Browsing by Author "Zhao, Yifan"
Now showing 1 - 20 of 154
Results Per Page
Sort Options
Item Open Access 2-D generating function of the zernike polynomials and their application for image classification(IEEE, 2019-12-19) Honarvar Shakibaei Asli, Barmak; Flusser, Jan; Zhao, YifanThis work proposes a new approach to find the generating function (GF) of the Zernike polynomials in two dimensional form. Combining the methods of GFs and discrete-time systems, we can develop two dimensional digital systems for systematic generation of entire orders of Zernike polynomials. We establish two different formulas for the GF of the radial Zernike polynomials based on both the degree and the azimuthal order of the radial polynomials. In this paper, we use four terms recurrence relation instead of the ordinary three terms recursion to calculate the radial Zernike polynomials and their GFs using unilateral 2D Z-transform. A spatio-temporal implementation scheme is developed for generation of the radial Zernike polynomials. Since Zernike moments (ZMs) are invariant with respect to rotation, translation and scaling, the experimental schemes show the image classification applications by using the proposed algorithm.Item Open Access An adaptive pig face recognition approach using convolutional neural networks(Elsevier, 2020-04-16) Marsot, Mathieu; Mei, Jiangqiang; Shan, Xiaocai; Ye, Liyong; Feng, Peng; Yan, Xuejun; Li, Chenfan; Zhao, YifanThe evolution of agriculture towards intensive farming leads to an increasing demand for animal identification associated with high traceability, driven by the need for quality control and welfare management in agricultural animals. Automatic identification of individual animals is an important step to achieve individualised care in terms of disease detection and control, and improvement of the food quality. For example, as feeding patterns can differ amongst pigs in the same pen, even in homogenous groups, automatic registration shows the most potential when applied to an individual pig. In the EU for instance, this capability is required for certification purposes. Although the RFID technology has been gradually developed and widely applied for this task, chip implanting might still be time-consuming and costly for current practical applications. In this paper, a novel framework composed of computer vision algorithms, machine learning and deep learning techniques is proposed to offer a relatively low-cost and scalable solution of pig recognition. Firstly, pig faces and eyes are detected automatically by two Haar feature-based cascade classifiers and one shallow convolutional neural network to extra high-quality images. Secondly, face recognition is performed by employing a deep convolutional neural network. Additionally, class activation maps generated by grad-CAM and saliency maps are utilised to visually understand how the discriminating parameters have been learned by the neural network. By applying the proposed approach on 10 randomly selected pigs filmed in farm condition, the proposed method demonstrates the superior performance against the state-of-art method with an accuracy of 83% over 320 testing images. The outcome of this study will facilitate the real-application of AI-based animal identification in swine production.Item Open Access Advanced sensing and control for connected and automated vehicles(MDPI, 2022-02-16) Huang, Chao; Du, Haiping; Zhao, Wanzhong; Zhao, Yifan; Yan, Fuwu; Lv, ChenIn recent years, connected and automated vehicles (CAV) have been a transformative technology that is expected to reduce emissions and change and improve the safety and efficiency of the mobilities [...]Item Open Access Advanced uncertainty quantification with dynamic prediction techniques under limited data for industrial maintenance applications.(Cranfield University, 2021-07) Grenyer, Alex; Erkoyuncu, John Ahmet; Zhao, YifanEngineering systems are expected to function effectively whilst maintaining reliability in service. These systems consist of various equipment units, many of which are maintained on a corrective or time-based basis. Challenges to plan maintenance accounting for turnaround times, equipment availability and resulting costs manifest varying degrees of uncertainty stemming from multiple quantitative and qualitative (compound) sources throughout the in-service life. Under or over-estimating this uncertainty can lead to increased failure rates or, more often, unnecessary maintenance being carried out. As well as the quality availability of data, uncertainty is driven by the influence of expert experience or assumptions and environmental operating conditions. Accommodating for uncertainty requires the determination of key contributors, their influence on interconnected units and how this might change over time. This research aims to develop a modelling approach to quantify, aggregate and forecast uncertainty given by a combination of historic equipment data and heuristic estimates for in-service engineering systems. Research gaps and challenges are identified through a systematic literature review and supported by a series of surveys and interviews with industrial