DCFM), the non-shared and shared convolution kernels are re-weighted dynamically. Finally, we embed the feature enhanced aggregation and dynamic kernel weights into the backbone network, which forms the FcFNet. Experimental evaluations conducted on the GTOT and RGBT234 datasets demonstrate a notable enhancement in detection performance, with an absolute improvement of 2.4% in success rate (SR) and 3.3% in precision rate (PR) when leveraging the CBFM and DCFM. The proposed FcFNet exhibits competitive performance compared to other state-of-the-art tracking algorithms, especially in continuous tracking tasks for non-cooperative targets. More information and the source code of the FcFNet will be released at https://github.com/saajoejge/FcFNet.

10 September 2025 in Remote Sensing
Verified
Journal Article
Open access
PDF file
YHYuxiang Hu
Yuxiang Hu
KDKefeng Deng
Kefeng Deng
QSQingguo Su
Qingguo Su
DZDi Zhang
Di Zhang
XSXinjie Shi
Xinjie Shi
KRKaijun Ren
Kaijun Ren
Tropical cyclones (TCs) rank among the most destructive natural hazards globally, with core damaging potential originating from regions of intense wind shear and steep wind speed gradients within the eyewall and spiral rainbands. Accurately characterizing these fine-scale structural features is therefore critical for understanding TC intensity evolution, wind hazard distribution, and disaster mitigation. Recently, the deep learning-based downscaling methods have shown significant advantages in efficiently obtaining high-resolution wind field distributions. However, existing methods are mainly used to downscale general wind fields, and research on downscaling extreme wind field events remains limited. There are two main difficulties in downscaling TC wind fields. The first one is that high-quality datasets for TC wind fields are scarce; the other is that general deep learning frameworks lack the ability to capture the dynamic characteristics of TCs. Consequently, this study proposes a novel deep learning framework, CycloneWind, for downscaling TC surface wind fields: (1) a high-quality dataset is constructed by integrating Cyclobs satellite observations with ERA5 reanalysis data, incorporating auxiliary variables like low cloud cover, surface pressure, and top-of-atmosphere incident solar radiation; (2) we propose CycloneWind, a dynamically constrained Transformer-based architecture incorporating three wind field dynamical operators, along with a wind dynamics-constrained loss function formulated to enforce consistency in wind divergence and vorticity; (3) an Adaptive Dynamics-Guided Block (ADGB) is designed to explicitly encode TC rotational dynamics using wind shear detection and wind vortex diffusion operators; (4) Filtering Transformer Layers (FTLs) with high-frequency filtering operators are used for modeling wind field small-scale details. Experimental results demonstrate that CycloneWind successfully achieves an 8-fold spatial resolution reconstruction in TC regions. Compared to the best-performing baseline model, CycloneWind reduces the Root Mean Square Error (RMSE) for the U and V wind components by 9.6% and 4.9%, respectively. More significantly, it achieves substantial improvements of 23.0%, 22.6%, and 20.5% in key dynamical metrics such as divergence difference, vorticity difference, and direction cosine dissimilarity.

3 September 2024 in IEEE Geoscience and Remote Sensing Letters
Verified
Journal Article
Closed access
QSQingguo Su
Qingguo Su
XSXinjie Shi
Xinjie Shi
WWWuxin Wang
Wuxin Wang
DZDi Zhang
Di Zhang
KDKefeng Deng
Kefeng Deng
KRKaijun Ren
Kaijun Ren
In the field of numerical weather prediction, fine-grained precipitation fields play a crucial role in forecasting and analyzing the spatial distribution and intensity of the precipitation. Historically, it is customary to employ the interpolation technique to downscale the low-resolution initial field output by assimilation systems, aligning with the requirements of a high-resolution forecasting model. Currently, data-driven deep learning methods offer novel solutions to address this challenge. In this letter, we propose a spatial downscaling algorithm for precipitation data generated from the North American Land Data Assimilation System (NLDAS), called Gaussian-Perceptual-based Generative Adversarial Network (GP-GAN). Specifically, the GP-GAN introduces a Siamese Gaussian-Perceptual Module (SGPM) which maps the data reconstructed from the generator and ground-truth to Gaussian latent space to learn the distribution of precipitation. Moreover, the Adaptive Weighted Loss Function (AWLF) is proposed to strengthen the emphasis and understanding of extreme precipitation events. Experimental results on the RainNet dataset comprising hourly precipitation over the USA demonstrate that GP-GAN provides better performance than other generative adversarial networks and diffusion models in improving spatial resolution.

10 November 2025 in IEEE Transactions on Geoscience and Remote Sensing
Verified
Journal Article
Closed access
DLDawei Li
Dawei Li
KDKefeng Deng
Kefeng Deng
DZDi Zhang
Di Zhang
HLHongze Leng
Hongze Leng
YLYudi Liu
Yudi Liu
JSJunqiang Song
Junqiang Song
Precipitation nowcasting is a long-standing challenge due to the inherent unpredictability, which often lead to significant risks and damage. Traditional approaches that model nonlinear relationships between initial and future precipitation states often fail to accurately capture precipitation dynamics, including distribution and intensity patterns. Current data-driven methods are limited in their ability to represent the chaotic nature of precipitation without guidance from physical theory. To address this, we present Rainfusion, a generative model that integrates Prandtl’s mixing length theory from fluid dynamics with computer vision diffusion models. This integration accounts for nonlinear interactions between large-scale evolution and turbulent fluctuations in precipitation, generating physically plausible predictions. Rainfusion significantly improves forecasting skill on two benchmark dataset over the next 3 hours. Furthermore, we enhance Rainfusion with a control network trained on multi-source data, particularly lightning observations, enabling more accurate and controllable predictions of precipitation’s spatial-temporal patterns. Weather forecasters can utilize Rainfusion to guide predictions toward either growth or decay based on their domain expertise. Our approach advances precipitation nowcasting, offering a robust framework that bridges physical theory with modern deep learning techniques.