Digital Image Research Articles

Developing Robust AI Applications for Clinical Use: The Special Case of Pathology

Robustness is a key requirement for any method in medicine, especially when the method in question is being used as part of a diagnostic process. This is particularly true for artificial intelligence-based decision support systems, which, although being used as a supportive tool, will ultimately influence diagnostic assessments. In pathology, attaining clinical robustness in AI methods poses a particularly challenging task, primarily due to the extensive diversity of digital images, which humans can adapt to far more easily. This paper presents factors that contribute to this challenge, but also identifies and evaluates common solutions to counteract domain shift, which is known to deteriorate the performance of artificial intelligence models.

Investigation of corrosion resistance and interfacial bonding properties of silver/silicone zinc-rich coating/substrate system in marine environment

This paper aims to study the corrosion resistance and interfacial bonding properties of the silver zinc-rich coating/substrate system and silicone zinc-rich coating/substrate system. The actual working state of the system in marine environment was simulated by the coupled effects of corrosion and pre-stretching. Scanning electron microscope (SEM) with an energy dispersive X-ray spectrometer (EDS) was used in corrosion test to analyze the surface morphology and corrosion products. Meanwhile, the degradation degree of the corrosion resistance was investigated based on the electrochemical impedance spectroscopy (EIS) and water absorption rate. Three-point bending test and tensile test combined with digital image correlation (DIC) were performed to evaluate the interfacial bonding properties of these two systems based on the elastic modulus and critical cracking strain. Results show that the dense accumulation of aluminum and zinc powders makes the silver zinc-rich coating have higher corrosion resistance compared to the silicone zinc-rich coating. The combined effects of corrosion and pre-stretching on the fracture toughness of silicone zinc-rich coating are larger than that of the silver zinc-rich coating. The interfacial bonding properties of the coating/substrate systems gradually degrade as the pre-stretching load approaches the critical cracking load of the coating.

Temporary division of roles in group hunting for fish eggs by a coral reef fish

AbstractDivision of roles was observed during group hunting by the false cleanerfish, Aspidontus taeniatus (Blenniidae), when they raid the nests of the damselfish (Pomacentridae) and eat their guarded eggs. In this paper, we provide the first description of the collaborative group egg-eating behavior by the false cleanerfish. When raiding the nests of the three-spot dascyllus, Dascyllus trimaculatus, whose eggs are guarded by parents, the false cleanerfish divided roles as follows: “decoy” or “watcher” to draw attention and attract attacks from the parents, and “hider” or “intruder” to avoid detection by the parents and invade the nest. The potential differential costs associated with each role are unique among examples of group hunting strategies in fishes. However, once any individual in the group successfully invaded the nest, all individuals quickly achieved successful predation of the eggs and gained immediate shared benefit. We propose that the group egg-eating behavior of the false cleanerfish not only reinforces the evidence that fish can collaborate with other individuals but also suggests the hypothesis that collaborative hunting can evolve through mutualism even in fishes. Digital video images related to the article are available at http://www.momo-p.com/showdetail-e.php?movieid=momo240411at01a, and http://www.momo-p.com/showdetail-e.php?movieid=momo240411at02a.

An Analytical Model for the Initial Stiffness of Bearing Connections with Slotted Holes

In steel structures, cover plate joints are mainly used with normal round holes, but slotted holes, where the slot is oriented perpendicular to the direction of the bearing force, make it possible to release displacements in selected directions and thus better control the complex state of stresses in the connected plates. This study investigates the initial stiffness of a cover plate with a single slotted hole. The aim of this paper is to propose an analytical model that predicts the initial stiffness of slotted holes with respect to experimental results obtained on a panel of different geometries. This approach makes it possible to evaluate the effects of the end distances on the bending of the plate compared to the bearing under the bolt. A large panel of geometries is tested to quantify the experimental value of initial stiffness using Digital Image Correlation (DIC) technique compared to Linear Variable Differential Transformer (LVDT) sensors. To evaluate the initial stiffness, a method of processing experimental data is developed that allows a reproducible determination of stiffnesses. An analytical model is then proposed based on the improvement of existing models available in the literature. It is validated for different specimen geometries considering the contribution to the initial stiffness due to bending, shearing and support of the steel plate under the bolt. The comparisons show that the proposed analytical model can predict the initial stiffness of different geometries of tested specimens with slotted holes.