practitioners. These are addressed by the development of two frameworks: (1) quantify and aggregate compound uncertainty, and (2) predict uncertainty under limited data. The two frameworks are brought together to produce the Multistep Compound Dynamic Uncertainty Quantification (MCDUQ) app, developed in MATLAB. Results demonstrate effective measurement of compound uncertainties and their impact on system reliability, along with robust predictions under limited data with an immersive visualisation of dynamic uncertainty. The embedded frameworks are each validated through implementation in two case studies. The app is verified with industrial experts through a series of interviews and virtual demonstrations.Item Open Access An AI-powered navigation framework to achieve an automated acquisition of cardiac ultrasound images(Springer Nature, 2023-09-11) Soemantoro, Raska; Kardos, Attila; Tang, Gilbert; Zhao, YifanEchocardiography is an effective tool for diagnosing cardiovascular disease. However, numerous challenges affect its accessibility, including skill requirements, workforce shortage, and sonographer strain. We introduce a navigation framework for the automated acquisition of echocardiography images, consisting of 3 modules: perception, intelligence, and control. The perception module contains an ultrasound probe, a probe actuator, and a locator camera. Information from this module is sent to the intelligence module, which grades the quality of an ultrasound image for different echocardiography views. The window search algorithm in the control module governs the decision-making process in probe movement, finding the best location based on known probe traversal positions and image quality. We conducted a series of simulations using the HeartWorks simulator to assess the proposed framework. This study achieved an accuracy of 99% for the image quality model, 96% for the probe locator model, and 99% for the view classification model, trained on an 80/20 training and testing split. We found that the best search area corresponds with general guidelines: at the anatomical left of the sternum between the 2nd and 5th intercostal space. Additionally, the likelihood of successful acquisition is also driven by how long it stores past coordinates and how much it corrects itself. Results suggest that achieving an automated echocardiography system is feasible using the proposed framework. The long-term vision is of a widely accessible and accurate heart imaging capability within hospitals and community-based settings that enables timely diagnosis of early-stage heart disease.Item Open Access Analysis of autopilot disengagements occurring during autonomous vehicle testing(IEEE, 2017-12-20) Lv, Chen; Cao, Dongpu; Zhao, Yifan; Auger, Daniel J.; Sullman, Mark; Wang, Huaji; Millen Dutka, Laura; Skrypchuk, Lee; Mouzakitis, AlexandrosIn present-day highly-automated vehicles, there are occasions when the driving system disengages and the human driver is required to take-over. This is of great importance to a vehicle U+02BC s safety and ride comfort. In the U.S state of California, the Autonomous Vehicle Testing Regulations require every manufacturer testing autonomous vehicles on public roads to submit an annual report summarizing the disengagements of the technology experienced during testing. On 1 January 2016, seven manufacturers submitted their first disengagement reports: Bosch, Delphi, Google, Nissan, Mercedes-Benz, Volkswagen, and Tesla Motors. This work analyses the data from these disengagement reports with the aim of gaining abetter understanding of the situations in which a driver is required to takeover, as this is potentially useful in improving the Society of Automotive Engineers U+0028 SAE U+0029 Level 2 and Level 3 automation technologies. Disengagement events from testing are classified into different groups based on attributes and the causes of disengagement are investigated and compared in detail. The mechanisms and time taken for take-over transition occurred in disengagements are studied. Finally, recommendations for OEMs, manufacturers, and government organizations are also discussed.Item Open Access Angle-closure assessment in anterior segment OCT images via deep learning(Elsevier, 2021-01-07) Hao, Huaying; Zhao, Yitian; Yan, Qifeng; Higashita, Risa; Zhang, Jiong; Zhao, Yifan; Xu, Yanwu; Li, Fei; Zhang, Xiulan; Liu, JiangPrecise characterization and analysis of anterior chamber angle (ACA) are of great importance in facilitating clinical examination and diagnosis of angle-closure disease. Currently, the gold standard for diagnostic angle assessment is observation of ACA by gonioscopy. However, gonioscopy requires direct contact between the gonioscope and patients’ eye, which is uncomfortable for patients and may deform the ACA, leading to false results. To this end, in this paper, we explore a potential way for grading ACAs into open-, appositional- and synechial angles by Anterior Segment Optical Coherence Tomography (AS-OCT), rather than the conventional gonioscopic examination. The proposed classification schema can be beneficial to clinicians who seek to better understand the