Zihan Deng

Zihan Deng

Auto-generated profile, based on publicly available information

Zihan Deng

Center of Ultra-Precision Optoelectronic Instrument Engineering and the Key Laboratory of Ultra-Precision Intelligent Instrumentation, Ministry of Industry and Information Technology, Harbin Institute of Technology, Harbin, China

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

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

Representative Publications

View All Publications
11 May 2025 in Expert Systems with Applications
Verified
Journal Article
Closed access
ZDZihan Deng
Zihan Deng
ZWZhisheng Wang
Zhisheng Wang
YSYuanlin Shan
Yuanlin Shan
GHGuohang He
Guohang He
TDTiantian Du
Tiantian Du
SWShunli Wang
Shunli Wang

15 July 2024 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
ZWZhisheng Wang
Zhisheng Wang
HYHaijun Yu
Haijun Yu
YHYixing Huang
Yixing Huang
SWShunli Wang
Shunli Wang
SNSong Ni
Song Ni
JCJunning Cui
Junning Cui
Micro-computed tomography (micro-CT) is a widely used state-of-the-art instrument employed to study the morphological structures of objects in various fields. However, its small field-of-view (FOV) cannot meet the pressing demand for imaging relatively large objects at high spatial resolutions. Recently, we devised a novel scanning mode called multiple source translation CT (mSTCT) that effectively enlarges the FOV of the micro-CT and correspondingly developed a virtual projection-based filtered backprojection (V-FBP) algorithm for reconstruction. Although V-FBP skillfully solves the truncation problem in mSTCT, it requires densely sampled projections to arrive at high-resolution reconstruction, which reduces imaging efficiency. In this paper, we develop two backprojection-filtration (BPF)-based algorithms for mSTCT—S-BPF (derivatives along source) and D-BPF (derivatives along detector). D-BPF can achieve high-resolution reconstruction with fewer projections than V-FBP and S-BPF. Through simulated and physical experiments conducted in this paper, we demonstrate that D-BPF can dramatically reduce the number of projection views compared with V-FBP at the same spatial resolution, which makes mSTCT more feasible in practice. Meanwhile, S-BPF can yield more stable results than D-BPF, which is similar to V-FBP.

15 April 2024 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
ZWZhisheng Wang
Zhisheng Wang
ZDZihan Deng
Zihan Deng
FLFenglin Liu
Fenglin Liu
YHYixing Huang
Yixing Huang
HYHaijun Yu
Haijun Yu
JCJunning Cui
Junning Cui
Recently, in some special situations, linear computed tomography (LCT) systems have provided better imaging solutions for internal structures compared with conventional CT instruments. To image the region of interest (ROI) and weaken projection truncation in the LCT reconstruction, the backprojection filtration (BPF) algorithm is an effective solution; however, in BPF for LCT, it is difficult to achieve stable interior reconstruction, and for differentiated backprojection (DBP) images of LCT. To satisfy multiple reconstruction scenarios for LCT, including interior ROI, complete object, and exterior region, we propose two types of reconstruction architectures. The first overlays multiple DBP images to obtain a complete DBP image and then uses a network to learn the overlying or composite Hilbert filtering function, referred to as the overlay-single network (OSNet). The second uses multiple networks to train different directional Hilbert filtering models for DBP images of multiple linear scanning trajectories, respectively, and then overlays the reconstructed results, i.e., multiple networks overlaying (MNetO). In two architectures, we introduce a Swin transformer (ST) block to the generator of pix2pixGAN to extract both local and global features from DBP images at the same time. We investigate the proposed architectures from different networks, FOV sizes, pixel sizes, number of projections, geometric magnification, and computation time. The two architectures are proven to recover images. OSNet outperforms BPF in multiple scenarios. In OSNet, ST-pix2pixGAN is superior to pix2pixGAN and CycleGAN for the different networks. The average peak signal-to-noise ratio (PSNR) value reaches 41.1949 dB, which improves by 31% compared to BPF. The average structural similarity index measure (SSIM) and root-mean-square error (RMSE) are improved at 0.9996 and 0.0093, respectively. In MNetO, one model in a certain direction can increase the PSNR by about 18.4% compared to BPF, enabling high-quality imaging of the exterior edge. Our architectures can provide great solutions or references for LCT high-resolution imaging and nondestructive measuring.

