Browsing by Author "Kuang, Boyu"
Now showing 1 - 20 of 22
Results Per Page
Sort Options
Item Open Access 3D reconstruction of rail tracks based on fusion of RGB and infrared sensors(IEEE, 2024-08-28) Wang, Yizhong; Kuang, Boyu; Durazo, Isidro; Zhao, YifanRail tracks, an essential part of the rail system, have remarkably demanded thorough inspections amid rising passenger volumes and high-speed rail development. Non-destructive testing (NDT), without disrupting train operations, aims to mitigate risks by employing safe physical properties like sound, electromagnetic, and light. However, each NDT technique is sensitive to specific damage types, offering limited diagnostic perspectives and placing considerable requirements on operators, resulting in a high cognitive load. To improve the above situation, this study proposes an innovative approach for rail inspection by developing a 3D RGB-T model that combines Visual Testing (VT) and Thermal Inspection (TI) through image registration, 3D reconstruction, sensor fusion, and non-destructive testing (NDT). Their fusion facilities a complementary assessment of rail tracks by capturing both surface texture and thermal radiation to identify damages effectively. The introduction of a novel RGB and IR registration method enables the spatial alignment of images from both, reconstructing the 3D RGB-T model. This model broadens the detection scope beyond the limitations of singular NDT methods, utilizing complementary data to locate and assess the damage extent effectively and accurately. This integrated approach reduces training requirements, minimizes human errors, and provides a clear and interpretable visualization of track conditions.Item Open Access Advanced semantic segmentation of aircraft main components based on transfer learning and data-driven approach(Springer, 2024-12-31) Thomas, Julien; Kuang, Boyu; Wang, Yizhong; Barnes, Stuart; Jenkins, KarlThe implementation of Smart Airport and Airport 4.0 visions relies on the integration of automation, artificial intelligence, data science, and aviation technology to enhance passenger experiences and operational efficiency. One essential factor in the integration is the semantic segmentation of the aircraft main components (AMC) perception, which is essential to maintenance, repair, and operations in aircraft and airport operations. However, AMC segmentation has challenges from low data availability, high-quality annotation scarcity, and categorical imbalance, which are common in practical applications, including aviation. This study proposes a novel AMC segmentation solution, employing a transfer learning framework based on a sophisticated DeepLabV3 architecture optimized with a custom-designed Focal Dice Loss function. The proposed solution remarkably suppresses the categorical imbalance challenge and increases the dataset variability with manually annotated images and dynamic augmentation strategies to train a robust AMC segmentation model. The model achieved a notable intersection over union of 84.002% and an accuracy of 91.466%, significantly advancing the AMC segmentation performance. These results demonstrate the effectiveness of the proposed AMC segmentation solution in aircraft and airport operation scenarios. This study provides a pioneering solution to the AMC semantic perception problem and contributes a valuable dataset to the community, which is fundamental to future research on aircraft and airport semantic perception.Item Open Access Classification of flow regimes using a neural network and a non-invasive ultrasonic sensor in an S-shaped pipeline-riser system(Elsevier, 2021-11-24) Nnabuife, Somtochukwu Godfrey; Kuang, Boyu; Rana, Zeeshan A.; Whidborne, James F.A method for classifying flow regimes is proposed that employs a neural network with inputs of extracted features from Doppler ultrasonic signals of flows using either the Discrete Wavelet Transform (DWT) or the Power Spectral Density (PSD). The flow regimes are classified into four types: annular, churn, slug, and bubbly flow regimes. The neural network used in this work is a feedforward network with 20 hidden neurons. The network comprises four output neurons, each of which corresponds to the target vector's element number. 