Fluidization characteristics of loaded vibrating fluidized bed and dewatering and deashing effect of moist fine-grained coal

ABSTRACT The influence of a thermal vibrating fluidized bed on the dewatering and deashing effect of moist fine-grained coal was investigated by digital image processing and signal analysis methods. Comparative analysis results of the pressure signals of a loaded vibrating fluidized bed indicated that high or low gas velocity and strong or weak vibration conditions easily caused the formation of numerous irregular bubbles and contributed to violent fluctuations in the bed. An interaction between gas velocity and vibration energy was found, with vibration energy being the dominant factor influencing the bubble behavior at low gas velocities. Moreover, at high gas velocities, the advantages arising from vibrations were weakened. Under the condition of f = 20 Hz, A = 2 mm, v = 18 cm⋅s−1, the contact efficiency between the gas – solid two phases was higher, and the pressure fluctuation of the bed became more stable. Under this condition, the ash content of clean coal and the moisture removal rate were 11.76% and 83.54%, respectively. Finally, comparison and analysis of the influence of different parameters on the calorific value of clean coal revealed that the influence of each parameter on heat generation of the selected product was found to be in the following order: f > A > V > T.

Effect of corrosion pit on fatigue damage and failure in powder bed fusion AlSi10Mg

AbstractThis paper is concerned with environmental corrosion during service of load‐bearing aluminum alloys fabricated by additive manufacturing (AM) with a focus on damage evolution characterization and identification of dominated failure mechanisms of pre‐corroded selective laser melted aluminum alloy. An experimental strategy for analysis of the damage and failure process is presented, combining 3D surface measurement, 3D digital image correlation, and scanning electron microscopy, to provide a multi‐source experimental characterization of corrosion morphology, strain field evolution, and fracture morphology. Statistics and machine learning methods were employed to process the measured multi‐source experimental data, showing that local average roughness has a strong influence on macro‐crack initiation position. The analysis focuses on four primary types of fatigue micro‐crack initiation, namely, internal defect, diffuse corrosion, corrosion micro‐pit, and corrosion jut; internal defect and diffuse corrosion are unique features for powder bed fusion (PBF) AlSi10Mg. The micro‐crack initiation mechanisms for all four types are (i) local stress concentrations due to adjacent material micro‐structural defects, (ii) interaction of underlying alloy microstructure and corrosion‐induced stress/strain, and (iii) local stress/strain concentration at corrosion micro‐pit and jut, respectively.

Combustion Characteristics of Vat-Photopolymerized Fuels with a Spiral Star Port for Hybrid Propulsion.

Despite hybrid rocket motors offering distinct advantages over solid or liquid rocket motors, their low regression rate and insufficient combustion efficiency remain significantly unimproved. This study focuses on the effects of the helix lead on the regression rate distribution and combustion efficiency of vat-polymerized fuel grains with a spiral star port for a hybrid rocket. Both experimental and numerical investigations were conducted to study the combustion characteristics and regression rate distribution of three-dimensional (3D) print grains. Spiral star grains with varying helix leads of 60, 90, and 120 mm were fabricated using light-curing 3D printing technology. A 3D simulation model was developed to obtain the temperature distribution, species mass distribution, and combustion efficiency. Furthermore, firing tests were performed on a two-dimensional radial hybrid combustion test stand to measure the regression rate. Digital image processing of computed tomography images was used to determine the regression rate. Simulation results indicated that the spiral star grain port helps to improve the combustion efficiency compared with those seen with round tube and straight star port grains. With an increase in the axial distance, the flame zone gradually shrinks, and the smaller the helix lead, the faster the shrinkage. At a mass flow rate of 1.50 g/s for oxygen, the regression rate of the spiral star grains is significantly higher than that of the straight star grain and the conventional round tubular grains, and the regression rate gradually increases with a decrease in the helix lead. This finding is expected to solve the problem of the low regression rate of solid fuels with spiral star pore-shaped grains prepared by the light-curing 3D printing method.

