The field of optical microscopy spans across numerous industries and research domains, ranging from education to healthcare, quality inspection and analysis. Nonetheless, a key limitation often cited by optical microscopists refers to the limit of its lateral resolution (typically defined as 200nm), with potential circumventions involving either costly external modules (e.g. confocal scan heads, etc) and/or specialized techniques [e.g. super-resolution (SR) fluorescent microscopy]. Addressing these challenges in a normal (non-specialist) context thus remains an aspect outside the scope of most microscope users facilities. This study thus seeks to evaluate an alternative economical approach to achieving SR optical microscopy, involving non-fluorescent phase-modulated microscopical modalities such as Zernike phase contrast (PCM) and differential interference contrast (DIC) microscopy. Two in silico deep neural network (DNN) architectures which we developed previously (termed O-Net and Theta-Net) are assessed on their abilities to resolve a custom-fabricated test target containing nanoscale features calibrated via atomic force microscopy (AFM). The results of our study demonstrate that although both O-Net and Theta-Net seemingly performed well when super-resolving these images, they were complementary (rather than competing) approaches to be considered for image SR, particularly under different image signal-to-noise ratios (SNRs). High image SNRs favoured the application of O-Net models, while low SNRs inclined preferentially towards Theta-Net models. These findings demonstrate the importance of model architectures (in conjunction with the source image SNR) on model performance and the SR quality of the generated images where DNN models are utilized for non-fluorescent optical nanoscopy, even where the same training dataset number of epochs are being used.

Optical microscopy is widely regarded to be an indispensable tool in healthcare and manufacturing quality control processes, although its inability to resolve structures separated by a lateral distance under ~200 nm has culminated in the emergence of a new field named fluorescence nanoscopy, while this too is prone to several caveats (namely phototoxicity, interference caused by exogenous probes and cost). In this regard, we present a triplet string of concatenated O-Net (‘bead’) architectures (termed ‘Θ-Net’ in the present study) as a cost-efficient and non-invasive approach to enhancing the resolution of non-fluorescent phase-modulated optical microscopical images in silico. The quality of the afore-mentioned enhanced resolution (ER) images was compared with that obtained via other popular frameworks (such as ANNA-PALM, BSRGAN and 3D RCAN), with the Θ-Net-generated ER images depicting an increased level of detail (unlike previous DNNs). In addition, the use of cross-domain (transfer) learning to enhance the capabilities of models trained on differential interference contrast (DIC) datasets [where phasic variations are not as prominently manifested as amplitude/intensity differences in the individual pixels unlike phase-contrast microscopy (PCM)] has resulted in the Θ-Net-generated images closely approximating that of the expected (ground truth) images for both the DIC and PCM datasets. This thus demonstrates the viability of our current Θ-Net architecture in attaining highly resolved images under poor signal-to-noise ratios while eliminating the need for a priori PSF and OTF information, thereby potentially impacting several engineering fronts (particularly biomedical imaging and sensing, precision engineering and optical metrology).

The enhancement of the mechanical and thermal characteristics of 3D printed polylactic acid (PLA) composites reinforced by coir fiber powder (CFP) has been investigated by varying the weight percentage (wt%) of the reinforcement and annealing process. CFP/PLA composite filaments with CFP compositions of 0.1, 0.3, and 0.5 wt% were fabricated. These filaments were used to print CFP/PLA test specimens. The specimens were annealed at 90 degrees C for 120 min in a hot air oven followed by cooling at room temperature. Mechanical, morphological, crystalline, and thermal characterizations were conducted on these specimens. The tensile and flexural strength of neat PLA were observed as 49.7 and 82.4 MPa which decreased by 6.4% and 8.13% respectively for printed composite specimens with 0.5 wt% CFP as reinforcement material. On the other hand, the annealed CFP/PLA composite specimen, with 0.1 wt% CFP as a reinforcement material, demonstrated higher tensile and flexural strength. Specifically, it exhibited a maximum tensile strength of 56.4 MPa and a maximum flexural strength of 92.9 MPa, which are 13.5% and 12.7% higher, respectively, than neat PLA. These strengths are 15.5% and 16.7% higher, respectively, than those of the unannealed CFP/PLA composite specimen with the same wt% of CFP reinforcement. The annealing process increased the crystallinity of composites by enhancing the crystallinity index (63%) and crystalline size (6.7 nm). The high thermal stability of composites (with a glass transition temperature of 256(degrees)C) makes them suitable for applications in food and medical packaging.