13 and 40 inputs are used for features extracted from PSD and DWT respectively. Experimental data were collected from an industrial-scale multiphase flow facility. Using the PSD features, the neural network classifier misclassified 3 out of 31 test datasets in the classification and gave 90.3% accuracy, while only one dataset was misclassified with the DWT features, yielding an accuracy of 95.8%, thus showing the superiority of the DWT in feature extraction of flow regime classification. The approach demonstrates the applicability of a neural network and DWT for flow regime classification in industrial applications using a clamp-on Doppler ultrasonic sensor. The scheme has significant advantages over other techniques as only a non-radioactive and non-intrusive sensor is used. To the best of our knowledge, this is the first known successful attempt for the classification of liquid-gas flow regimes in an S-shape riser system using an ultrasonic sensor, PSD-DWTs features, and a neural network.Item Open Access A comparative analysis of different hydrogen production methods and their environmental impact(MDPI, 2023-11-29) Nnabuife, Somtochukwu Godfrey; Darko, Caleb Kwasi; Obiako, Precious Chineze; Kuang, Boyu; Sun, Xiaoxiao; Jenkins, Karl W.This study emphasises the growing relevance of hydrogen as a green energy source in meeting the growing need for sustainable energy solutions. It foregrounds the importance of assessing the environmental consequences of hydrogen-generating processes for their long-term viability. The article compares several hydrogen production processes in terms of scalability, cost-effectiveness, and technical improvements. It also investigates the environmental effects of each approach, considering crucial elements such as greenhouse gas emissions, water use, land needs, and waste creation. Different industrial techniques have distinct environmental consequences. While steam methane reforming is cost-effective and has a high production capacity, it is coupled with large carbon emissions. Electrolysis, a technology that uses renewable resources, is appealing but requires a lot of energy. Thermochemical and biomass gasification processes show promise for long-term hydrogen generation, but further technological advancement is required. The research investigates techniques for improving the environmental friendliness of hydrogen generation through the use of renewable energy sources. Its ultimate purpose is to offer readers a thorough awareness of the environmental effects of various hydrogen generation strategies, allowing them to make educated judgements about ecologically friendly ways. It can ease the transition to a cleaner hydrogen-powered economy by considering both technological feasibility and environmental issues, enabling a more ecologically conscious and climate-friendly energy landscape.Item Open Access A computational fluid dynamics study of flared gas for enhanced oil recovery using a micromodel(MDPI, 2022-12-19) Were, Stephanie; Nnabuife, Somtochukwu Godfrey; Kuang, BoyuThe current handling of gas associated with oil production poses an environmental risk. This gas is being flared off due to the technical and economic attractiveness of this option. As flared gases are mainly composed of methane, they have harmful greenhouse effects when released into the atmosphere. This work discusses the effectiveness of using this gas for enhanced oil recovery (EOR) purposes as an alternative to flaring. In this study, a micromodel was designed with properties similar to a sandstone rock with a porosity of 0.4, and computational fluid dynamics (CFD) techniques were applied to design an EOR system. Temperature effects were not considered in the study, and the simulation was run at atmospheric pressure. Five case studies were carried out with different interfacial tensions between the oil and gas (0.005 N/m, 0.017 N/m, and 0.034 N/m) and different injection rates for the gas (1 × 10−3 m/s, 1 × 10−4 m/s, and 1 × 10−6 m/s). The model was compared with a laboratory experiment measuring immiscible gas flooding. Factors affecting oil recoveries, such as the interfacial tension between oil and gas, the viscosity, and the pressure, were studied in detail. The results showed that the surface tension between the oil and gas interphase was a limiting factor for maximum oil recovery. The lower surface tension recovered 33% of the original oil in place. The capillary pressure was higher than the pressure in the micromodel, which lowered the amount of oil that was displaced. The study showed the importance of pressure maintenance to increase oil recovery for immiscible gas floods. It is recommended that a wider set of interfacial tensions between oil and gas be tested to obtain a range at which oil recovery is maximum for EOR with flared gas.Item Open Access A dataset for autonomous aircraft refueling on the ground (AGR)(IEEE, 2023-09-01) Kuang, Boyu; Barnes, Stuart; Tang, Gilbert; Jenkins, Karl W.Automatic aircraft ground refueling (AAGR) can improve the safety, efficiency, and cost-effectiveness of aircraft ground refueling (AGR), a critical and frequent operation on almost all aircraft. Recent AAGR relies on machine vision, artificial intelligence, and robotics to implement automation. An essential step for automation is AGR scene recognition, which can support further component detection, tracking, process monitoring, and environmental awareness. As in many practical and commercial applications, aircraft refueling