Analysis of the Meso‐Mechanical Deformation Behavior of Al Foams and Ni/Al Hybrid Foams Using Time‐Lapse Micro‐Computed Tomography Measurements and 3D Optical Flow

Metal foams are an interesting class of bio‐inspired materials for lightweight construction and energy absorption. Commonly, aluminum (Al) foams are used. However, the mechanical properties have been improved by coating pure Al foams with a nanocrystalline nickel (Ni) coating resulting in Ni/Al hybrid foams. Herein, the meso‐mechanical deformation mechanisms in Al foams and the changes in the mechanisms in Ni/Al hybrid foams are studied using time‐lapse micro‐computed tomography measurements in comparison between numerical modeling using voxel finite element models and evaluation of displacement fields using 3D optical flow. This gives never‐seen insights into the highly localized 3D deformation mechanisms within the entire volume of the foams and not only on the surfaces as given by conventional digital image correlation methods. Displacements calculated by the 3D optical flow algorithm demonstrate its possibility to reveal a significant concentration of plastic deformation, particularly evident when deformation occurs within a distinct, slightly inclined band in the central region. Common numerical simulations using standard plasticity models do not accurately capture this localized deformation, underscoring the need to integrate damage and softening models into the simulation.

Surface Bubbles Emergence as an Indicator for Optimal Concrete Compaction

Compaction quality significantly influences the strength and durability of concrete in structures. Under-compacting can retain entrapped air, reducing strength, while over-compacting can lead to segregation, creating local variances in strength distribution and modulus of elasticity in the concrete structure. This study examines the widely adopted concept that compaction is optimal when bubbles cease to emerge on the concrete surface. We recorded the surface activity of six comparable concrete specimens during the compaction process using a 4K video camera. Four specimens were compacted using a table vibrator and two with a poker vibrator. From the video frames, we isolated the bubbles for analysis, employing digital image processing techniques to distinguish newly risen bubbles per frame. It was found that the bubbles continuously rose to the surface in all specimens throughout the compaction process, suggesting a need for extended compaction, with some specimens showing a slow in the rate of the bubbles’ emergence. However, upon examining the segregation levels, it was discovered that all the specimens were segregated, some severely, despite the continued bubble emergence. These findings undermine the reliability of using bubble emergence as a principle to stop compaction and support the need for developing online measurement tools for evaluating compaction quality.

Study on Roof Deformation and Failure Law of Close Distance Coal Seams Mining Based on Digital Image Correlation

Study on Roof Deformation and Failure Law of Close Distance Coal Seams Mining Based on Digital Image Correlation

Isothermal and thermomechanical fatigue crack growth behavior of 316LN stainless steel under load‐ and strain‐controlled modes

AbstractCrack growth tests of 316LN stainless steel under load‐controlled and strain‐controlled thermo‐mechanical fatigue (TMF) loadings at the temperature range of 350°C ~ 550°C were conducted. Moreover, the isothermal fatigue (IF) crack growth tests at the maximum temperature of 550°C were performed for comparison. The differences in IF and TMF crack propagation behavior under different loading conditions were thoroughly discussed, and the underlying mechanisms were revealed by a comprehensive characterization of the deformation state of the crack tip region using digital image correlation (DIC) and electron backscatter diffraction (EBSD) methods. Furthermore, it was found that the EBSD characterization was simply based on a single plane made it difficult to evaluate the complex crack growth behavior. In addition, the limitations of the classical parameter of stress intensity factor under TMF loading were presented.

