Lei Chen

Dr. Lei Chen

orcid

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

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

Representative Publications

View All Publications
28 July 2020 in IEEE Transactions on Circuits and Systems I: Regular Papers
Verified
Journal Article
Closed access
CXCanran Xu
Canran Xu
ZPZizhao Peng
Zizhao Peng
XHXuanzhen Hu
Xuanzhen Hu
WZWei Zhang
Wei Zhang
LCLei Chen
Lei Chen
FAFengwei An
Fengwei An
In the natural and practical scenario, the captured video sequence under bad weather situations or low light conditions often suffers from poor visibility and low-contrast problems. This hurts the performance of the high-level processing, e.g. object tracking or recognition. In this paper, we develop an FPGA-based low-visibility enhancement accelerator for video sequence by adaptive histogram equalization with dynamic clip-threshold (AHEwDC) which is determined by the visibility assessment. The main goal is to improve the low visibility with high image quality for both hazy and low-light video sequences in real-time. Firstly, a concept to quantify the visual perception based on supervised learning is to estimate the visibility score. Then, to avoid the problem of noise amplification in the conventional method, we propose a visibility assessment model to find an optimal clip-threshold. The contrast energy of gray channel, yellow-blue channel and red-green channel, average saturation, and gradients are statistical features in the model to describe the visibility of an image. Finally, to meet the speed requirement for video sequence processing, a specified hardware architecture for both visibility assessment and AHEwDC is implemented on FPGA. Besides, a mean spatial filter for cumulative distribution functions (CDFs) of the AHE is developed for suppressing the noise caused by a single-color local region. The demonstration system on the DE1-SoC platform with the Intel Cyclone V FPGA device with the max working frequency of 75.84 MHz is capable of processing 30 fps FHD (1920x 1080) video.

3 August 2021 in IEEE Transactions on Circuits and Systems I: Regular Papers
Verified
Journal Article
Closed access
PDPingcheng Dong
Pingcheng Dong
ZCZhuoyu Chen
Zhuoyu Chen
ZLZhuoao Li
Zhuoao Li
YFYuzhe Fu
Yuzhe Fu
LCLei Chen
Lei Chen
FAFengwei An
Fengwei An
The semi-global matching (SGM) algorithm in stereo vision is a well-known depth-estimation method since it can generate dense and robust disparity maps. However, the real-time processing and low power dissipation, the specifications of the Internet-of-Thing (IoT) applications, are challenging for their computational complexity. In this paper, we propose a hardware-oriented SGM algorithm with pixel-level pipeline and region-optimized cost aggregation for high-speed processing and low hardware-resource usage. Firstly, the matching costs in a region are integrated with an optimization strategy to significantly reduce memory usage and improve the processing speed of the cost aggregation. Then, a two-layer parallel two-stage pipeline (TPTP) architecture, which enables pixel-level processing, is designed to calculate two directions (0° and 135°) aggregation to further solve the crucial computational bottleneck of the SGM algorithm. Finally, the architecture is demonstrated on a low-cost XILINX Spartan-7 device and an advanced Stratix-V FPGA device for VGA ( $640\times 480$ ) depth estimation. The experimental results show that the proposed architecture with compact hardware architecture also ensures accuracy. The pixel-level pipeline architecture enables a processing speed of 355 frames per second (fps) at 109MHz on the Spartan-7 FPGA device and 508 fps at 156MHz on the Stratix-V FPGA. Besides, the coprocessor respectively achieves an energy efficiency of 4.74 nJ/pixel with a power dissipation of 517mW and 4.29nJ/pixel with a power dissipation of 669mW on these two FPGAs.

16 August 2022 in IEEE Transactions on Circuits and Systems II: Express Briefs
Verified
Journal Article
Closed access
YTYonghao Tan
Yonghao Tan
HDHuanshihong Deng
Huanshihong Deng
MSMengying Sun
Mengying Sun
MZMinghao Zhou
Minghao Zhou
YCYifei Chen
Yifei Chen
FAFengwei An
Fengwei An
Much effort overlaps in designing different hardware to implement different Simultaneous localization and mapping (SLAM) algorithms. In this brief, a reconfigurable architecture with dedicated instruction sets allows the coprocessor to satisfy the pose estimation of a sample class of the SLAM algorithms, feature-based or learning-based methods, which can be decomposed to basic common operations. Furthermore, a memory-reused strategy in instructions was designed to avoid the demand for temporary memory for complex operations. Finally, two parallel computing cores are implemented to perform matrix operations and special computation about the pose estimation in floating-point and fixed-point arithmetic. These contribute to the low hardware resource usage and memory requirement, as illustrated in the experimental results.