progression of the spectrum of angle-closure disease types, so as to further assist the assessment and required treatment at different stages of angle-closure disease. To be more specific, we first use an image alignment method to generate sequences of AS-OCT images. The ACA region is then localized automatically by segmenting an important biomarker - the iris - as this is a primary structural cue in identifying angle-closure disease. Finally, the AS-OCT images acquired in both dark and bright illumination conditions are fed into our Multi-Sequence Deep Network (MSDN) architecture, in which a convolutional neural network (CNN) module is applied to extract feature representations, and a novel ConvLSTM-TC module is employed to study the spatial state of these representations. In addition, a novel time-weighted cross-entropy loss (TC) is proposed to optimize the output of the ConvLSTM, and the extracted features are further aggregated for the purposes of classification. The proposed method is evaluated across 66 eyes, which include 1584 AS-OCT sequences, and a total of 16,896 images. The experimental results show that the proposed method outperforms existing state-of-the-art methods in applicability, effectiveness, and accuracy.Item Open Access Attention mechanism enhanced spatiotemporal-based deep learning approach for classifying barely visible impact damages in CFRP materials(Elsevier, 2024-03-14) Deng, Kailun; Liu, Haochen; Cao, Jun; Yang, Lichao; Du, Weixiang; Xu, Yigeng; Zhao, Yifan; This work was partially supported by the Royal Academy of Engineering Industrial Fellowship [#grant IF2223B-110], and partially supported by the Science and Technology Department of Gansu Province Science and Technology Project Funding, 22YF7GA072.Most existing machine learning approaches for analysing thermograms mainly focus on either thermal images or pixel-wise temporal profiles of specimens. To fully leverage useful information in thermograms, this article presents a novel spatiotemporal-based deep learning model incorporating an attention mechanism. Using captured thermal image sequences, the model aims to better characterise barely visible impact damages (BVID) in composite materials caused by different impact energy levels. This model establishes the relationship between patterns of BVID in thermography and their corresponding impact energy levels by learning from spatial and temporal information simultaneously. Validation of the model using 100 composite specimens subjected to five different low-velocity impact forces demonstrates its superior performance with a classification accuracy of over 95%. The proposed approach can contribute to Structural Health Monitoring (SHM) community by enabling cause analysis of impact incidents based on predicting the potential impact energy levels. This enables more targeted predictive maintenance, which is especially significant in the aviation industry, where any impact incidents can have catastrophic consequences.Item Open Access Automated tortuosity analysis of nerve fibers in corneal confocal microscopy(IEEE, 2020-02-17) Zhao, Yitian; Zhang, Jiong; Pereira, Ella; Zheng, Yalin; Su, Pan; Xie, Jianyang; Zhao, Yifan; Shi, Yonggang; Qi, Hong; Liu, Jiang; Liu, YonghuaiPrecise characterization and analysis of corneal nerve fiber tortuosity are of great importance in facilitating examination and diagnosis of many eye-related diseases. In this paper we propose a fully automated method for image-level tortuosity estimation, comprising image enhancement, exponential curvature estimation, and tortuosity level classification. The image enhancement component is based on an extended Retinex model, which not only corrects imbalanced illumination and improves image contrast in an image, but also models noise explicitly to aid removal of imaging noise. Afterwards, we take advantage of exponential curvature estimation in the 3D space of positions and orientations to directly measure curvature based on the enhanced images, rather than relying on the explicit segmentation and skeletonization steps in a conventional pipeline usually with accumulated pre-processing errors. The proposed method has been applied over two corneal nerve microscopy datasets for the estimation of a tortuosity level for each image. The experimental results show that it performs better than several selected state-of-the-art methods. Furthermore, we have performed manual gradings at tortuosity level of four hundred and three corneal nerve microscopic images, and this dataset has been released for public access to facilitate other researchers in the community in carrying out further research on the same and related topics.Item Open Access Automatic 2-D/3-D vessel enhancement in multiple modality images using a weighted symmetry filter(IEEE, 2017-09-26) Zhao, Yitian; Zhao, Yitian; Zheng, Yalin; Liu, Yonghuai; Zhao, Yifan; Luo, Lingling; Yang, Siyuan; Na, Tong; Wang, Yongtian; Liu, JiangAutomated detection of vascular structures is of great importance in understanding the mechanism, diagnosis and treatment of many vascular pathologies. However, automatic vascular detection continues to be an open issue because of