Bolt-Hole Elongation of Woven Carbon-Epoxy Composite Plates and Joints Using the Digital Image Correlation Technique

The elongation of the bolt hole is an important parameter for assessing the failure of bolted joints. However, direct experimental measurement using strain gauges and extensometers is difficult. This article shows that digital image correlation (DIC) can overcome the difficulties and provide important indications of the failure mechanisms of bolted joints. Hole elongation was measured using DIC in the following carbon/epoxy composite configurations: standard open-hole tensile (OHT) and filled-hole tensile (FHT), single-lap shear only-bolted (OB), and single-lap shear hybrid-bolted/bonded (HBB) joints. For each configuration, the hole-elongation changes were tracked for cross-ply (CP) and quasi-isotropic (QI) stacking sequences with two thicknesses. In the tensile load direction for OHT and FHT cases, CP showed a greater hole elongation than QI. However, the opposite trend was observed in the transverse direction. In OB joints, bypass loads contributed more to the hole elongation than bearing action. In HBB joints, it has been observed that the adhesive significantly reduces hole elongation, particularly for CP configurations. Moreover, it was found that in HBB joints, hole elongation was independent of laminate lay-up, while it was very determinative in OB joints.

Object Detection using Convolutional Neural Network

During the last years, a noticeable growth is observed in the field of computer vision research. In computer vision, object detection is a task of classifying and localizing the objects in order to detect the same. The widely used object detection applications are human– computer interaction, video surveillance, satellite imagery, transport system, and activity recognition. In the wider family of deep learning architectures, convolutional neural network (CNN) made up with set of neural network layers is used for visual imagery. Deep CNN architectures exhibit impressive results for detection of objects in digital image. This paper represents a comprehensive review of the recent development in object detection using convolutional neural networks

Space/ground-borne techniques and petrographic microscopic dissection for geologic mapping in Gabal Ras Abda area, Northeastern Desert, Egypt

BackgroundGabal Ras Abda area as a part of the Red Sea Mountain range, is characterized by inaccessible and rugged terrains. The exposed rock units are hardly followed in the field because of the rigid topography. Thus, the present work proposes and develops an integrated approach to map the exposed rock units and extract the geologic structures using satellite imagery data followed by both field and petrographic verification, saving time, efforts and cost.ResultsTo achieve the target, both the measured spectral signature curves with Landsat-8 and Sentinel-2A data were used to develop and create the most enhanced Band Ratios and Principal Components for lithological discrimination and mapping which were (((Band7 + Band 4)/(Band 7), (Band 2)/(Band 2 + Band 5) and (Band 5) in RGB) and ((Band 9 + Band 11 + Band 12)/(Band 1), (Band 4 − Band 2) and (Band 11/Band 6) + (Band 6) in RGB)) with Principal Component Bands ((PC1, PC2 and PC3 in RGB) and (PC3, PC2 and PC1 in RGB)), respectively. Also, georeferenced Google Earth Pro, panchromatic band of Landsat-8 and ALOS PALSAR Digital Elevation Model images were used to extract the structural lineaments. Geologic, petrographic and field structural studies were emphasized the remote sensing results, indicating that the main rock types cropped out in Ras Abda area from the oldest to the youngest are older granitoids (quartz-diorites, tonalites and granodiorites), Dokhan volcanics (andesites, rhyodacites, rhyolites and their related tuffs), younger gabbros, younger granites (monzogranites, syenogranites and alkali-feldspar granites), post-granite dykes and offshoots (acidic, microgranitic and basic types) and Phanerozoic sedimentary rocks. Also, the study emphasized that the E-W trend is the main structural trend controlling the investigated area followed by WNW-ESE and NE-SW directions.ConclusionsThe results of remote sensing achieved compliance with the geologic, petrographic and structural investigation through distinctly differentiating the different rocks and extracting the lineaments, indicating the accuracy of the remote sensing results and emphasizing their importance and effective role in producing a precise and highly accurate geologic map.Graphical abstract

Resin infiltration versus fluoride varnish for visual improvement of white spot lesions during multibracket treatment. A randomized-controlled clinical trial