data is usually confidential, and no standardized workflow or definition is available. These are the prerequisites and critical challenges to deploying and benefitting advanced data-driven AGR. This study presents a dataset (the AGR Dataset) for AGR scene recognition using image crawling, augmentation, and classification, which has been made available to the community. The AGR dataset crawled over 3k images from 13 databases (over 26k images after augmentation), and different aircraft, illumination, and environmental conditions were included. The ground-truth labeling is conducted manually using a proposed tree-formed decision workflow and six specific AGR tags. Various professionals have independently reviewed the AGR dataset to keep it no-bias. This study proposes the first aircraft refueling image dataset, and an image labeling software with a UI to automate the labeling workflow.Item Open Access Development of an integrated energy management system for off-grid solar applications with advanced solar forecasting, time-of-use tariffs, and direct load control(Elsevier, 2024-06-19) Falope, Tolulope Olumuyiwa; Lao, Liyun; Huo, Da; Kuang, BoyuEffectively managing and maximizing the integration of renewable energy sources is essential for a sustainable power grid due to the stochastic and intermittent nature of renewable energy generation. This study develops a comprehensive Integrated Energy Management System incorporating supply-demand side management in the form of time-of-use credit, direct load control, and generator control to enhance photovoltaic utilization in off-grid applications. A novel three-step solar energy forecasting approach is proposed in this paper, utilizing low-level data fusion and regression models to predict next-day photovoltaic generation with improved accuracy, and a rule-based decision algorithm is developed to correct forecast errors and manage loads dynamically. A techno-economic analysis covering a 20-year duration is carried out for scenarios with and without the integrated energy management system; three configurations are investigated for supplying an off-grid residential home, including diesel generator, diesel generator/photovoltaic system, and diesel generator/photovoltaic system/integrated energy management system. Results reveal that the hybrid configuration with integrated energy management system achieved 44 % and 46 % reductions in costs and carbon dioxide emissions compared to the diesel generator alone, and 8 % and 9 % compared to the diesel generator/photovoltaic setup respectively. The Integrated Energy Management System further enhanced photovoltaic utilisation rate by over 113 % when compared to the diesel generator/photovoltaic system. Further evaluations include customer behaviour impacts, demonstrating that a fully automated system with 100 % compliance significantly outperforms systems with manual customer control, highlighting the detrimental effect of overrides on the efficiency of direct load control. The flexibility of the Integrated Energy Management System framework allows potential adaptation for on-grid applications, showcasing its utility in diverse operational contexts.Item Open Access Development of gas-liquid flow regimes identification using a noninvasive ultrasonic sensor, belt-shape features, and convolutional neural network in an S-shaped riser(IEEE, 2021-07-14) Nnabuife, Somtochukwu Godfrey; Kuang, Boyu; Whidborne, James F.; Rana, Zeeshan A.The problem of classifying gas-liquid two-phase flow regimes from ultrasonic signals is considered. A new method, belt-shaped features (BSFs), is proposed for performing feature extraction on the preprocessed data. A convolutional neural network (CNN/ConvNet)-based classifier is then applied to categorize into one of the four flow regimes: 1) annular; 2) churn; 3) slug; or 4) bubbly. The proposed ConvNet classifier includes multiple stages of convolution and pooling layers, which both decrease the dimension and learn the classification features. Using experimental data collected from an industrial-scale multiphase flow facility, the proposed ConvNet classifier achieved 97.40%, 94.57%, and 94.94% accuracy, respectively, for the training set, testing set, and validation set. These results demonstrate the applicability of the BSF features and the ConvNet classifier for flow regime classification in industrial applications.Item Unknown Gas-liquid flow regimes identification using non-intrusive Doppler ultrasonic sensor and convolutional recurrent neural networks in an S-shaped riser(Elsevier, 2022-01-19) Kuang, Boyu; Nnabuife, Somtochukwu Godfrey; Sun, Shuang; Whidborne, James F.; Rana, Zeeshan A.The problem of gas-liquid (two-phase) flow regime identification in an S-shaped riser using an ultrasonic sensor and convolutional recurrent neural networks (CRNN) is addressed. This research systematically