Surface Imaging Research Articles

Road adhesion coefficient Estimation: Physics-informed deep learning method with vehicle dynamics model

Road adhesion coefficient (RAC) is not only a significant basis for trajectory planning and decision-making of autonomous vehicles but also a critical parameter to determine the control potential of autonomous vehicles under complex operating conditions. Accurate, timely, and reliable estimation of RAC facilitates many function units of autonomous vehicles and improves driving safety. To address the issues of low accuracy and transparency of RAC estimation method based on road surface image recognition, a vehicle dynamics model-informed deep learning (VDMIDL) method for RAC estimation is proposed in this work. The physics information related to vehicle dynamics is integrated into a CNN-based deep learning model via the fusion of vehicle physical features and high-dimensional road surface image features to realize accurate estimation of RAC. The results of the model training show that the proposed VDMIDL model for RAC estimation possesses higher prediction accuracy compared with the conventional deep learning (CDL) model. The vehicle test results indicate that the proposed VDMIDL model can provide more prospective and precise RAC for trajectory planning, decision-making, and control of the vehicle in response to sudden variations in road conditions. The VDMIDL model is applicable to a wide range of road conditions. RAC estimated by the VDMIDL model is accurate and stable under poor road condition in particular. The proposed fusion estimation model incorporates more physics information into deep learning model, endowing more interpretability and transparency to the black-box model.

Evaluation of floatability characteristics of gastroretentive tablets using VIS imaging with artificial neural networks

Gastroretentive dosage forms are recommended for several active substances because it is often necessary for the drug to be released from the carrier system into the stomach over an extended period. Among gastroretentive dosage forms, floating tablets are a very popular pharmaceutical technology. In this study, it was investigated whether a rapid, nondestructive method can be used to characterize the floating properties of a tablet.To accomplish our objective, the same composition was compressed, and varied compression forces were applied to achieve the desired tablet. In addition to physical examinations, digital microscopic images of the tablets were captured and analyzed using image analysis techniques, allowing the investigation of the floatability of the dosage form. Image processing algorithms and artificial neural networks (ANNs) were utilized to classify the samples based on their strength and floatability. The input dataset consisted solely of the acquired images.It has been shown by our research that visible imaging coupled with pattern recognition neural networks is an efficient way to categorize these samples based on their floatability. Rapid and non-destructive digital imaging of tablet surfaces is facilitated by this method, offering insights into both crushing strength and floating properties.

Prior-shape-guided photometric stereo model for 3D damage measurement of worn surfaces

Three-dimensional (3D) damage measurement often encounters limited accuracy due to the non-Lambertian effects on worn surfaces. To address this issue, a prior-shape-guided surface reconstruction model is introduced with photometric stereo vision. Initially, an illumination calibration method is constructed to mitigate the uneven brightness on worn surface images. Subsequently, a prior-shape-guided network is developed that employs the feature aggregation strategy and prior shape uncertainty to guide worn surface reconstruction. Experimental results reveal that the proposed model can effectively recover the typical morphologies of worn surfaces, and the root mean square error is reduced from 7.39 to 1.59 when compared to existing methods. Most importantly, this model has been applied to detect aero-engine bearings and achieve 3D quantitative characterization for worn damages.

Myoglobin redox form prediction in fresh beef using computer vision systems and artificial intelligence

To development a computer vision system (CVS) to determine the myoglobin redox forms on beef surfaces, the reflectance spectra of reference samples of deoxymyoglobin (DMb), oxymyoglobin (OMb), and metmyoglobin (MMb) were recorded, and the surface images captured by a digital camera (CVScam) and a cell phone camera (CVScel) to train the algorithm. Meat color changes during blooming and display storage were also recorded. Higher k-fold accuracy was observed for CVScam (90.98 %) than for CVScel (86.53 %), with significantly correlation with colorimeter for OMb (r = 0.77 and 0.71), DMb (r = 0.84 and 0.71), and MMb (r = 0.87 and 0.88). The CVS MMb was lower and the OMb was higher than colorimeter, but the redox form behaviors were more consistent with the expected chemical changes on the surface. The constructed CVS showed satisfactory performance as a useful and accurate tool for predicting the myoglobin redox forms on the beef surface.

