Yuchu He

Yuchu He

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
3 April 2024 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Verified
Journal Article
Open access
JJJia Jia
Jia Jia
GLGeunho Lee
Geunho Lee
ZWZHIBO WANG
ZHIBO WANG
ZLZhi Lyu
Zhi Lyu
YHYuchu He
Yuchu He
In recent years, the application of deep learning to change detection (CD) has significantly progressed in remote sensing images. CD tasks have mostly used architectures such as CNN and Transformer to locate image changes. However, these architectures have shortcomings in representing boundary details and are prone to false alarms and missed detections under complex lighting and weather conditions. For that, we propose a new network, Siamese Meets Diffusion Network (SMDNet), a CD model that combines discriminative and generative architecture. By leveraging the power of the Siam-U2Net Feature Differential Encoder (SU-FDE) and Denoising Diffusion Implicit Model (DDIM), it not only improves the accuracy of object edge detection but also enhances the data through iterative denoising and thinning reconstruction Detail detection accuracy. Improves the model's robustness under environmental changes. First, we propose an SU-FDE module that uses shared weight features to capture differences between time series images, refine edge detection, and combine it with the attention mechanism to identify vital coarse features, thereby improving model sensitivity and accuracy. Finally, the progressive sampling of DDIM is used to integrate further these key features, and the adaptability of the model in different environments is enhanced with the help of the denoising ability of the diffusion model and the accurate capture of the probability distribution of image data. The performance evaluation of SMDNet on LEVIR-CD, DSIFN-CD, and CDD datasets yields validated F1 scores of 89.17%, 88.48%, and 88.23%, respectively. This substantiates the advanced capabilities of our model in accurately identifying variations and intricate details.

9 March 2018 in IEEE Access
Verified
Journal Article
Open access
YHYuchu He
Yuchu He
DSDihua Sun
Dihua Sun
MZMin Zhao
Min Zhao
SCSenlin Cheng
Senlin Cheng
The vicinity of traffic signals is one of the most critical areas in road systems. New information technology paradigms like vehicle-to-vehicle and vehicle-to-infrastructure communication are applied to improve traffic operations at intersections; for example, a typical cyber-physical system enables efficient traffic state estimation and traffic modeling. A new model of vehicle cooperative driving under a typical scenario vicinity of traffic signals and Vehicle to X environment was proposed in this paper. Based on the intelligent driver model, the model planned trajectories for vehicles in the vicinity of traffic signals in advance to reduce stopping frequency and travel time and improve the throughput of the road according to traffic conditions, such as signal cycle state, the distance to traffic signals, and the situation regarding adjacent vehicles. In accordance with the relationship between the host vehicle and the surrounding vehicles, the model also planed the cooperative lane changing strategy in this scenario to improve safety and comfort in the process of lane changing. A simulation experiment compared the proposed model with the traditional intelligent driver model, and analyzed its performance in different traffic conditions. The feasibility and superiority of the model were confirmed.

4 October 2021 in IEEE Transactions on Intelligent Transportation Systems
Verified
Journal Article
Closed access
YHYuchu He
Yuchu He
ZJZhijuan Jia
Zhijuan Jia
MHMingsheng Hu
Mingsheng Hu
CCChi Cui
Chi Cui
YCYage Cheng
Yage Cheng
YYYanyan Yang
Yanyan Yang
Controller area network (CAN) is the most commonly used bus technology for In-vehicle network and uses multicast communication without corresponding security measures. Therefore, the message data field is vulnerable to tampering and other attacks. Recent machine learning-based intrusion detection methods for CAN bus messages only use the information contained in the message data field and do not take into account the contribution made by the neighboring information of CAN bus messages. In addition, previous models considered the data domain information of CAN bus messages as separate features and did not consider the unique weight of each feature, as well as the second-order interaction information between features. Therefore, we propose a novel intrusion detection model, The Hybrid Similar Neighborhood Robust Factorization Machine Model (HSNRFM), for detecting anomalies in CAN bus messages to address the shortcomings and problems of the previous models. To be able to incorporate the contribution of similar neighborhood information and learn the unique weight parameters of each feature in the model decision process, as well as additional second-order interaction information between features, the HSNRFM model solves the above problem using a similarity calculation method and a factorization machine model. Comprehensive experimental results are compared on real vehicle datasets. The HSNRFM model has AUC values of 0.9216 and 0.901 and AUPR values of 0.9194 and 0.9018 on two real datasets, respectively. And the results show that our proposed HSNRFM model has excellent detection efficiency for intrusion detection of CAN bus messages.