AimsThis study aimed to evaluate the visual improvement of resin infiltration of white spot lesions (WSL) during orthodontic treatment with the multibracket appliance (MBA) compared to fluoride varnish.MethodsPatients aged 12–17 years with at least one WSL with an International Caries Detection and Assessment System (ICDAS) score of 1–2 during an active MBA treatment were included and randomized to receive either resin infiltration (Icon) or fluoride application (Flairesse). Standardized digital images were obtained before, one-day, one-week, one-month, three-months and six-months after treatment using a DSLR camera and a matching polarization filter. A grey reference card was used for color standardization. A Matlab routine was used to measure the color difference between adjacent healthy enamel and treated WSL. The independent-samples t-test was used for intergroup and paired-samples t-test for intragroup comparison.ResultsImages of 116 teeth from 36 patients were analyzed. The ΔE for the “Icon” treated WSL was smaller (T1ICON = 5.0 ± 1.4) than in the fluoride group (T1Fluoride = 8.4 ± 3.2). Caries infiltration significantly improved the aesthetic appearance of WSL (p < 0.001), which remained satisfactory at six months (T5ICON = 5.2 ± 1.6).ConclusionWSL infiltration management during orthodontic treatment was superior to topical fluoridation in not only arresting the enamel lesions but also significantly improving the aesthetic appearance of demineralized regions around the brackets.Clinical relevanceWSL treatment in orthodontic patients is usually initiated after debonding. Research has shown that the earlier WSL is treated, the better the aesthetic outcome. There is limited data on the efficacy of resin infiltration of WSL during orthodontic treatment.

Experimental analysis of the compressive behaviour of perforated brick masonry using digital image correlation

The paper presents an experimental program on the characterisation of the mechanical behaviour of perforated brick masonry (PBM) under uniaxial compression. The study is aimed to provide a better understanding on the compressive strength and elastic behaviour of a kind of masonry hardly dealt with in the literature. The mechanical response of the investigated masonry shows actually two complexities, linked respectively to the perforated nature of the bricks and the asymmetric cross-section of the units. The latter feature is particularly recurrent in exposed brickwork composed of perforated bricks, whenever one of the face shells has a larger thickness than the other one. For this purpose, an experimental campaign was carried out on units and PBM subjected to uniaxial compressive loading. The mechanical response of PBM is investigated to accurately determine the masonry Young’s modulus and the compressive strength. Measurements obtained with conventional sensors are compared with those derived from the Digital Image Correlation (DIC) technique. Lastly, the experimental findings in terms of compressive strength and Young’s modulus are cross-referenced with the estimations derived from empirical criteria available in Eurocode 6 and ACI 530.1–02.

Label-free, high-throughput holographic imaging to evaluate mammalian gametes and embryos.

Assisted reproduction is one of the significant tools to treat human infertility. Morphological assessment is the primary method to determine sperm and embryo viability during IVF cycles. It has the advantage of being a quick, convenient, and inexpensive means of assessment. However, visual observation is of limited predictive value for early embryo morphology. It has led many to search for other imaging tools to assess the reproductive potential of a given embryo. The limitations of visual assessment apply to both humans and animals. One recent innovation in ART imaging is interferometric phase microscopy (IPM), also known as holographic microscopy. IPM/QPI is the next likely progression of analytical microscopes for the assisted reproduction laboratory. The IPM system analyzes waves produced by the light as it passes through the specimen observed. The microscope collects the light waves produced and uses the algorithm to create a hologram of the specimen. Recently, IPM has been combined with quantitative phase imaging (QPI), which joins phase contrast microscopy with holographic microscopy. These microscopes collect light waves produced and use the algorithm to create a hologram of the specimen. Unlike other systems, IPM can provide a quantitative digital image, and it can make 2-D and 3-D images of the samples. This review summarizes some newer and more promising QPI microscopy systems for evaluating gametes and embryos. Studies clearly show that QPI is superior to bright field microscopy-based evaluation methods when evaluating sperm and oocytes prior to IVF and embryos prior to transfer. However, further assessment of these systems for efficacy, reproducibility, cost-effectiveness, and embryo/gamete safety must take place before they are widely adopted.