evaluates three different schemes with four CRNN-based classifiers over fourteen experiments. Four metrics are used as the evaluation criteria: categorical accuracy, categorical cross-entropy, mean square error (MSE), and computation graph complexity. Compared with existing results, a compatible performance is achieved while considerably reducing the model complexity. The testing and validation accuracies were 98.13% and 98.06%, while the complexity decreased by 98.4% (only 117,702 parameters). The proposed approach is i) accurate, low complexity, and non-intrusive and hence suitable for industry, and ii) could provide a benchmark for flow regime identification.Item Metadata only Integration of renewable energy sources in tandem with electrolysis: a technology review for green hydrogen production(Elsevier, 2024) Nnabuife, Somtochukwu Godfrey; Hamzat, Abdulhammed K.; Whidborne, James; Kuang, Boyu; Jenkins, Karl W.The global shift toward sustainable energy solutions emphasises the urgent need to harness renewable sources for green hydrogen production, presenting a critical opportunity in the transition to a low-carbon economy. Despite its potential, integrating renewable energy with electrolysis to produce green hydrogen faces significant technological and economic challenges, particularly in achieving high efficiency and cost-effectiveness at scale. This review systematically examines the latest advancements in electrolysis technologies—alkaline, proton exchange membrane electrolysis cell (PEMEC), and solid oxide—and explores innovative grid integration and energy storage solutions that enhance the viability of green hydrogen. The study reveals enhanced performance metrics in electrolysis processes and identifies critical factors that influence the operational efficiency and sustainability of green hydrogen production. Key findings demonstrate the potential for substantial reductions in the cost and energy requirements of hydrogen production by optimising electrolyser design and operation. The insights from this research provide a foundational strategy for scaling up green hydrogen as a sustainable energy carrier, contributing to global efforts to reduce greenhouse gas emissions and advance toward carbon neutrality. The integration of these technologies could revolutionise energy systems worldwide, aligning with policy frameworks and market dynamics to foster broader adoption of green hydrogen.Item Open Access Labelled image dataset of the pressure refuelling port from an Airbus A320(Cranfield University, 2023-10-05 07:48) Kuang, BoyuThe dataset is for a conference publication. The file contains two types of files: - images (".jpg") - labels (".json")Item Open Access Non-intrusive classification of gas-liquid flow regimes in an S-shaped pipeline-riser using doppler ultrasonic sensor and deep neural networks(Elsevier, 2020-07-26) Nnabuife, Somtochukwu Godfrey; Kuang, Boyu; Whidborne, James F.; Rana, ZeeshanThe problem of predicting the regime of a two-phase flow is considered. An approach is proposed that classifies the flow regime using Deep Neural Networks (DNNs) operating on features extracted from Doppler ultrasonic signals of the flow using the Fast Fourier Transform (FFT) is proposed. The features extracted are categorised into one of the four flow regime classes: the annular, churn, slug, and bubbly flow regimes. The scheme was tested on signals from an experimental facility. To increase the number of samples without losing key classification information, this paper proposes a Twin-window Feature Extraction (TFE) technique. To further distinguish the performance of the proposed approach, the classifier was compared to four conventional machine learning classifiers: namely, the AdaBoost classifier, bagging classifier, extra trees classifier, and decision tree classifier. Using the TFE features, the DNNs classifier achieved a higher recognition accuracy of 99.01% and greater robustness for the overfitting challenge, thereby showing the superiority of the DNNs in flow regime classification when compared to the four conventional machine-learning classifiers, which had classification accuracies of 55.35%, 86.21%, 82.41%, and 80.03%, respectively. This approach demonstrates the application of DNNs for flow regime classification in chemical and petroleum engineering fields, using a clamp-on Doppler ultrasonic sensor. This appears to be the first known successful attempt to identify gas-liquid flow regimes in an S-shaped riser using Continuous Wave Doppler Ultrasound (CWDU) and DNNsItem Open Access A novel aircraft wing inspection framework based on multiple view geometry and convolutional neural network(Council of European Aerospace Societies (CEAS), 2020-02-28) Kuang, Boyu; Rana, Zeeshan; Zhao, YifanTo achieve greener and safer aeronautical operations, this paper considers the problem of reconstructing