HZHonghao Zhu

Honghao Zhu

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Honghao Zhu has a total of 50 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Integration of Variational Methods and Deep Learning for Image Processing: A Task-Oriented Review

    by Lu Tan, Hong-Hao Zhu, Jian-Nan Xiao, Wan-Quan Liu

    Variational methods and deep learning represent two distinct paradigms that have each shaped the development of modern image processing. Variational formulations are grounded in carefully designed energy functionals and partial differential equations (PDEs) that encode prior knowledge about image structure; they offer interpretability, theoretical guarantees, and strong performance. Deep learning, by contrast, learns image priors implicitly from data through highly expressive parametric models and has achieved impressive improvements across a wide range of tasks. Over roughly the past decade, a growing body of work has sought to integrate the two paradigms, combining the mathematical rigor of variational models with the representational power of deep networks. This article provides a task-oriented review of this integration. We survey the principal mathematical frameworks—total variation (TV) and its generalisations, the Mumford–Shah functional, Euler’s elastica, level set methods, and nonlinear diffusion PDEs—and describe how each has been coupled with deep architectures. We then organise the literature by tasks, covering image denoising, segmentation, reconstruction, super-resolution, inpainting, and deblurring. For each task, we survey major learning paradigms and representative deep learning integration frameworks—centred on variational loss formulation, algorithm unrolling, PDE-driven architectures, plug-and-play priors, deep image prior, and score-based generative modelling—highlighting the synergy between mathematically rigorous variational principles and empirical model performance. We further establish theoretical links: residual networks to ordinary differential equations, and score-based diffusion models to stochastic PDEs. Standard benchmarks and metrics are summarised, followed by open challenges such as cross-domain generalisation, robustness, interpretability, hallucination control, and data-efficient learning. Finally, conclusions are drawn based on the available evidence, and the maturity of each research direction is explicitly discussed. This review differs from existing surveys in three respects. It classifies methods by the locus at which the variational and the learned component are coupled—the training objective, the network architecture, or the energy minimised at inference—rather than by task or by assumption on the forward operator. It applies that single classification uniformly across six tasks, including segmentation and inpainting, which carry no explicit forward operator and therefore fall outside the scope of existing surveys of learned inverse problems. Moreover, it uses the resulting coverage map, audited entry by entry against a formal membership criterion, to distinguish combinations that have been realised from those that remain unexplored, those the taxonomy excludes, and those connected only loosely.

  • Open Access

    Optimized Small Object Detection Algorithm for UAV Perspective in Complex Traffic Scenarios

    by Shijun Liu, Honghao Zhu, Pengcheng Li

    Small object detection in UAV vision systems faces difficulties when the traffic environment is complex because the detection accuracy is limited and the feature representation is not sufficient. Balancing real-time performance with robustness can be hard. The paper proposes OSOD-UAV, an optimized small object detection algorithm for complex traffic scenarios. First, we introduce Spatial Adaptive Dynamic Convolution module into the backbone network. It can adjust the convolution weights. Second, the object distribution is discrete from UAV perspective and modeling long-range dependencies is difficult. The paper designs a Bi-Directional Axial Attention module to handle this. This module decouples the spatial dimensions. It enhances global contextual information interaction and can improve the model's perception ability in occluded and dense scenes. Finally, the paper uses WIoU (Wise-IoU) loss function instead of the traditional CIoU. A dynamic focusing mechanism is used to optimize the bounding box regression process and the detection accuracy is expected to improve. The experiments show that the paper's model achieves 32.8% in mAP@0.5. It is worth noting that the accuracy and robustness of small object detection in complex UAV scenes can be enhanced. Also, it provides a feasible solution for UAV visual detection in complex traffic flow environments.

