Browsing by Author "Li, Jun"

Now showing 1 - 12 of 12
  • Thumbnail Image
    ItemOpen Access
    A data-driven approach for automatic aircraft engine borescope inspection defect detection using computer vision and deep learning
    (MDPI, 2025-03-01) Schaller, Thibaud; Li, Jun; Jenkins, Karl W.
    Regular aircraft engine inspections play a crucial role in aviation safety. However, traditional inspections are often performed manually, relying heavily on the judgment and experience of operators. This paper presents a data-driven deep learning framework capable of automatically detecting defects on reactor blades. Specifically, this study develops Deep Neural Network models to detect defects in borescope images using various datasets, based on Computer Vision and YOLOv8n object detection techniques. Firstly, reactor blade images are collected from public resources and then annotated and preprocessed into different groups based on Computer Vision techniques. In addition, synthetic images are generated using Deep Convolutional Generative Adversarial Networks and a manual data augmentation approach by randomly pasting defects onto reactor blade images. YOLOv8n-based deep learning models are subsequently fine-tuned and trained on these dataset groups. The results indicate that the model trained on wide-shot blade images performs better overall at detecting defects on blades compared to the model trained on zoomed-in images. The comparison of multiple models’ results reveals inherent uncertainties in model performance that while some models trained on data enhanced by Computer Vision techniques may appear more reliable in some types of defect detection, the relationship between these techniques and subsequent results cannot be generalized. The impact of epochs and optimizers on the model’s performance indicates that incorporating rotated images and selecting an appropriate optimizer are key factors for effective model training. Furthermore, models trained solely on artificially generated images from collages perform poorly at detecting defects in real images. A potential solution is to train the model on both synthetic and real images. Future work will focus on improving the framework’s performance and conducting a more comprehensive uncertainty analysis by utilizing larger and more diverse datasets, supported by enhanced computational power.
  • Thumbnail Image
    ItemOpen Access
    Analysis of China’s high-speed railway network using complex network theory and graph convolutional networks
    (MDPI, 2025-04-16) Xu, Zhenguo; Li, Jun; Moulitsas, Irene; Niu, Fangqu
    This study investigated the characteristics and functionalities of China’s High-Speed Railway (HSR) network based on Complex Network Theory (CNT) and Graph Convolutional Networks (GCN). First, complex network analysis was applied to provide insights into the network’s fundamental characteristics, such as small-world properties, efficiency, and robustness. Then, this research developed three novel GCN models to identify key nodes, detect community structures, and predict new links. Findings from the complex network analysis revealed that China’s HSR network exhibits a typical small-world property, with a degree distribution that follows a log-normal pattern rather than a power law. The global efficiency indicator suggested that stations are typically connected through direct routes, while the local efficiency indicator showed that the network performs effectively within local areas. The robustness study indicated that the network can quickly lose connectivity if key nodes fail, though it showed an ability initially to self-regulate and has partially restored its structure after disruption. The GCN model for key node identification revealed that the key nodes in the network were predominantly located in economically significant and densely populated cities, positively contributing to the network’s overall efficiency and robustness. The community structures identified by the integrated GCN model highlight the economic and social connections between official urban clusters and the communities. Results from the link prediction model suggest the necessity of improving the long-distance connectivity across regions. Future work will explore the network’s socio-economic dynamics and refine and generalise the GCN models.