the three-dimensional (3D) geometric structure of aeronautical components. A novel framework that recovers the 3D shapes by means of convolutional neural network (ConvNets) and multiple view geometry (MVG) operating on Mask-R-CNN-segmented two-dimensional images is proposed. To achieve more accurate 3D aircraft’s surface and exclude the invalid background structures, this paper innovatively integrates the environmental robustness of ConvNets and geometric adaptation of Mask-R- CNN into the MVG theory. The preliminary experiments show that the proposed framework is visual-comfortable, and it also accurately reconstructs the regions with damage to catch up with the inspection purpose.Item Open Access OG-SLAM: a real-time and high-accurate monocular visual SLAM framework(Peertechz, 2022-07-26) Kuang, Boyu; Chen, Yuheng; Rana, Zeeshan A.The challenge of improving the accuracy of monocular Simultaneous Localization and Mapping (SLAM) is considered, which widely appears in computer vision, autonomous robotics, and remote sensing. A new framework (ORB-GMS-SLAM (or OG-SLAM)) is proposed, which introduces the region-based motion smoothness into a typical Visual SLAM (V-SLAM) system. The region-based motion smoothness is implemented by integrating the Oriented Fast and Rotated Brief (ORB) features and the Grid-based Motion Statistics (GMS) algorithm into the feature matching process. The OG-SLAM significantly reduces the absolute trajectory error (ATE) on the key-frame trajectory estimation without compromising the real-time performance. This study compares the proposed G-SLAM to an advanced V-SLAM system (ORB-SLAM2). The results indicate the highest accuracy improvement of almost 75% on a typical RGB-D SLAM benchmark. Compared with other ORB-SLAM2 settings (1800 key points), the OG-SLAM improves the accuracy by around 20% without losing performance in real-time. The OG-SLAM framework has a significant advantage over the ORB-SLAM2 system in that it is more robust for rotation, loop-free, and long ground-truth length scenarios. Furthermore, as far as the authors are aware, this framework is the first attempt to integrate the GMS algorithm into the V-SLAM.Item Open Access The prospects of hydrogen in achieving net zero emissions by 2050: a critical review(Elsevier, 2023-05-25) Nnabuife, Godfrey Somtochukwu; Oko, Eni; Kuang, Boyu; Bello, Abdulrauf; Onwualu, Azikiwe Peter; Oyagha, Sherry; Whidborne, James F.Hydrogen (H2) usage was 90 tnes (Mt) in 2020, almost entirely for industrial and refining uses and generated almost completely from fossil fuels, leading to nearly 900 Mt of carbon dioxide emissions. However, there has been significant growth of H2 in recent years. Electrolysers' total capacity, which are required to generate H2 from electricity, has multiplied in the past years, reaching more than 300 MW through 2021. Approximately 350 projects reportedly under construction could push total capacity to 54 GW by the year 2030. Some other 40 projects totalling output of more than 35 GW are in the planning phase. If each of these projects is completed, global H2 production from electrolysers could exceed 8 Mt by 2030. It's an opportunity to take advantage of H2S prospects to be a crucial component of a clean, safe, and cost-effective sustainable future. This paper assesses the situation regarding H2 at the moment and provides recommendations for its potential future advancement. The study reveals that clean H2 is experiencing significant, unparalleled commercial and political force, with the amount of laws and projects all over the globe growing quickly. The paper concludes that in order to make H2 more widely employed, it is crucial to significantly increase innovations and reduce costs. The practical and implementable suggestions provided to industries and governments will allow them to fully capitalise on this growing momentum.Item Open Access Prospects of low and zero-carbon renewable fuels in 1.5-degree net zero emission actualisation by 2050: a critical review(Elsevier, 2022-10-12) Anika, Ogemdi Chinwendu; Nnabuife, Somtochukwu Godfrey; Bello, Abdulrauf; Okoroafor, Esuru Rita; Kuang, Boyu; Villa, RaffaellaThe Paris Climate Agreement seeks to keep global temperature increases under 2° Celsius, ideally 1.5° Celsius. This goal necessitates significant emission reductions. By 2030, emissions are expected to range between 52 and 58 GtCO2e from their 2016 level of approximately 52 GtCO2e. This review paper explores a number of low and zero-carbon renewable fuels, such as hydrogen, green ammonia, green methanol, biomethane, natural gas, and synthetic methane (with natural gas and synthetic methane subject to CCUS both at processing and at final use) as alternative solutions for providing a way to rebalance transition paths in order to achieve the goals of the Paris Agreement while also reaping the benefits of other