difficulties posed by multiple factors such as poor contrast, inhomogeneous backgrounds, anatomical variations, and the presence of noise during image acquisition. In this paper, we propose a novel 2D/3D symmetry filter to tackle these challenging issues for enhancing vessels from different imaging modalities. The proposed filter not only considers local phase features by using a quadrature filter to distinguish between lines and edges, but also uses the weighted geometric mean of the blurred and shifted responses of the quadrature filter, which allows more tolerance of vessels with irregular appearance. As a result, this filter shows a strong response to the vascular features under typical imaging conditions. Results based on 8 publicly available datasets (six 2D datasets, one 3D dataset and one 3D synthetic dataset) demonstrate its superior performance to other state-ofthe- art methods.Item Open Access Automatic reconstruction of irregular shape defects in pulsed thermography using deep learning neural network(Springer, 2022-07-25) Liu, Haochen; Li, Wenhan; Yang, Lichao; Deng, Kailun; Zhao, YifanQuantitative defect and damage reconstruction play a critical role in industrial quality management. Accurate defect characterisation in Infrared Thermography (IRT), as one of the widely used Non-Destructive Testing (NDT) techniques, always demands adequate pre-knowledge which poses a challenge to automatic decision-making in maintenance. This paper presents an automatic and accurate defect profile reconstruction method, taking advantage of deep learning Neural Networks (NN). Initially, a fast Finite Element Modelling (FEM) simulation of IRT is introduced for defective specimen simulation. Mask Region-based Convolution NN (Mask-RCNN) is proposed to detect and segment the defect using a single thermal frame. A dataset with a single-type-shape defect is tested to validate the feasibility. Then, a dataset with three mixed shapes of defect is inspected to evaluate the method’s capability on the defect profile reconstruction, where an accuracy over 90% on Intersection over Union (IoU) is achieved. The results are compared with several state-of-the-art of post-processing methods in IRT to demonstrate the superiority at detailed defect corners and edges. This research lays solid evidence that AI deep learning algorithms can be utilised to provide accurate defect profile reconstruction in thermography NDT, which will contribute to the research community in material degradation analysis and structural health monitoring.Item Open Access Brain functional and effective connectivity based on electroencephalography recordings: a review(Wiley, 2021-10-20) Cao, Jun; Zhao, Yifan; Shan, Xiaocai; Wei, Hua-Liang; Guo, Yuzhu; Chen, Liangyu; Erkoyuncu, John Ahmet; Sarrigiannis, Ptolemaios GeorgiosFunctional connectivity and effective connectivity of the human brain, representing statistical dependence and directed information flow between cortical regions, significantly contribute to the study of the intrinsic brain network and its functional mechanism. Many recent studies on electroencephalography (EEG) have been focusing on modeling and estimating brain connectivity due to increasing evidence that it can help better understand various brain neurological conditions. However, there is a lack of a comprehensive updated review on studies of EEG-based brain connectivity, particularly on visualization options and associated machine learning applications, aiming to translate those techniques into useful clinical tools. This article reviews EEG-based functional and effective connectivity studies undertaken over the last few years, in terms of estimation, visualization, and applications associated with machine learning classifiers. Methods are explored and discussed from various dimensions, such as either linear or nonlinear, parametric or nonparametric, time-based, and frequency-based or time-frequency-based. Then it is followed by a novel review of brain connectivity visualization methods, grouped by Heat Map, data statistics, and Head Map, aiming to explore the variation of connectivity across different brain regions. Finally, the current challenges of related research and a roadmap for future related research are presented.Item Open Access Characterisation of cognitive load using machine learning classifiers of electroencephalogram data(MDPI, 2023-10-17) Wang, Qi; Smythe, Daniel; Cao, Jun; Hu, Zhilin; Proctor, Karl J.; Owens, Andrew P.; Zhao, YifanA high cognitive load can overload a person, potentially resulting in catastrophic accidents. It is therefore important to ensure the level of cognitive load associated with safety-critical tasks (such as driving a vehicle) remains manageable for drivers, enabling them to respond appropriately to changes in the driving environment. Although electroencephalography (EEG) has attracted significant interest in cognitive load research, few studies have used EEG to investigate cognitive load in the context of driving. This paper presents a feasibility study on the simulation of various levels of cognitive load through designing and implementing four driving tasks. We employ machine