We present herein a triplet string of concatenated O-Net (‘ bead ’) architectures (formulated as discussed in our previous study) which we term ‘Θ-Net’ as a means of improving the viability of generated super-resolved (SR) images in silico . In the present study, we assess the quality of the afore-mentioned SR images with that obtained via other popular frameworks (such as ANNA-PALM, BSRGAN and 3D RCAN). Models developed from our proposed framework result in images which more closely approach the gold standard of the SEM-verified test sample as a means of resolution enhancement for optical microscopical imaging, unlike previous DNNs. In addition, cross-domain (transfer) learning was also utilized to enhance the capabilities of models trained on DIC datasets, where phasic variations are not as prominently manifested as amplitude/intensity differences in the individual pixels [unlike phase contrast microscopy (PCM)]. The present study thus demonstrates the viability of our current multi-paradigm architecture in attaining ultra-resolved images under poor signal-to-noise ratios, while eliminating the need for a priori PSF & OTF information. Due to the wide-scale use of optical microscopy for inspection & quality analysis in various industry sectors, the findings of this study would be anticipated to exhibit a far-ranging impact on several engineering fronts.### Competing Interest StatementThe authors have declared no competing interest.

AbstractWe present herein atripletstring of concatenated O-Net (‘bead’) architectures (formulated as discussed in our previous study) which we term ‘Θ-Net’ as a means of improving the viability of generated super-resolved (SR) imagesin silico. In the present study, we assess the quality of the afore-mentioned SR images with that obtained via other popular frameworks (such as ANNA-PALM, BSRGAN and 3D RCAN). Models developed from our proposed framework result in images which more closely approach the gold standard of the SEM-verified test sample as a means of resolution enhancement for optical microscopical imaging, unlike previous DNNs. In addition,cross-domain (transfer) learningwas also utilized to enhance the capabilities of models trained on DIC datasets, where phasic variations are not as prominently manifested as amplitude/intensity differences in the individual pixels [unlike phase contrast microscopy (PCM)]. The present study thus demonstrates the viability of our current multi-paradigm architecture in attaining ultra-resolved images under poor signal-to-noise ratios, while eliminating the need fora prioriPSF & OTF information. Due to the wide-scale use of optical microscopy for inspection & quality analysis in various industry sectors, the findings of this study would be anticipated to exhibit a far-ranging impact on several engineering fronts.

Hydraulic turbines contribute to 60% of renewable energy in the world; however, they also entail some adverse effects on the aquatic ecology system. One such effect is their excessive noise and vibration. To minimize this effect, one of the most effective and feasible solutions is to modify the design of the turbine rotor blade by introducing a skew. In this study, two 0.3-meter tidal turbines with 0-degree (no-skewness) and positive 90-degree skewness made of stainless steel 316L were designed and printed using a 3-D printing powder bed fusion technique. These rotors were then tested at the Emerson Cavitation Tunnel (ECT) at Newcastle University, UK, and the variation in the skewness of the blades of the turbines as a function of the power coefficient on a given tip speed ratio (TSR) value was ascertained. Results showed that the highly skewed rotor had significantly lower drag and torque fluctuations, with a slight decrease in efficiency compared to the non-skewed one, which warrants further investigation on the effect of added skew to reduce vibration and noise. Numerical simulations were also performed for verification and validation of the experimental tests, using the H45 dynamometer at the ECT. A comprehensive software code for propellers and tidal turbines, ROTORYSICS, was used to examine the cavitation effect of the two rotors; a comparison was made for both, with and without cavitation. The results indicate that for a high immersion depth of tidal turbine rotors, cavitation rarely occurs, but for hydrokinetic turbines that are installed on dams in rivers and falls, cavitation could be a serious concern. It was concluded that the 0-degree skewed rotor is more hydrodynamically efficient than the 90-degree skewed rotor.