sustainability targets. The results show renewables will need to account for approximately 90% of total electricity generation by 2050 and approximately 25% of non-electric energy usage in buildings and industry. However, low and zero-carbon renewable fuels currently only contributes about 15% to the global energy shares, and it will take about 10% more capacity to reach the 2050 goal. The transportation industry will need to take important steps toward energy efficiency and fuel switching in order to achieve the 20% emission reduction. Therefore, significant new commitments to efficient low-carbon alternatives will be necessary to make this enormous change. According to this paper, investing in energy efficiency and low-carbon alternative energy must rise by a factor of about five by 2050 in comparison to 2015 levels if the 1.5 °C target is to be realised.Item Open Access Pseudo-image-feature-based identification benchmark for multi-phase flow regimes(Elsevier, 2020-12-08) Kuang, Boyu; Nnabuife, Somtochukwu Godfrey; Rana, ZeeshanMultiphase flow is a prevalent topic in many disciplines, and flow regime identification is an essential foundation in multiphase flow research. Computer vision and deep learning have achieved numerous excellent models, but many have not demonstrated satisfactory performance in fundamental research, including flow regime identification. This research proposes an advanced pseudo-image feature (PIF) as the flow regime descriptor and a benchmark of multiple deep learning classifiers. The PIF simulates the image format and compactly encodes the flow regime to a pseudo-image, which explicitly displays the implicit flow regime signals. This research further evaluates three proposed and five existing popular deep learning classifiers. The proposed benchmark provides a baseline for applying deep learning in flow regime identification. The proposed fully convolutional network (FCN) classifier achieved state-of-the-art performance, and the testing and verification accuracy respectively reached 99.95% and 99.54%. This research suggests that PIF has an excellent capability for flow regime representation, and the proposed deep learning classifiers achieve superior performance in flow regime identification compared to the existing classifiers. Industries can utilize the proposed multiphase flow identification technology to obtain greater production efficiency, productivity, and financial gainItem Open Access Rock segmentation in the navigation vision of the planetary rovers(MDPI, 2021-11-24) Kuang, Boyu; Wisniewski, Mariusz; Rana, Zeeshan A.; Zhao, YifanVisual navigation is an essential part of planetary rover autonomy. Rock segmentation emerged as an important interdisciplinary topic among image processing, robotics, and mathematical modeling. Rock segmentation is a challenging topic for rover autonomy because of the high computational consumption, real-time requirement, and annotation difficulty. This research proposes a rock segmentation framework and a rock segmentation network (NI-U-Net++) to aid with the visual navigation of rovers. The framework consists of two stages: the pre-training process and the transfer-training process. The pre-training process applies the synthetic algorithm to generate the synthetic images; then, it uses the generated images to pre-train NI-U-Net++. The synthetic algorithm increases the size of the image dataset and provides pixel-level masks—both of which are challenges with machine learning tasks. The pre-training process accomplishes the state-of-the-art compared with the related studies, which achieved an accuracy, intersection over union (IoU), Dice score, and root mean squared error (RMSE) of 99.41%, 0.8991, 0.9459, and 0.0775, respectively. The transfer-training process fine-tunes the pre-trained NI-U-Net++ using the real-life images, which achieved an accuracy, IoU, Dice score, and RMSE of 99.58%, 0.7476, 0.8556, and 0.0557, respectively. Finally, the transfer-trained NI-U-Net++ is integrated into a planetary rover navigation vision and achieves a real-time performance of 32.57 frames per second (or the inference time is 0.0307 s per frame). The framework only manually annotates about 8% (183 images) of the 2250 images in the navigation vision, which is a labor-saving solution for rock segmentation tasks. The proposed rock segmentation framework and NI-U-Net++ improve the performance of the state-of-the-art models. The synthetic algorithm improves the process of creating valid data for the challenge of rock segmentation. All source codes, datasets, and trained models of this research are openly available in Cranfield Online Research Data (CORD).Item Open Access Self-supervised learning-based two-phase flow regime identification using ultrasonic sensors in an S-shape riser(Elsevier, 2023-09-07) Kuang, Boyu; Nnabuife, Somtochukwu G.; Whidborne, James