learning-based classification techniques using EEG recordings to differentiate driving conditions. An EEG dataset containing these four driving tasks from a group of 20 participants was collected to investigate whether EEG can be used as an indicator of changes in cognitive load. The collected dataset was used to train four Deep Neural Networks and four Support Vector Machine classification models. The results showed that the best model achieved a classification accuracy of 90.37%, utilising statistical features from multiple frequency bands in 24 EEG channels. Furthermore, the Gamma and Beta bands achieved higher classification accuracy than the Alpha and Theta bands during the analysis. The outcomes of this study have the potential to enhance the Human–Machine Interface of vehicles, contributing to improved safety.Item Open Access Characterization of driver neuromuscular dynamics for human-automation collaboration design of automated vehicles(IEEE, 2018-03-05) Lv, Chen; Wang, Huaji; Cao, Dongpu; Zhao, Yifan; Auger, Daniel J.; Sullman, Mark; Matthias, Rebecca; Skrypchuk, Lee; Mouzakitis, AlexandrosIn order to design an advanced human-automation collaboration system for highly automated vehicles, research into the driver's neuromuscular dynamics is needed. In this paper a dynamic model of drivers' neuromuscular interaction with a steering wheel is firstly established. The transfer function and the natural frequency of the systems are analyzed. In order to identify the key parameters of the driver-steering-wheel interacting system and investigate the system properties under different situations, experiments with driver-in-the-loop are carried out. For each test subject, two steering tasks, namely the passive and active steering tasks, are instructed to be completed. Furthermore, during the experiments, subjects manipulated the steering wheel with two distinct postures and three different hand positions. Based on the experimental results, key parameters of the transfer function model are identified by using the Gauss-Newton algorithm. Based on the estimated model with identified parameters, investigation of system properties is then carried out. The characteristics of the driver neuromuscular system are discussed and compared with respect to different steering tasks, hand positions and driver postures. These experimental results with identified system properties provide a good foundation for the development of a haptic take-over control system for automated vehicles.Item Open Access Classification of barely visible impact damage in composite laminates using deep learning and pulsed thermographic inspection(Springer, 2023-01-31) Deng, Kailun; Liu, Haochen; Yang, Lichao; Addepalli, Sri; Zhao, YifanWith the increasingly comprehensive utilisation of Carbon Fibre-Reinforced Polymers (CFRP) in modern industry, defects detection and characterisation of these materials have become very important and draw significant research attention. During the past 10 years, Artificial Intelligence (AI) technologies have been attractive in this area due to their outstanding ability in complex data analysis tasks. Most current AI-based studies on damage characterisation in this field focus on damage segmentation and depth measurement, which also faces the bottleneck of lacking adequate experimental data for model training. This paper proposes a new framework to understand the relationship between Barely Visible Impact Damage features occurring in typical CFRP laminates to their corresponding controlled drop-test impact energy using a Deep Learning approach. A parametric study consisting of one hundred CFRP laminates with known material specification and identical geometric dimensions were subjected to drop-impact tests using five different impact energy levels. Then Pulsed Thermography was adopted to reveal the subsurface impact damage in these specimens and recorded damage patterns in temporal sequences of thermal images. A convolutional neural network was then employed to train models that aim to classify captured thermal photos into different groups according to their corresponding impact energy levels. Testing results of models trained from different time windows and lengths were evaluated, and the best classification accuracy of 99.75% was achieved. Finally, to increase the transparency of the proposed solution, a salience map is introduced to understand the learning source of the produced models.Item Open Access A clustering approach to detect faults with multi-component degradations in aircraft fuel systems(Elsevier, 2020-12-18) Zaporowska, Anna; Liu, Haochen; Skaf, Zakwan; Zhao, YifanAccurate fault diagnosis and prognosis can significantly increase the safety and reliability of engineering systems and also reduce the maintenance costs. There is very limited relative research reported on the fault diagnosis of a complex system with multi-component degradation. The Complex Systems (CS) problem, which features multiple components simultaneously and nonlinearly interacting with each other and corresponding environment on multiple levels, has become an essential challenge in system engineering. In CS, even a single