19 November 2021 in Security and Communication Networks
Verified
Journal Article
Open access
PDF file
YCYage Cheng
Yage Cheng
BGBei Gong
Bei Gong
ZJZhijuan Jia
Zhijuan Jia
YYYanYan Yang
YanYan Yang
YHYuchu He
Yuchu He
XZXiaofei Zhang
Xiaofei Zhang
In this article, we analysed the problems of electronic medical records (EMRs) and found that the EMRs generated by different hospitals for the same patient are mutually independent and duplication and data sharing are difficult among hospitals. In order to solve this problem, this paper proposes an efficient and secure cross-domain sharing scheme of EMRs based on edge computing. The program allows the doctor to access the personal history EMRs through the patient’s authorization so that the doctor can understand the patient’s history of illness and, on this basis, generate a new medical record for the patient. Then, the doctor sends the EMRs to the edge server, and the server calculates the ciphertext and adds it to the patient’s personal medical record to complete the case update. Analysis shows that this solution can effectively prevent data tampering and forgery through blockchain and avoid privacy leakage problems in plaintext sharing by using searchable encryption and by relying on edge servers to solve nearby computing tasks and divert the computing capacity of cloud servers to improve efficiency. The security proof shows that the scheme satisfies the complex problem of the BDH assumption. Performance analysis shows that the scheme is feasible and efficient.

28 November 2023 in IEEE Access
Verified
Journal Article
Open access
CCChi Cui
Chi Cui
Haiping DuHaiping Du
Haiping Du
ZJZhijuan Jia
Zhijuan Jia
XZXiaofei Zhang
Xiaofei Zhang
YHYuchu He
Yuchu He
YYYanyan Yang
Yanyan Yang
As a state-of-the-art distributed learning approach, federated learning has gained much popularity in connected and autonomous vehicles (CAVs). In federated learning, models are trained locally, and only model parameters instead of raw data are exchanged to aggregate a global model. Compared with traditional learning approaches, the enhanced privacy protection and relieved network bandwidth provided by federated learning make it more favorable in CAVs. On the other hand, poisoning attack, which can break the integrity of the trained model by injecting crafted perturbations to the training samples, has become a major threat to deep learning in recent years. It has been shown that the distributed nature of federated learning makes it more vulnerable to poisoning attacks. In view of this situation, the strategies and attacking methods of the adversaries are worth studying. In this paper, two novel optimization-based black-box and clean-label data poisoning attacking methods are proposed. Poisoning perturbations are generated using particle swarm optimization hybrid with simulated annealing and genetic algorithm respectively. The attacking methods are evaluated by experiments conducted on the example of traffic sign recognition system on CAVs, and the results show that the prediction accuracy of the global model is significantly downgraded even with a small portion of poisoned data using the proposed methods.

24 August 2018 in IET Intelligent Transport Systems
Verified
Journal Article
Open access
PDF file
DSDihua Sun
Dihua Sun
YHYuchu He
Yuchu He
MZMin Zhao
Min Zhao
SCSenlin Cheng
Senlin Cheng
The vicinity of traffic signals is one of the most special and critical areas in the whole road system. Considering that the driver can obtain traffic signals information through two approaches: vision and the vehicle‐to‐vehicle (V2X) communication equipment in the vehicle, this study proposes two vehicle cooperative driving models that apply to the vicinity of traffic signals: the intelligent driver model (IDM) in the vicinity of traffic signals (IDM‐VT) and IDM in the vicinity of traffic signals under V2X environment (IDM‐VT‐V2X). These two models are both based on the intelligent driver model. In accordance with different situations of vehicles in the vicinity of traffic signals, such as distance from the traffic lights, whether there was another vehicle in front and different states of traffic lights, the pertinent analysis was conducted, and the cooperative driving strategy that satisfied the complicated situations of vehicle was proposed. These two models were verified and compared in the simulation experiment. The results of simulation show that when comparing with the IDM‐VT, the IDM‐VT‐V2X can reduce the average travel time and the average stop delay time by 12.98 and 98.32%, respectively, and it can also reduce 22.53% fuel consumption.