“Image-Spectral” fusion monitoring of small cotton samples nitrogen content based on improved deep forest

Accurate monitoring of nitrogen nutrition is crucial for improving cotton yield and quality, as well as the ecological environment. The mainstream method for monitoring nutrition is to establish traditional machine learning (ML) models using a single data source. However, this approach has limitations such as limited access to feature information, model-fitting problems, and limited generalization. Deep learning (DL), on the other hand, has shown promise in complex nonlinear modeling tasks due to its flexible structure. However, it has its limitations, such as the fact that agronomic sample collection and testing are usually labor- and material-intensive, resulting in sample sizes that are too small to meet its training conditions. Therefore, there is an urgent need for DL models that can effectively integrate features from multiple data sources and accurately monitor crop nitrogen content, especially in scenarios with small samples. In this study, we conducted indoor pot experiments using the cotton variety Xinluzao 53 and subjected it to six nitrogen treatments. The data sources for our analysis included hyperspectral and digital images of the cotton leaves. To enhance representation learning capabilities, we enriched the multi-class base learners within each layer of the deep forest (DF) model and introduced skip connections. These enhancements improved the quality of inversion for both hyperspectral and digital image datasets. We then developed image-spectral fusion models, which combined the DF structure with stacking ensemble learning. Our focus was on three levels of fusion: feature-level fusion, decision-level fusion, and secondary decision-level fusion. This approach aimed to further enhance the accuracy and stability of nitrogen content inversion. The DF model satisfied the training condition for small samples. Compared to traditional ML algorithms and the original DF algorithm, the improved DF model achieved an increase in validation set R2 of 13.4–28.5% and 10.9–14.9%, respectively. These findings highlight the enhanced accuracy and stability of the improved DF model. Additionally, compared to the optimal inversion model using two single data sources, the “Image-Spectral” three-level fusion models exhibited improvements in validation set R2 of 8.6–9.3%, 10.5–11.2%, and 11.8–12.5% for feature-level, decision-level, and secondary decision-level fusion, respectively. The improved DF and three-level fusion model collectively contributed to the increased accuracy of cotton nitrogen content inversion. Among these models, the secondary decision-level fusion model demonstrated the most marked improvement. This methodology provides valuable insights into monitoring crop phenotypic parameters in situations with limited sample sizes.

A methodology for phase characterization in pellet feed using digital microscopy and deep learning

This work proposes a new method for characterizing pellet feed, employing Deep Learning (DL) and Convolutional Neural Networks (CNNs). The main minerals in the composition of the studied pellet feed are hematite, magnetite, goethite, and quartz. Over time, several characterization methodologies have been developed that use Digital Microscopy and Image Analysis tools. The greatest difficulties in this characterization lie in differentiating the textures of hematite particles, the different shapes of their crystals, or discriminating between quartz and resin in reflected light optical microscopy images. This work proposes a mineral characterization methodology based on the Mask R-CNN algorithm. The goal is to perform instance segmentation, that is, to identify, classify, and segment objects in the images. Two DL models were combined: the BF Model performs instance segmentation for the compact, porous, martite, and goethite classes in images obtained in Bright Field mode, and the CPOL Model uses images acquired in Circularly Polarized Light to segment the monocrystalline, polycrystalline, and martite classes. An F1-score of around 80% was obtained for the BF Model and around 90% for the CPOL Model. The results were promising and can be improved as the training image database increases.

The toughness of interlocking metasurfaces

Interlocking metasurfaces (ILMs) are arrays of autogenous latching unit cells patterned across a surface. These create structural joints similar to bio‐inspired suture joints but patterned over a 2D surface rather than a 1D seam. This enables ILMs to be an alternative to conventional joining technologies such as bolts, welds, and adhesives. However, compared to conventional joining methods, relatively little is known of the engineering considerations for designing structural ILMs. In this study, the interfacial toughness of an archetypal ILM was examined for the first time. Under the conditions studied here, the ILM was substantially tougher than the material from which it was made; in this case exhibiting up to a 50% increase in interfacial crack initiation energy over the solid base material, a photocured 3D printed polymer. Through experimental tests using in‐situ digital image correlation along with complementary computational analyses, the mechanism of toughening in the ILM structure and the origins of toughness anisotropy were revealed. The increase in toughness is associated with cross‐cell interactions, i.e. load‐sharing across unit cells, which gives rise to a finite process zone length with different effective material properties. In this way, ILM toughening is analogous to crack blunting in ductile materials or fiber bridging in composites; yet here, the ILM was composed of a single‐phase base material and so the architected toughening is geometric in nature and hence amenable to future topological optimization.This article is protected by copyright. All rights reserved.