The present work focusses on developing a completely sustainable, ecological, biodegradable, and eco-friendly composite material for edible as well as health product packaging utilizing an existing production technology. Ground chitin and chitosan powder were blended with eco-friendly polylactic acid (PLA) pellets and transformed into filaments by an extrusion process; thereafter composite parts produced by 3D printing were subjected to mechanical testing, morphological examination, crystallographic examination, chemical composition analysis and thermal analysis. Composite specimens containing chitin and chitosan, respectively, were obtained with varying mass fraction. The 3D printed composite specimen containing respectively chitin and chitosan yielded reduced tensile and flexural properties; for instance, for Chitin/PLA composite with 0.5 wt% (chitin), the tensile strength and flexural strength were 6.9 MPa and 10.4 MPa respectively. The addition of chitin reinforcement to the eco-friendly composite led to increased density with a maximum value of 1.285 g/cm3 for Chitin/PLA composite at 0.5 wt% of chitin reinforcement. However, both Chitosan/PLA and Chitin/PLA composite exhibited enhanced ductility compared to neat PLA. Additionally, the micrographs of the composites revealed the presence of voids and impurity-like particle reflecting decomposed chitin and chitosan. Insights into the chemical bonding and composition were gained by analysing the Fourier transform infrared spectra; analysis of the X-ray diffraction pattern revealed the crystal size and crystallinity index with values of 7.7 nm and 62%, respectively (e.g. Chitin/PLA composite at 0.5 wt%). Finally, the thermogravimetric analysis revealed that both composite materials exhibited a thermal stability of around 257 °C. Altogether, these findings support the use of the Chitosan/PLA and Chitin/PLA composites in applications such as food product packaging.

A new portable resin injection device has been developed to repair barely visible impact damage (BVID) in carbon fiber reinforced polymer (CFRP) laminates. Repair was conducted on damaged 16- and 24-ply laminates at two different ambient pressures, that is, atmospheric pressure and vacuum, using three different adhesive types, that is, neat epoxy, epoxy blended with respective halloysite nanotubes (E1HNT) and carbon nanotubes (NF100). The repair effectiveness was assessed by infrared thermography and in-plane compression testing. All damaged laminates suffered significant degradation in fracture properties as compared to pristine laminates. According to Taguchi method, the main factor affecting the repair efficiency is the laminate ply number. The findings revealed that the optimal repair parameters corresponds to using neat epoxy or E1HNT in vacuum could restore the elastic properties and using E1HNT adhesive at atmospheric pressure could restore the fracture properties of damaged 16-ply laminate after repair. Regardless of repair parameters, the repair efficiency for the mechanical properties of 24-ply laminates revealed minimal restoration to the laminate pristine conditions after repairing the damaged laminates due to the complexity of BVID for allowing the adhesive to infiltrate and filled the crack structures. The choice of repair parameters was discussed based on the effectiveness of restoring elastic and fracture properties.

Convolutional neural networks (CNN) have been utilized to detect anomalies on the railway track surfaces whose conditions must be monitored to ensure the safety of railway systems. While CNN has advantages over conventional image processing methods in self-learning features for detecting railway track anomalies, the CNN model and parameters still need to be carefully constructed and examined for the effective application with railway track images. This study presents a systematic investigation of CNN model parameters for detecting anomalies on railway tracks. Parameters such as number of convolutional layers, convolutional kernel size, pooling kernel size and number of epochs were examined. Experiments and analyses were performed to determine how these parameters affect the detection accuracy. The experimental procedures and findings demonstrated the effects of individual parameters, as well as the potential interactions between the factors; thus, systematic procedures are needed to investigate and improve CNN models deployed to detect and classify anomalies on railway tracks.