component degradation could cause misidentification of the fault severity level and lead to serious consequences. This paper introduces a new test rig to simulate multi-component degradations of the aircraft fuel system. A data analysis approach based on machine learning classification of both the time and frequency domain features is then proposed to detect and identify the fault severity level of CS with multi-component degradation. Results show that a) the fault can be sensitively detected with an accuracy > 99%; b) the severity of fault can be identified with an accuracy of 100%.Item Open Access Code and Data for "A Dementia Classification Framework using Frequency and Time-frequency Features based on EEG signals"(Cranfield University, 2019-02-05 16:33) Zhao, YifanThis is the code and data that are used for the paper "A Dementia Classification Framework using Frequency and Time-frequency Features based on EEG signals", in IEEE Transactions on Neural Systems & Rehabilitation Engineering, available at https://doi.org/10.1109/TNSRE.2019.2909100.Item Open Access A coefficient clustering analysis for damage assessment of composites based on pulsed thermographic inspection(Elsevier, 2016-06-11) Zhao, Yifan; Tinsley, Lawrence; Addepalli, Sri; Mehnen, Jorn; Roy, RajkumarThis paper introduces a coefficient clustering analysis method to detect and quantitatively measure damage occurring in composite materials using pulsed thermographic inspection. This method is based on fitting a low order polynomial model for temperature decay curves, which (a) provides an enhanced visual confirmation and size measurement of the damage, (b) provides the reference point for sound material for further damage depth measurement, (c) and reduces the burden in computational time. The performance of the proposed method is evaluated through a practical case study with carbon fibre reinforced polymer (CFRP) laminates which were subjected to a drop impact test with varying energy levels. A novel method for reducing an entire thermogram sequence into a single image is introduced, which provides an enhanced visualisation of the damage area.Item Open Access A combined control systems and machine learning approach to forecasting iceberg flux off Newfoundland(MDPI, 2021-07-09) Ross, Jennifer B.; Bigg, Grant R.; Zhao, Yifan; Hanna, EdwardIcebergs have long been a threat to shipping in the NW Atlantic and the iceberg season of February to late summer is monitored closely by the International Ice Patrol. However, reliable predictions of the severity of a season several months in advance would be useful for planning monitoring strategies and also for shipping companies in designing optimal routes across the North Atlantic for specific years. A seasonal forecast model of the build-up of seasonal iceberg numbers has recently become available, beginning to enable this longer-term planning of marine operations. Here we discuss extension of this control systems model to include more recent years within the trial ensemble sample set and also increasing the number of measures of the iceberg season that are considered within the forecast. These new measures include the seasonal iceberg total, the rate of change of the seasonal increase, the number of peaks in iceberg numbers experienced within a given season, and the timing of the peak(s). They are predicted by a range of machine learning tools. The skill levels of the new measures are tested, as is the impact of the extensions to the existing seasonal forecast model. We present a forecast for the 2021 iceberg season, predicting a medium iceberg year.Item Open Access A compactness based saliency approach for leakages detection in fluorescein angiogram(Springer Verlag, 2016-07-26) Zhao, Yitian; Su, Pan; Yang, Jian; Zhao, Yifan; Zheng, Yalin; Wang, YongtianThis study has developed a novel saliency detection method based on compactness feature for detecting three common types of leakage in retinal fluorescein angiogram: large focal, punctate focal, and vessel segment leakage. Leakage from retinal vessels occurs in a wide range of retinal diseases, such as diabetic maculopathy and paediatric malarial retinopathy. The proposed framework consists of three major steps: saliency detection, saliency refinement and leakage detection. First, the Retinex theory is adapted to address the illumination inhomogeneity problem. Then two saliency cues, intensity and compactness, are proposed for the estimation of the saliency map of each individual superpixel at each level. The saliency maps at different levels over the same cues are fused using an averaging operator. Finally, the leaking sites can be detected by masking the vessel and optic disc regions. The effectiveness of this framework has been evaluated by applying it to different types of leakage images with cerebral malaria. The sensitivity in detecting large focal, punctate focal and vessel segment leakage is 98.1, 88.2 and 82.7 %, respectively, when compared to a reference standard of manual annotations by expert human observers. The developed framework will become a new powerful tool for studying retinal conditions involving retinal leakage.