  • Thumbnail Image
    ItemOpen Access
    Decision tool of medical endoscope maintenance service in Chinese hospitals: a conjoint analysis
    (BioMed Central, 2023-12-15) Zheng, Jun; Wei, Jingming; Xie, Ying; Chen, Siyao; Li, Jun; Lou, Ligang; Sun, Jing; Feng, Jingyi
    Medical devices are instruments, apparatus, appliances, software, implants, reagents, materials or other articles that are intended for use in the treatment or diagnosis of disease or injury in humans. Concerning medical endoscope devices, which enable doctors to observe and manipulate the area under examination through a puncture hole in the body cavity or organ, hospitals predominantly consider the quality and cost of maintenance services when making their selection. The effective and efficient provision of maintenance services plays a crucial role in ensuring cost-effective and high-quality management of medical devices. In this study, we have developed an innovative decision tool that analyzed key factors impacting the choice of medical devices’ maintenance service. This tool assists hospitals in evaluating and selecting appropriate maintenance services for medical device, specifically endoscopy devices. Moreover, it also serves as a valuable resource for manufacturers and suppliers to enhance their after-sales service offerings.
  • Thumbnail Image
    ItemOpen Access
    A deep-learning-based approach for aircraft engine defect detection
    (MDPI, 2023-02-01) Upadhyay, Anurag; Li, Jun; King, Steve; Addepalli, Sri
    Borescope inspection is a labour-intensive process used to find defects in aircraft engines that contain areas not visible during a general visual inspection. The outcome of the process largely depends on the judgment of the maintenance professionals who perform it. This research develops a novel deep learning framework for automated borescope inspection. In the framework, a customised U-Net architecture is developed to detect the defects on high-pressure compressor blades. Since motion blur is introduced in some images while the blades are rotated during the inspection, a hybrid motion deblurring method for image sharpening and denoising is applied to remove the effect based on classic computer vision techniques in combination with a customised GAN model. The framework also addresses the data imbalance, small size of the defects and data availability issues in part by testing different loss functions and generating synthetic images using a customised generative adversarial net (GAN) model, respectively. The results obtained from the implementation of the deep learning framework achieve precisions and recalls of over 90%. The hybrid model for motion deblurring results in a 10× improvement in image quality. However, the framework only achieves modest success with particular loss functions for very small sizes of defects. The future study will focus on very small defects detection and extend the deep learning framework to general borescope inspection.
  • Thumbnail Image
    ItemOpen Access
    An empirical evaluation of generative adversarial nets in synthesizing X-ray chest images
    (IEEE, 2022-12-12) Belmekki, Zakariae; Li, Jun; Jenkins, Karl W.; Reuter, Patrick; Gómez Jáuregui, David Antonio
    In the last decade, Generative Adversarial Nets (GAN) have become a subject of growing interest in multiple research fields. In this paper, we focus on applications in the medical field by attempting to generate realistic X-ray chest images. A heuristic approach is adopted to perform an extensive evaluation of different architecture configurations and optimization algorithms and we propose an optimal configuration of the baseline Deep Convolutional GAN (DCGAN) based on empirical findings. Additionally, we highlight the technical limitations of GAN and provide an analysis of the high memory requirements, which we reduce by a range of 1.2-7 percent by removing unnecessary layers. Finally, we verify that the loss of the discriminator can be used as an assessment metric.
  • Thumbnail Image
    ItemOpen Access
    Federated reinforcement learning enhanced human-robotic systems: a comprehensive review
    (IEEE, 2024-10-11) Upadhyay, Anurag; Abafat, Soheil; Baradaranshokouhi, Yashar; Lu, Xin; Jing, Yanguo; Li, Jun
    Federated Reinforcement learning (FRL) presents a transformative approach for leveraging Human-robot collaboration (HRC) systems by addressing critical challenges in traditional learning paradigms. This paper provides a comprehensive review of the current state of FRL technology and its potential applications within HRC systems. The adaptation of FRL in HRC system is still in its infancy. This review systematically analyses the development trends, current challenges, and future prospects of various learning approaches within HRC systems. The paper highlights the critical factors of developing a conceptual frame-work for FRL within HRC systems to fully realise the potential of FRL. This paper aims to provide valuable insights and guidance for future research efforts focused on advancing FRL technology for human-robotic collaboration.