We present a fast and precise deep-learning architecture, which we term O-Net, for obtaining super-resolved images from conventional phase-modulated optical microscopical techniques, such as phase-contrast microscopy and differential interference contrast microscopy. O-Net represents a novel deep convolutional neural network that can be trained on both simulated and experimental data, the latter of which is being demonstrated in the present context. The present study demonstrates the ability of the proposed method to achieve super-resolved images even under poor signal-to-noise ratios and does not require prior information on the point spread function or optical character of the system. Moreover, unlike previous state-of-the-art deep neural networks (such as U-Nets), the O-Net architecture seemingly demonstrates an immunity to network hallucination, a commonly cited issue caused by network overfitting when U-Nets are employed. Models derived from the proposed O-Net architecture are validated through empirical comparison with a similar sample imaged via scanning electron microscopy (SEM) and are found to generate ultra-resolved images which came close to that of the actual SEM micrograph.

Growing concerns like depleting mineral resources, increased materials wastage, and structural light-weighting requirements due to emission control regulations drive the development of sustainable metal matrix composites. Al and Mg based alloys with relatively lower melting temperatures qualify for recycling applications and hence are considered as the matrix material for developing sustainable composites. The recent trend also explores various industrial by-products and agricultural wastes as green reinforcements, and this article presents insights on the properties of Al and Mg based sustainable metal matrix composites with special emphasis on green reinforcements and processing methods.

The detection and classification of rail track surface defects is critical to the safety and productivity of urban transport operations. The requirement of fast and accurate detection and classification from the huge number of images taken drives for an automated solution. A method for detection and classification of rail surface anomalies and defects adopting deep learning approach based on the convolutional neural networks has been presented in this paper. The training and testing images are acquired from an automated video recording setup on the train. We proposed a convolutional neural network model trained to learn the features which will then be utilized to detect and classify defects from images taken by image acquisition device mounted on the train. The experimental results are promising and can be integrated to the current workflow of rail maintenance operations to improve the productivity.

Rail track surfaces could suffer from defects such as abrasion and deformation, thus to ensure rail system safety, the conditions of rail tracks must be monitored. With the advancement of deep learning and computer vision technologies, automatic detection and classification techniques are being tested to replace or complement manual patrolling for productivity improvement. However, classic neural network approaches require a large amount of data which could be time-consuming and limit the application of deep learning techniques. This paper proposes applying one-shot learning using a Siamese convolutional neural network to the identification of rail surface defects. The results show the reduced requirement of training speed and possess potentials for real-time applications.

The aim of this study is to develop a completely sustainable, biodegradable, eco-friendly and green composite material for packaging of food and medical products by an additive manufacturing technique such as 3D printing. This report presents the mechanical, crystalline, chemical bonding and thermal characteristics of a novel pineapple leaf fiber (PALF) reinforcing polylactic acid (PLA) green composite manufactured by 3D printing technique. Both powdered raw and alkali-treated PALFs were respectively mixed with PLA and extruded as filaments for 3D printing into composite test specimens. The characterization study reveals that the 3D printed composite with 3 wt% (alkalitreated) PALF reinforcement exhibited maximum tensile and flexural characteristics. The density of the 3D printed composite specimens increased with increase in wt% of PALF. On the other hand, the 3D printed composite specimens blended with raw PALF showed enhanced elongation at break compared to alkali-treated PALF reinforced composite specimens. The microstructural images of the 3D printed composite specimens confirms the existence of impurities, voids and fiber degradation phenomenon. The Fourier transform infrared spectra revealed the chemical bonding nature of the 3D printed composite specimens. The X-Ray diffraction was used to calculate the crystalline size and crystallinity index. Thermogravimetric analysis reveals that the 3D printed composite specimen possessed adequate thermal stability for use in packaging applications.