  • Open Access

    Character-Level Adversarial Samples Generation Method based on Dual-Layer Text Watermark

    by Houyue Wu, Xianwei Li, Shunxiang Zhang, Honghao Zhu

    Textual adversarial samples are widely used to assess the robustness and security of language models. Most existing methods generate these samples by substitution or deletion. However, such approaches are often easy to detect and show limited attack effectiveness. To address these issues, we propose a character-level adversarial attack framework based on dual-layer watermarking. The method embeds explicit and implicit watermarks in the original text to interfere with downstream models. First, we apply character-level gradient optimization to perturb explicit statistical features. Then, we use reinforcement learning to fine-tune semantic and encoding characteristics, guided by feedback from the target model. This approach ensures that the generated samples significantly reduce the performance of target models while preserving semantic meaning. Experiments on four benchmark datasets show that our method outperforms state-of-the-art baselines. It reduces average word distance (WD) by 12.5%, improves BERTScore by 0.9%, lowers the attack success rate (TMASR) by 11.0%, and increases the concealment rate (Hide) by 5.6%. Our code is publicly available at: https://github.com/why3664/adv-text-watermark.


  • Top Cited Papers

    A Transactional-Behavior-Based Hierarchical Gated Network for Credit Card Fraud Detection

    by Yu Xie, MengChu Zhou, Guanjun Liu, Lifei Wei, Honghao Zhu, Pasquale De Meo

    The task of detecting fraud in credit card transactions is crucial to ensure the security and stability of a financial system, as well as to enforce customer confidence in digital payment systems. Historically, credit card companies have used rule-based approaches to detect fraudulent transactions, but these have proven inadequate due to the complexity of fraud strategies and have been replaced by much more powerful solutions based on machine learning or deep learning algorithms. Despite significant progress, the current approaches to fraud detection suffer from a number of limitations: for example, it is unclear whether some transaction features are more effective than others in discriminating fraudulent transactions, and they often neglect possible correlations among transactions, even though they could reveal illicit behaviour. In this paper, we propose a novel credit card fraud detection (CCFD) method based on a transaction behaviour-based hierarchical gated network. First, we introduce a feature-oriented extraction module capable of identifying key features from original transactions, and such analysis is effective in revealing the behavioural characteristics of fraudsters. Second, we design a transaction-oriented extraction module capable of capturing the correlation between users' historical and current transactional behaviour. Such information is crucial for revealing users' sequential behaviour patterns. Our approach, called transactional-behaviour-based hierarchical gated network model (TbHGN), extracts two types of new transactional features, which are then combined in a feature interaction module to learn the final transactional representations used for CCFD. We have conducted extensive experiments on a real-world credit card transaction dataset with an increase in average F1 between 1.42% and 6.53% and an improvement in average AUC between 0.63% and 2.78% over the state of the art.


  • Hybrid Quantum Classical Optimization for Low-Carbon Sustainable Edge Architecture in RIS-Assisted AIoT Healthcare Systems

    by Keping Yu, Chinmay Chakraborty, Dongyang Xu, Tiantian Zhang, Honghao Zhu, Osama Alfarraj, Amr Tolba

    Healthcare systems, empowered by the integration of Artificial Intelligence (AI) and Internet of Things networks, are undergoing significant advancements, ushering in a new era of enhanced treatment experiences and improved quality of life. Edge computing plays a pivotal role as an architectural enabler; however, it also presents numerous energy-related challenges spanning sensors, communication, and edge devices. One of the most formidable challenges is the proliferation of complex communication protocols across various devices, including sensors, reconfigurable intelligent surfaces, smart devices, and edge servers, leading to substantial carbon emissions and energy consumption. To address this challenge, this paper introduces a low-carbon, sustainable edge architecture leveraging AI techniques. Specifically, we develop a deep learning-based radio frequency fingerprint access protocol to facilitate real-time and energy-efficient device access between smart devices and edge gateways. Building upon this foundation, we propose a hybrid quantum-classical optimization algorithm to achieve green data transmission at lower layers for artificial intelligence of things healthcare systems. Simulation results demonstrate that our optimized architecture achieves over 99% identification accuracy using a signal dataset of 50GB obtained from real-world smart devices and practical gateways in a real-world environment, all while maintaining energy-efficient data delivery.


  • A Self-Adapting and Efficient Dandelion Algorithm and Its Application to Feature Selection for Credit Card Fraud Detection

    by Honghao Zhu, MengChu Zhou, Yu Xie, Aiiad Albeshri

    A dandelion algorithm (DA) is a recently developed intelligent optimization algorithm for function optimization problems. Many of its parameters need to be set by experience in DA, which might not be appropriate for all optimization problems. A self-adapting and efficient dandelion algorithm is proposed in this work to lower the number of DA's parameters and simplify DA's structure. Only the normal sowing operator is retained; while the other operators are discarded. An adaptive seeding radius strategy is designed for the core dandelion. The results show that the proposed algorithm achieves better performance on the standard test functions with less time consumption than its competitive peers. In addition, the proposed algorithm is applied to feature selection for credit card fraud detection (CCFD), and the results indicate that it can obtain higher classification and detection performance than the-state-of-the-art methods.