13 November 2024 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
ZWZhisheng Wang
Zhisheng Wang
ZDZihan Deng
Zihan Deng
LLLegeng Lin
Legeng Lin
SLShangyu Li
Shangyu Li
SWShunli Wang
Shunli Wang
JCJunning Cui
Junning Cui
For a linear scanning mode—multiple Source Translation Computed Tomography (mSTCT)—when extending a large field-of-view (FOV) with a high geometry magnification, we propose some practical filtered backprojection type reconstruction algorithms. More specially, we formulate a fan-beam Differentiation, Hilbert transform, and Backprojection reconstruction formula for STCT, whose filter is along the detector rather than the source (referred to as D-DHB) to avoid acquiring large amounts of projections, and then generalize it to the cone-beam scene. On the one hand, we apply D-DHB in the full-scan mSTCT (F-mSTCT) mode. On the other hand, we eliminate the unstable position-dependent backprojection weight in the D-DHB formula by introducing an additional weight function. In fact, the backprojection weight of analytical algorithms with filtering along the detector implies the distance between the reconstructed point and the X-ray source focus. Especially where the extended FOV is very close to the source trajectory, removing the position-dependent backprojection weight can avoid some intolerable artifacts around the reconstructed image and improve computation efficiency without losing more accuracy. To meet the requirement of extremely extended FOV imaging, we fully evaluated our two methods from different aspects of the experiments. The results, overall, reveal that our methods are the most practical for mSTCT.

12 November 2024 in Higher Education Research
Verified
Journal Article
Open access
PDF file
ZDZihan Deng
Zihan Deng
TDTiantian Du
Tiantian Du
KDKanghui Deng
Kanghui Deng
LLLegeng Lin
Legeng Lin
SLShangyu Li
Shangyu Li
ZWZhisheng Wang
Zhisheng Wang
This study aims to enhance the learning motivation of engineering graduate students, particularly those specializing in Instrument Science and Engineering. It does so by evaluating the impact of a combined teaching approach that integrates multiple innovative methods. A total of 156 master’s students and 44 doctoral students were randomly assigned to either an experimental group or a control group. The experimental group engaged in small-group discussions, personalized online courses, and flipped classrooms, while the control group followed traditional teaching methods. Over a 20-week intervention period, several aspects were measured, including changes in learning motivation, teacher perceptions, satisfaction of basic psychological needs, and academic emotions. The results clearly indicate that the intervention significantly reduced non-regulatory behaviors and led to improvements in students’ self-regulation, decision-making abilities, and overall academic performance. Furthermore, students in the experimental group demonstrated superior outcomes in various critical areas such as research skills, engineering thinking, communication, and cooperation, all of which were reflected in their test scores when compared to the control group. Additionally, achievement test results in the small-group discussion model showed a negative correlation with class size, implying that the effectiveness of this method may vary depending on student personalities and group dynamics. The study concludes that small-group discussions positively influence engineering thinking, enhance goal clarity, and foster both student initiative and motivation. This teaching approach effectively meets the students’ needs for independent problem-solving and substantially enhances learning motivation, thereby supporting the educational goals in graduate studies related to Instrument Science and Engineering.

10 January 2024 in TechRxiv
Verified
Preprint
Open access
PDF file
ZDZihan Deng
Zihan Deng
ZWZhisheng Wang
Zhisheng Wang
LLLegeng Lin
Legeng Lin
SWShunli Wang
Shunli Wang
JCJunning Cui
Junning Cui
Sparse-view computed tomography (CT) significantly reduces radiation doses to the human body, whereas its analytical reconstruction exhibits severe streak artifacts. Recently, deep learning methods have shown exciting effects in CT reconstruction. The Dual-Domain (DuDo) deep learning method is one of the representative methods, and it can process the information in both the sinogram and image domains. However, the existing DuDo methods do not pay enough attention to the allocation of training costs and strategies for the two domains. In this paper, we propose a Computation-Overhead Optimization (COO) DuDo training strategy for sparse-view CT reconstruction, i.e., COO-DuDo. The training ratio of different domains is controlled by calculating their computation overhead, loss, and gradient variation of the loss. To make our COO-DuDo strategy enable sparse-view CT reconstruction better, we adopt a DuDo-Network (COO-DDNet) structure based on two coding-decoding-type subnetworks. As specific contributions, we design a Multilevel Cross-domain Connection (MCC) method to connect the decoding layers of the same scale in the two subnetworks and adopt the two-channel method for upsampling, which enables more fine-grained control of model updates and suppresses checkerboard artifacts. The evaluation results validate the effectiveness of our training strategy and methods. The peak signal-to-noise ratio (PSNR) and structural similarity (SSIM) of the reconstruction results increased by 38.8% and 0.37%, respectively, and the model convergence time decreased by 11.8%. Our research provides a broader perspective for dual-domain image restoration tasks from the perspective of computational overhead.