20 July 2020 in IEEE Transactions on Circuits and Systems II: Express Briefs
Verified
Journal Article
Closed access
ZXZhihua Xiao
Zhihua Xiao
PXPeng Xu
Peng Xu
XWXianglong Wang
Xianglong Wang
LCLei Chen
Lei Chen
FAFengwei An
Fengwei An
A critical mission for mobile robot vision is to detect and classify different objects with low power consumption. In this paper, a multi-class object detection coprocessor is proposed by using the Histogram of Oriented Gradient (HOG) and Local Binary Pattern (LBP) together with a weighted Softmax classifier. Its architecture is compact and hardware-friendly suitable for energy-constrained applications. Firstly, the cell-based feature extraction unit and block-level normalization reuse the SRAMs for storing one-row cell and one-row block. Meanwhile, the working frequency of the feature extraction and block-normalization unit is synchronized to the image sensor for low dynamic power. Then, a block-level one-time sliding-detection-window (OTSDW) mechanism is developed for partial classification with scalable object size. Finally, the Softmax classifier, which is implemented by the look-up table, linear fitting methods, and the fixed-point number, is tested in the Fashion MNIST dataset to evaluate its performance in multi-class classification problems and it reached an accuracy of over 86.2% with 10,180 parameters. The experimental result shows that the hardware-resource usage of the FPGA implementation is capable of 60 fps VGA video with 80.98 mW power consumption. This method uses similar or even fewer hardware resources than that of previous work using only the HOG feature and single-class classifier.

8 November 2022 in Sensors
Verified
Journal Article
Open access
PDF file
YMYunhao Ma
Yunhao Ma
XFXiwei Fang
Xiwei Fang
XGXinyu Guan
Xinyu Guan
KLKe Li
Ke Li
LCLei Chen
Lei Chen
FAFengwei An
Fengwei An
Binocular stereoscopic matching is an essential method in computer vision, imitating human binocular technology to obtain distance information. Among plentiful stereo matching algorithms, Semi-Global Matching (SGM) is recognized as one of the most popular vision algorithms due to its relatively low power consumption and high accuracy, resulting in many excellent SGM-based hardware accelerators. However, vision algorithms, including SGM, are still somewhat inaccurate in actual long-range applications. Therefore, this paper proposes a disparity improvement strategy based on subpixel interpolation and disparity optimization post-processing using an area optimization strategy, hardware-friendly divider, split look-up table, and the clock alignment multi-directional disparity occlusion filling, and depth acquisition based on floating-point operations. The hardware architecture based on optimization algorithms is on the Stratix-IV platform. It consumes about 5.6 K LUTs, 12.8 K registers, and 2.5 M bits of on-chip memory. Meanwhile, the non-occlusion error rate of only 4.61% is about 1% better than the state-of-the-art works in the KITTI2015 dataset. The maximum working frequency can reach up to 98.28 MHz for the 640 × 480 resolution video and 128 disparity range with the power dissipation of 1.459 W and 320 frames per second processing speed.

4 March 2022 in Sensors
Verified
Journal Article
Open access
PDF file
YLYiwei Luo
Yiwei Luo
PLPing Li
Ping Li
GSGang Shi
Gang Shi
ZLZuowei Liang
Zuowei Liang
LCLei Chen
Lei Chen
FAFengwei An
Fengwei An
Vision-based Lane departure warning system (LDWS) has been widely used in modern vehicles to improve drivability and safety. In this paper, a novel LDWS with precise positioning is proposed. Calibration strategy is first presented through a 3D camera imaging model with only three parallel and equally spaced lines, where the three angles of rotation for the transformation from the camera coordinate system to the world coordinate system are deduced. Then camera height is calculated compared to the previous works using a measured one with potential errors. A criterion for lane departure warning with only one of the two lane-markings is proposed to estimate both yaw angle and distance between the lane-markings and the vehicle. Experiments show that calibration strategy can be easily set up and achieve an average of 98.95% accuracy on the lane departure assessment.