  • Thumbnail Image
    ItemOpen Access
    Generalized quadrature spatial modulation and its application to vehicular networks with NOMA
    (IEEE, 2020-07-16) Li, Jun; Dang, Shuping; Yan, Yier; Peng, Yuyang; Al-Rubaye, Saba; Tsourdos, Antonios
    Quadrature spatial modulation (QSM) is recently proposed to increase the spectral efficiency (SE) of SM, which extends the transmitted symbols into in-phase and quadrature domains. In this paper, we propose a generalized QSM (GQSM) scheme to further increase the SE of QSM by activating more than one transmit antenna in in-phase or quadrature domain. A low-complexity detection for GQSM is provided to mitigate the detection burden of the optimal maximum-likelihood (ML) detection method. An upper bounded bit error rate is analyzed to discover the system performance of GQSM. Moreover, by collaborating with the non-orthogonal multiple access (NOMA) technique, we investigate the practical application of GQSM to cooperative vehicular networks and propose the cooperative GQSM with OMA (C-OMA-GQSM) and cooperative GQSM with NOMA (C-NOMA-GQSM) schemes. Computer simulation results verify the reliability of the proposed low-complexity detection as well as the theoretical analysis, and show that GQSM outperforms QSM in the entire SNR region. The superior BER performance of the proposed C-NOMA-GQSM scheme make it a promising modulation candidate for next generation vehicular networks.
  • Thumbnail Image
    ItemOpen Access
    Joint optimization of depth and ego-motion for intelligent autonomous vehicles
    (IEEE, 2022-03-24) Gao, Yongbin; Tian, Fangzheng; Li, Jun; Fang, Zhijun; Al-Rubaye, Saba; Song, Wei; Yan, Yier
    The three-dimensional (3D) perception of autonomous vehicles is crucial for localization and analysis of the driving environment, while it involves massive computing resources for deep learning, which can't be provided by vehicle-mounted devices. This requires the use of seamless, reliable, and efficient massive connections provided by the 6G network for computing in the cloud. In this paper, we propose a novel deep learning framework with 6G enabled transport system for joint optimization of depth and ego-motion estimation, which is an important task in 3D perception for autonomous driving. A novel loss based on feature map and quadtree is proposed, which uses feature value loss with quadtree coding instead of photometric loss to merge the feature information at the texture-less region. Besides, we also propose a novel multi-level V-shaped residual network to estimate the depths of the image, which combines the advantages of V-shaped network and residual network, and solves the problem of poor feature extraction results that may be caused by the simple fusion of low-level and high-level features. Lastly, to alleviate the influence of image noise on pose estimation, we propose a number of parallel sub-networks that use RGB image and its feature map as the input of the network. Experimental results show that our method significantly improves the quality of the depth map and the localization accuracy and achieves the state-of-the-art performance.
  • Thumbnail Image
    ItemOpen Access
    Multiclass sentiment prediction of airport service online reviews using aspect-based sentimental analysis and machine learning
    (MDPI, 2024-03-06) Alanazi, Mohammed Saad M.; Li, Jun; Jenkins, Karl W.
    Airport service quality ratings found on social media such as Airline Quality and Google Maps offer invaluable insights for airport management to improve their quality of services. However, there is currently a lack of research analysing these reviews by airport services using sentimental analysis approaches. This research applies multiclass models based on Aspect-Based Sentimental Analysis to conduct a comprehensive analysis of travellers’ reviews, in which the major airport services are tagged by positive, negative, and non-existent sentiments. Seven airport services commonly utilised in previous studies are also introduced. Subsequently, various Deep Learning architectures and Machine Learning classification algorithms are developed, tested, and compared using data collected from Twitter, Google Maps, and Airline Quality, encompassing travellers’ feedback on airport service quality. The results show that the traditional Machine Learning algorithms such as the Random Forest algorithm outperform Deep Learning models in the multiclass prediction of airport service quality using travellers’ feedback. The findings of this study offer concrete justifications for utilising multiclass Machine Learning models to understand the travellers’ sentiments and therefore identify airport services required for improvement.
