Yucheng Jiang

Mr. Yucheng Jiang

orcid

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

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

Representative Publications

View All Publications
9 February 2024 in IEEE Transactions on Circuits and Systems II: Express Briefs
Verified
Journal Article
Closed access
YJYucheng Jiang
Yucheng Jiang
YYYu Yu
Yu Yu
HLHan Li
Han Li
ZDZehua Dong
Zehua Dong
SMSongping Mai
Songping Mai
High-precision stereo matching implementation requires substantial computational resources. To achieve precise real-time depth estimation on embedded devices, we propose a hardware accelerator for local-expansion-move (LE) that can be utilized for accurate disparity extraction. By inferring per-pixel 3D plane labels, it is able to obtain smooth disparities. With the efficient multi-level pipeline design and sophisticated memory structures, we have implemented two key techniques, localization and spatial propagation, which significantly enhance the efficiency and precision of the inference. Compared to the software implementations of LE on CPU and GPU, the accelerator exhibits speedups of 524x and 27.5x respectively. It achieves state-of-the-art matching accuracy in embedded systems by integrating with a lightweight neural network, while maintaining quasi-real-time efficiency. Moreover, the accelerator is implemented using high-level synthesis, which allows for easy modification of accelerator parameters based on the image resolutions and performance requirements in different applications.

14 November 2024 in IEEE Sensors Journal
Verified
Journal Article
Closed access
JYJie Yu
Jie Yu
SMSongping Mai
Songping Mai
YXYuxiang Xu
Yuxiang Xu
YJYucheng Jiang
Yucheng Jiang
The structured light three-dimensional measurement technology, which combines phase-shift and gray code, is widely adopted in industrial measurement and quality inspection due to its features such as high accuracy, rapid phase unwrapping, robustness, and suitability for complex shapes. However, the requirement for projecting multiple frames restricts its application in real-time systems. Furthermore, it cannot simultaneously yield high-accuracy measurement results for complex scenes comprising static and dynamic objects, without introducing any motion-related artifacts. Therefore, this paper introduces a real-time, fast, and highly accurate single-camera structured light 3D measurement system to address these challenges. This system employs a strategy of cross-projecting phase-shifting and gray-coded patterns to enhance the 3D measurement frame rate. It also introduces an effective two-dimensional phase misalignment correction method to address the issue of misalignment between gray code and phase-shift at the boundaries. By comparing the phase differences among different combinations of phase-shifting patterns, the system can distinguish between static and dynamic areas with a small time span. Subsequently, during the decoding process, the standard method and a motion compensation approach are respectively applied to static and dynamic regions. Experimental results demonstrate that our system achieves a point cloud frame rate of 30fps, which is five times higher than the sequential projection solution and notably enhances the measurement accuracy of dynamic areas by employing motion compensation, all while maintaining accurate measurements in static areas.

2 July 2024 in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
Verified
Journal Article
Closed access
ZDZehua Dong
Zehua Dong
BZBoyu Zhang
Boyu Zhang
YJYucheng Jiang
Yucheng Jiang
YYYu Yu
Yu Yu
HLHan Li
Han Li
SMSongping Mai
Songping Mai
The push-relabel algorithm is recognized as one of the efficient algorithms in the field of graph cut, finding widespread applications in computer vision. While its pixel-level parallel implementations are prevalent, existing methods predominantly rely on checkerboard scheduling, imposing inherent constraints on neighborhood size, limited to four. This limitation compromises both algorithm precision and efficiency, hindering real-time and high-precision applications. To address these issues, this article introduces a novel approach to accelerate push-relabel algorithm implementation on FPGA in a more universal and efficient manner, supporting variable-sized image block operations. First, by introducing the deferred update strategy, we realize the pulse-enhanced parallel push-relabel (PEPPR) algorithm to address data contention and conflict in parallel processing. Second, the simultaneous weighted push method is proposed, further enhancing parallel operations. Lastly, we introduce the efficient diffusion wave search (DWS) algorithm to expedite algorithm convergence and reduce redundancy. While achieving a modest $1.7\times $ acceleration compared to state-of-the-art implementations, the proposed algorithm (PEPPR-DWS) successfully overcomes the inherent limitations of checkerboard scheduling in full pixel-level parallelism. In the test based on Middlebury benchmark V3, the proposed 8-neighborhood implementation exhibits a reduction of error rate by over 1% compared to the typical 4-neighborhood implementation. It provides a versatile and efficient solution for high-precision and real-time applications, holding substantial potential for practical applications.