  • Open Access

    A Finger Vein Recognition Algorithm Based on the Histogram of Variable Curvature Directional Binary Statistics

    by Min Li, Xue Jiang, Honghao Zhu, Fei Liu, Huabin Wang, Liang Tao, Shijun Liu

    Structural features are capable of effectively capturing the overall texture variations in images. However, in locally prominent areas with visible veins, other characteristics such as directionality, convexity–concavity, and curvature also play a crucial role in recognition, and their impact cannot be overlooked. This paper introduces a novel approach, the histogram of variable curvature directional binary statistical (HVCDBS), which combines the structural and directional features of images. The proposed method is designed for extracting discriminative multifeature information in vein recognition. First, a multidirection and multicurvature Gabor filter is introduced for convolution with vein images, yielding directional and convexity–concavity information at each pixel, along with curvature information for the corresponding curve. Simultaneously incorporating the original image feature information, these four aspects of information are fused and encoded to construct a variable curvature binary pattern (VCBP) with multifeatures. Second, the feature map containing multifeature information is blockwise processed to build variable curvature binary statistical features. Finally, competitive Gabor directional binary statistical features are combined, and a matching score‐level fusion scheme is employed based on maximizing the interclass distance and minimizing the intraclass distance to determine the optimal weights. This process fuses the two feature maps into a one‐dimensional feature vector, achieving an effective representation of vein images. Extensive experiments were conducted on four widely utilized vein databases, and the results indicate that the proposed algorithm, compared with solely extraction of structural features, achieved higher recognition rates and lower equal error rates.


  • NUS: Noisy-Sample-Removed Undersampling Scheme for Imbalanced Classification and Application to Credit Card Fraud Detection

    by Honghao Zhu, MengChu Zhou, Guanjun Liu, Yu Xie, Shijun Liu, Cheng Guo

    Since minority samples are substantially less common than majority samples, many industrial applications, such as credit card fraud detection (CCFD) and defective part identification, call for imbalanced classification. The performance of a classifier tends to suffer from the noisy samples in majority or minority classes. This work proposes a new undersampling scheme, called a clustering-based noisy-sample-removed undersampling scheme (NUS) for imbalanced classification. The majority class samples are first clustered. The distance of the majority class sample from the cluster center that is furthest away is used as the radius to build a hypersphere, with each cluster’s center assumed to be a spherical center. We determine the Euclidean distance between the center of a cluster and each minority sample to find whether they are in the hypersphere or not. Afterward, we exclude noisy samples from the minority class. The noisy samples of majority classes are removed by using the same procedure. Second, we propose an NUS, which combines noisy sample removal with undersampling techniques. Finally, to prove the effectiveness of NUS, we integrate NUS with the basic classifiers random forest (RF), decision tree (DT), and logistics regression (LR). We conduct their comparison with seven undersampling, oversampling, and noisy-sample-removed methods. This work performs experiments on 13 public and three real transaction datasets related to e-commerce. The results show that NUS plays a positive role in promoting existing classifiers’ performance.


  • Open Access

    Dynamic elite strategy mayfly algorithm

    by Qianhang Du, Honghao Zhu

    The mayfly algorithm (MA), as a newly proposed intelligent optimization algorithm, is found that easy to fall into the local optimum and slow convergence speed. To address this, an improved mayfly algorithm based on dynamic elite strategy (DESMA) is proposed in this paper. Specifically, it first determines the specific space near the best mayfly in the current population, and dynamically sets the search radius. Then generating a certain number of elite mayflies within this range. Finally, the best one among the newly generated elite mayflies is selected to replace the best mayfly in the current population when the fitness value of elite mayfly is better than that of the best mayfly. Experimental results on 28 standard benchmark test functions from CEC2013 show that our proposed algorithm outperforms its peers in terms of accuracy speed and stability.