Tieding Lu

Tieding Lu

Auto-generated profile, based on publicly available information

Tieding Lu

Key Laboratory of Mine Environmental Monitoring and Improving around Poyang Lake of Ministry of Natural Resources, East China University of Technology, Nanchang, China

Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

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

Representative Publications

View All Publications
6 October 2025 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Verified
Journal Article
Open access
XZXiufang Zhou
Xiufang Zhou
ZLZhuotao Liu
Zhuotao Liu
XGXunqiang Gong
Xunqiang Gong
SQShunan Qin
Shunan Qin
TLTieding Lu
Tieding Lu
YZYanfei Zhong
Yanfei Zhong
As the core carriers of human activities, buildings represent not only the fundamental components of urban spatial structures but also serve critical functions in global resource management, urban planning decisions, disaster risk assessment, and the monitoring of sustainable development. Consequently, they constitute significant features with substantial value in the analysis and application of remote sensing imagery. The issues of mistake and omission extraction and blurred margins that are caused by the UNet's insufficient utilization of features in different scales prompt the proposal of an improved UNet, i.e., RFTransUNet, which is in support of the Feature Cross Transformer block based on residual network and Vision Transformer. This net, based on UNet, takes residual blocks as the network backbone, uses the FTrans block in the skip connection part to conduct multi-scale feature fusion, and adopts the Feature Pyramid Network as deep supervision in training. Among them, the encoder and decoder based on residual blocks can better retain semantic information when extracting detailed features of images, the FTrans block fuses shallow detailed information and deep semantic information, and the FPN introduces reference labels to each layer of the network during training. The contrast experiments, aiming at verifying the proposed method, are conducted on two publicly available datasets and a self-built dataset. Versus other comparative methods, the proposed method has clearer and more accurate extracted results with inconspicuous error extraction and better marginal maintenance. The Intersection of Union (IoU) of the public satellite and aerial imagery datasets and the self-built UAV imagery dataset achieves 71.7862%, 90.6190%, and 84.7210%, respectively. The link to the code and dataset is available at https://github.com/RSIDEA-ECUT/RFTransUNet.

21 March 2026 in SSRN Electronic Journal
Verified
Preprint
Open access
HYHouming Yang
Houming Yang
BJBing Ji
Bing Ji
TZTao Zhang
Tao Zhang
ZLZhaoquan Li
Zhaoquan Li
TLTieding Lu
Tieding Lu
MTMiao Tang
Miao Tang
Aiming at the problems that complex noise in gravity and gravity gradient data restricts interpretation accuracy, and traditional denoising methods are low in precision and prone to causing loss of effective signals, we applied a residual learning-based DnCNN deep learning method to denoise gravity and gravity gradient data. This model employs residual learning and batch normalization techniques to directly estimate noise residuals from noisy data, enabling effective separation of signal and noise. The method was compared against two-dimensional variational mode decomposition (2D-VMD), bidimensional empirical mode decomposition (BEMD), bidimensional wavelet denoising (BWD), two-dimensional singular spectrum analysis (2D-SSA), IIR filter and Gaussian filter. The study also analyzed the impact of different wavelet bases, threshold selection methods, and threshold types on the performance of BWD. Two sets of simulation experiments and multiple sets of experimental data collected under different noise levels results demonstrate that the residual learning-based DnCNN method achieves significantly superior performance across four evaluation metrics—root mean square error (RMSE), mean absolute error (MAE), peak signal-to-noise ratio (PSNR), and structural similarity index (SSIM) compared to the other six methods. It effectively suppresses noise while preserving the valid signals and local detail features in the original data more completely. In gravity gradient data denoising, data processed by this method still satisfies the Laplace equation constraint.

18 October 2026 in SSRN Electronic Journal
Verified
Preprint
Open access
HYHouming Yang
Houming Yang
BJBing Ji
Bing Ji
SBShaofeng Bian
Shaofeng Bian
LWLin Wu
Lin Wu
TLTieding Lu
Tieding Lu
JZJingwen Zong
Jingwen Zong
Shipborne gravimetry is an important means of obtaining high-precision marine gravity field information. However, disturbances from the complex marine environment introduce intense non-stationary noise into gravimeter observations, severely limiting the measurement accuracy. To address this issue, we use simulated and field- measured data from the LaCoste Sea-III gravimeter and systematically study error analysis, correction processing, and adaptive filtering methods. Based on the analysis of main error sources, several corrections are completed, including normal gravity, Eötvös effect, and drift corrections. We propose a joint denoising method combining crested porcupine optimizer-optimized variational mode decomposition with multiscale permutation entropy (CPO-VMD-MPE). The CPO globally optimizes the number of modes and penalty factor of the VMD, while the MPE quantifies the complexity of each decomposed mode to distinguish signal from noise. The signal-dominated components are smoothed by a Savitzky-Golay (SG) filter and reconstructed to achieve effective extraction of the gravity signal. PSO, GOA, and SSA algorithms are introduced for comparison of optimization performance. The results show that the CPO converges faster and achieves higher accuracy than PSO, GOA, and SSA. In both the simulation and sea trials, the proposed method yields lower mean and standard deviation of residuals than existing methods, demonstrating stronger noise suppression and signal preservation capabilities, with its denoising performance verified by multiple evaluation metrics.