24 March 2023 in Sensors
Verified
Journal Article
Open access
PDF file
YJYucheng Jiang
Yucheng Jiang
ZDZehua Dong
Zehua Dong
SMSongping Mai
Songping Mai
Stereo matching in binocular endoscopic scenarios is difficult due to the radiometric distortion caused by restricted light conditions. Traditional matching algorithms suffer from poor performance in challenging areas, while deep learning ones are limited by their generalizability and complexity. We introduce a non-deep learning cost volume generation method whose performance is close to a deep learning algorithm, but with far less computation. To deal with the radiometric distortion problem, the initial cost volume is constructed using two radiometric invariant cost metrics, the histogram of gradient angle and amplitude descriptors. Then we propose a new cross-scale propagation framework to improve the matching reliability in small homogenous regions without increasing the running time. The experimental results on the Middlebury Version 3 Benchmark show that the performance of the combination of our method and Local-Expansion, an optimization algorithm, ranks top among non-deep learning algorithms. Other quantitative experimental results on a surgical endoscopic dataset and our binocular endoscope show that the accuracy of the proposed algorithm is at the millimeter level which is comparable to the accuracy of deep learning algorithms. In addition, our method is 65 times faster than its deep learning counterpart in terms of cost volume generation.

25 December 2024 in IEEE Transactions on Instrumentation and Measurement
Verified
Journal Article
Closed access
JHJunwei Hu
Junwei Hu
YJYucheng Jiang
Yucheng Jiang
YYYongliang Ye
Yongliang Ye
SMSongping Mai
Songping Mai
Phase-shifting profilometry (PSP) is extensively utilized in three-dimensional measurements. While grayscale-based PSP ensures high measurement accuracy, it suffers from inefficiency. Conversely, color-based PSP enhances efficiency but compromises on accuracy and robustness. To promote both measurement accuracy and efficiency, we propose an orthogonal self-equalizing color fringe phase-shifting profilometry (OSCF-PSP) technique. This method employs orthogonal fringes to achieve precise channel decrosstalk while minimizing the condition number of the decrosstalk matrix, thereby improving noise immunity. Moreover, by leveraging the orthogonal nature of fringes, we develop a signal-to-noise ratio estimator independent of the measured object’s surface shape, texture information, and background illumination. This enables quantitative assessment of the reliability of each channel’s computed results and dynamic adaptive weighting of channel outputs, thereby achieving immunity to the object’s colored texture. Compared to other color fringe profilometry methods, OSCF-PSP exhibits superior robustness, being insensitive to colored texture variations of the object, cross-channel interference, and projector illumination imbalances. In scenarios where efficiency improves by 30% (3 patterns to 2 patterns)over conventional three-step grayscale PSP, most measurement errors decrease by 32%(0.0717 to 0.0488 Rad,measuring white objects)-50%(0.1242 to 0.0606 Rad,measuring color objects). Even in the worst-case scenario of our algorithm, measurement errors do not exceed those of the classical three-step phase-shifting method. Our proposed approach effectively mitigates the drawbacks of color-based PSP while maintaining its relative efficiency.

30 September 2021
Verified
Conference Paper
Open access
RSRuoxu Sun
Ruoxu Sun
JZJinyu Zhan
Jinyu Zhan
WJWei Jiang
Wei Jiang
YJYucheng Jiang
Yucheng Jiang
In this paper, we approach to improve the fault tolerance of Deep Neural Networks (DNNs) for safety-critical artificial intelligent applications. We propose to remap the range of 32-bit float to weights to reduce the influence of invalid weights caused by bit-flip faults. From preliminary experiments, we observe that weakening bit-flip faults which make positive weights larger can help to improve the reliability of DNNs. Then, we propose a gaussian distribution based mapping method to prevent weights from being influenced by bit-flip faults, in which a novel function is formulated to remap the relation between 32-bit float and the values of weights. Extensive experiments demonstrate that our approach can improve the accuracy of VGG16 from 13.5% to 80.5%, which is better than the other six tolerance approaches of DNNs.