F.; Sun, Shuang; Zhao, Junjie; Jenkins, Karl W.Two-phase flow regime identification is an essential transdisciplinary topic that spans digital signal processing, artificial intelligence, chemical engineering, and energy. Multiphase flow systems significantly impact pipeline safety, heat transfer, and pressure drop; therefore, precisely identifying the governing flow regime is crucial for effective modeling and design. However, it is challenging due to the geometrical complexity of flow regimes in multiphase flow. With the advances in sensor measurement and machine learning, applying non-destructive tests and self-supervised learning to practical industrial problems has become technically feasible and cost-effective. This study applies a weak-supervised learning-based two-phase flow regime identification solution using a non-destructive tests ultrasonic sensor in an S-shape riser experimental bed by proposing a self-supervised feature extraction algorithm. The proposed self-supervised feature extraction algorithm reduces time/labor consumption and human error in data annotation using SSL, which provides full supervision without manual annotation. The self-supervised feature extraction algorithm uses a bottlenecked neural network and encoder-decoder structure to extract compact features. The self-supervised feature extraction algorithm performance is evaluated using an established convolutional neural network-based classifier. The source data was collected from a 10 × 50 m riser experimental rig. The dataset is made available to the community as part of this study. The performance of the approach is comparable with state-of-the-art methods and is also the first successful attempt to apply self-supervised learning to multiphase flow regime ultrasonic signal identification. This study achieved 98.84%, 0.000663, 0.00312, and 7.71 × 10^5 in accuracy, root mean square error, categorical cross-entropy, and model complexity, respectively. The practical experiment justifies the robustness, fairness, and practicability in the practical application environment. The proposed self-supervised feature extraction brings new approaches and inspirations for the feature extraction step in identifying a two-phase flow regime, and it will be beneficial to generalize this study in different riser shapes in the future.Item Open Access Semantic terrain segmentation in the navigation vision of planetary rovers – a systematic literature review(MDPI, 2022-11-01) Kuang, Boyu; Gu, Chengzhen; Rana, Zeeshan A.; Zhao, Yifan; Sun, Shuang; Nnabuife, Somtochukwu GodfreyBackground: The planetary rover is an essential platform for planetary exploration. Visual semantic segmentation is significant in the localization, perception, and path planning of the rover autonomy. Recent advances in computer vision and artificial intelligence brought about new opportunities. A systematic literature review (SLR) can help analyze existing solutions, discover available data, and identify potential gaps. Methods: A rigorous SLR has been conducted, and papers are selected from three databases (IEEE Xplore, Web of Science, and Scopus) from the start of records to May 2022. The 320 candidate studies were found by searching with keywords and bool operators, and they address the semantic terrain segmentation in the navigation vision of planetary rovers. Finally, after four rounds of screening, 30 papers were included with robust inclusion and exclusion criteria as well as quality assessment. Results: 30 studies were included for the review, and sub-research areas include navigation (16 studies), geological analysis (7 studies), exploration efficiency (10 studies), and others (3 studies) (overlaps exist). Five distributions are extendedly depicted (time, study type, geographical location, publisher, and experimental setting), which analyzes the included study from the view of community interests, development status, and reimplementation ability. One key research question and six sub-research questions are discussed to evaluate the current achievements and future gaps. Conclusions: Many promising achievements in accuracy, available data, and real-time performance have been promoted by computer vision and artificial intelligence. However, a solution that satisfies pixel-level segmentation, real-time inference time, and onboard hardware does not exist, and an open, pixel-level annotated, and the real-world data-based dataset is not found. As planetary exploration projects progress worldwide, more promising studies will be proposed, and deep learning will bring more opportunities and contributions to future studies. Contributions: This SLR identifies future gaps and challenges by proposing a methodical, replicable, and transparent survey, which is the first review (also the first SLR) for semantic terrain segmentation in the navigation vision of planetary rovers.