The panic buying during Covid-19 caused farmers to amped-up production. However, farm equipment is costly to purchase. Therefore, some farmers utilized Additive Manufacturing (AM) to manufacture farming tools at low cost. However, the lack of in-situ monitoring in AM to stop printing failed parts can waste materials and time. Thus, this research aims to deploy a low-cost smart remote monitoring system using OctoPrint and Node-red to integrate a 3D printer and Teachable Machine and train a model to pre-emptively detect print errors. The result was satisfactory as the 3D printer stopped when the camera detected a defect with 75% accuracy. Furthermore, the user can easily customize the model to enhance the system versatility via the developed code-free platform.

Rail track is a critical component of rail systems. Accidents or interruptions caused by rail track anomalies usually possess severe outcomes. Therefore, rail track condition monitoring is an important task. Over the past decade, deep learning techniques have been rapidly developed and deployed. In the paper, we review the existing literature on applying deep learning to rail track condition monitoring. Potential challenges and opportunities are discussed for the research community to decide on possible directions. Two application cases are presented to illustrate the implementation of deep learning to rail track condition monitoring in practice before we conclude the paper.

Robots have been reportedly seen serving food in several restaurants in many parts of the world. New ventures have been deploying mechanical partners which promote the growth in service robotics. However, robots are considerably incompetent when it comes to beverage and soup delivery. The physical challenge behind the clumsy motion of these machines is found to be its jerky motion control. Jerk control solutions are widely studied in a constrained environment but not well introduced in dynamic environments. In this paper, we will begin by examining developed kinematics solutions, open-source packages from Robot Operating System and the constraints of motion planning. The proposed solution in this paper provides a quick system response with jerk limits using spline velocity profiles. The solution will introduce the concepts of a state machine design that enables the robot to behave and move reactively; effectively balancing its desired velocity and position without spilling a drop of customer satisfaction. Experiments have proven that robots can move at higher velocity without any crashing, spilling, or docking issues. The smooth velocity control proposed will improve the capabilities of waiter robot and service operations in restaurants.

With the development of metal matrix composites (MMCs), the mechanical properties of MMCs with a small volume fraction of nano-sized particles were found to be even superior to that with a larger content of micron-sized particles. However, the improved mechanical properties of MMCs bring tremendous challenges such as premature tool failure in Ti machining process. This study exhibits an investigation on tool wear in micro milling of magnesium-based MMCs reinforced with 1.98 Vol.% of nano-sized titanium particles using 0.5-mm diameter two-flute tungsten carbide micro endmills. The tool wear was characterised both quantitatively and qualitatively by observing tool wear patterns and analysing the effect of cutting parameters on flank wear, reduction in tool diameter, cutting forces, surface roughness, and burr formation. A finite element model was established to understand matrix deformation and interaction between tool-particle and further explain the tool wear phenomena observed. Additionally, cutting performance using AlTiN coated and uncoated was also investigated. The results indicated that the main wear mechanisms were identified as flank wear and edge chipping due to abrasive wear and chip adhesion in uncoated micro endmills. These wear mechanisms were confirmed by chip formation process produced by finite element modelling (FEM). It was also observed that the largest tool wear occurred at the smallest feed per tooth (0.75 μm/tooth) and smallest wear occurred at the largest feed per tooth (3 μm/tooth). Also, the effect of BUE on tool wear and surface generation was studied.

In the last two decades, metal matrix composites (MMCs) have been utilised in various industrial applications owing to their high strength-to-weight ratio and superior wear resistance. However, these superior mechanical properties bring poor machinability. The purpose of this paper is to perform a simulation study on the cutting mechanism of magnesium based MMCs reinforced with SiC nanoparticles. A two-dimensional micromechanical finite element (FE) model is established using ABAQUS/Explicit to simulate the micro machining process with consideration of cutting edge radius. The simulated results present tool-particles interaction, chip formation process, cutting force, Von Mises stress and strain distribution within workpiece under the effect of uncut chip thickness. The model is validated by comparing with experimental data in terms of cutting force and chip morphology.