  • Thumbnail Image
    ItemOpen Access
    Physical-layer counterattack strategies for the internet of bio-nano things with molecular communication
    (IEEE, 2023-06-06) Huang, Yu; Wen, Miaowen; Lin, Lin; Li, Bin; Wei, Zhuangkun; Tang, Dong; Li, Jun; Duan, Wei; Guo, Weisi
    Molecular communication (MC) is an emerging new communication paradigm where information is conveyed by chemical signals. It has been recognized as one of the most promising physical layer techniques for the future Internet of Bio-Nano Things (IoBNT), which enables revolutionary applications beyond our imagination. Compared with conventional communication systems, MC typically demands a higher security level as the IoBNT is deeply associated with the biochemical process. Against this background, this article first discusses the security and privacy issues of IoBNT with MC. Then, the physical-layer countermeasures against the threat are presented from an interdisciplinary perspective concerning data science, signal processing techniques, and the biochemical properties of MC. Correspondingly, both the keyless and key-based schemes are conceived and revisited. Finally, some open research issues and future research directions for secrecy enhancement in IoBNT with MC are put forward.
  • Thumbnail Image
    ItemOpen Access
    Predicting passengers’ feedback rate for airport service quality
    (Elsevier, 2024-02-24) Alanazi, Mohammed Saad M.; Jenkins, Karl W.; Li, Jun
    Airport service quality evaluation is commonly found on social media sites, including Google Maps. The reviews by users of Google Maps are longer in terms of the number of words than those found on Twitter. They also include a rating, whereas those on Twitter need to be labelled. However, they are less well known than those on Twitter amongst researchers who focus on sentimental analysis. This study attempts to fill the gap in the current literature and develops architecture that is based on Long-Short Term Memory Neural Networks and Convolution Neural Networks. The combined model developed receives meta-data, such as the number of words in the review and the number of likes the review receives in addition to the key review words. The two models, the first of which predicts polarity and the second reviews ratings, were tested under several variations of parameters and showed consistency in results. The dataset was collected from Google Maps and focused on two crowded airports in the Arabic Peninsula (Doha and Dubai). They were found to be unbalanced, with positive reviews being more abundant than negative reviews.
  • Thumbnail Image
    ItemOpen Access
    Vortex and core detection using computer vision and machine learning methods
    (River Publishers, 2023-12-30) Xu, Zhenguo; Maria, Ayush; Chelli, Kahina; De Premare, Thibaut Dumouchel; Bilbao, Xabadin; Petit, Christopher; Zoumboulis-Airey, Robert; Moulitsas, Irene; Teschner, Tom-Robin; Asif, Seemal; Li, Jun
    The identification of vortices and cores is crucial for understanding airflow motion in aerodynamics. Currently, numerous methods in Computer Vision and Machine Learning exist for detecting vortices and cores. This research develops a comprehensive framework by combining classic Computer Vision and state-of-the-art Machine Learning techniques for vortex and core detection. It enhances a CNN-based method using Computer Vision algorithms for Feature Engineering and then adopts an Ensemble Learning approach for vortex core classification, through which false positives, false negatives, and computational costs are reduced. Specifically, four features, i.e., Contour Area, Aspect Ratio, Area Difference, and Moment Centre, are employed to identify vortex regions using YOLOv5s, followed by a hard voting classifier based on Random Forest, Adaptive Boosting, and Xtreme Gradient Boosting algorithms for vortex core detection. This novel approach differs from traditional Computer Vision approaches using mathematical variables and image features such as HAAR and SIFT for vortex core detection. The findings show that vortices are detected with a high degree of statistical confidence by a fine-tuned YOLOv5s model, and the integrated technique produces an accuracy score of 97.56% in detecting vortex cores conducted on a total of 133 images generated from a rotor blade NACA0012 simulation. Future work will focus on framework generalisation with a larger and more diverse dataset and intelligent threshold development for more efficient vortex and core detection.