A visual measurement method for grinding surface roughness combining filter and branch convolution network

ABSTRACT The visual measurement method of grinding surface roughness is mainly predicted by designing image feature indicators. In contrast, the self-extraction method of grinding surface features based on deep learning has problems such as noise, low resolution and poor perception of details. This paper proposes a visual measurement method for grinding surface roughness to address these problems by combining filters and branching convolutional networks. The method adopts a two-branch convolutional neural network, inputs the images using filter denoising and the original grinding workpiece surface images simultaneously, self-extracts roughness correlation features and fuses them to achieve feature enhancement, effectively improves the recognition accuracy and generalisation ability of the model, and verifies the robustness of the model in the case of rusting on the workpiece surface. The experimental results show that the method can effectively solve the problem of the grinding process’s weak and difficult recognition of surface roughness feature information and provide a basis for automated visual online measurement of grinding surface roughness.

Enhancing concrete structure maintenance through automated crack detection: A computer vision approach

This paper presents the development of an Artificial Intelligence (AI) and Machine Learning (ML) model designed to detect cracks on concrete surfaces. The objective is to enhance the automation, precision, and performance of crack detection using the computer vision algorithm. Employing a ML approach and the YOLOv9 algorithm, this study developed a system to accurately identify concrete cracks from a diverse dataset. A total of 16,301 images of concrete surfaces, balanced between those with and without cracks, were utilized. The dataset was split into various sets with different ratios to ensure comprehensive model training. A transfer-learning methodology was employed to optimize the model's performance. The accuracy of the model was measured in each experiment to determine the optimal result. The most successful experiment resulted in a model with a mean Average Precision (mAP) of 94.6%, a Precision of 94.1%, and a Recall of 88.4%. These results demonstrate the effectiveness of AI and ML in concrete crack detection.

A comparative evaluation of human enamel remineralization ability of biomimetic nacre against casein phosphopeptide-amorphous calcium phosphate: An in vitro study.

This study aimed to assess and compare the efficacy of Nacre and casein phosphopeptide-amorphous calcium phosphate (CPP-ACP) on the remineralization of enamel using surface microhardness analysis, scanning electron microscopy (SEM), and energy dispersive X-ray (EDX) spectroscopy. Twenty human maxillary premolars extracted for orthodontic reasons were collected. Under cool water spray, the crowns were sectioned mesiodistally into buccal and palatal halves using a diamond disc. The samples were subsequently mounted in self-cure acrylic resin. The samples were then subjected to Vickers hardness testing and SEM-EDX for baseline. To simulate carious lesions, all of the samples were acid-etched with 37% phosphoric acid for 30 s in a specific area on the enamel samples and subjected to surface microhardness testing and SEM-EDX. The enamel samples were randomly assigned to Group 1: Nacre water-soluble matrix (WSM), Group 2: Nacre varnish, and Group 3: CPP-ACP for remineralization. After 21 days, remineralization assessment of the test samples was done using SMH analysis and SEM-EDX analysis. Data obtained were statistically analyzed using the one-way analysis of variance to reveal the significant differences between the groups. Tukey's test was used for post hoc comparisons. All three groups showed a significant increase in surface microhardness. All three groups showed a significant calcium and phosphorous ratio increase after remineralization. Among the three groups, the highest Ca:P ratio was seen in the Nacre WSM group (0.58) followed by the Nacre Varnish (0.57) and CPP-ACP group (0.57). SEM images of the Nacre surface revealed the presence of extensive interlocking. A layer of packed hydroxyapatite particles was formed on the surface of the nacre through surface reactions. All the groups in the present study showed some extent of remineralizing ability irrespective of the different materials and mechanisms of action. Nacre WSM showed a remarkable hardness spike close to natural enamel after demineralization.

Patterns of osteoarthritis in long finger joints: an observational cadaveric study with potential biomechanical considerations

PurposeTo explore the distribution and prevalence of osteoarthritis in metacarpophalangeal, proximal interphalangeal and distal interphalangeal joints in long fingers in a cadaveric study, and to discuss potential biomechanical influences on these patterns. MethodsThis cadaveric study evaluated 144 metacarpophalangeal, proximal interphalangeal and distal interphalangeal joints from 12 embalmed cadaver hands. A dorsal dissection approach was used to expose the joints, which were then marked with color-coded pegs for consistent orientation during imaging. High-resolution digital images of the distal articular surfaces were captured for analysis. The images were analyzed using custom software to quantify osteoarthritic areas, distinguishing between radial and ulnar aspects. Percentage affected joint surface was calculated using pixel-based measurements. Statistical analysis was used the Student t-test and ANOVA, with the significance threshold set at p < 0.05 and 95% confidence intervals ResultsThe ulnar side of the proximal interphalangeal joint in digits 2 and 3 showed higher prevalence of osteoarthritis (59.31% ± 15.48%) than the radial side (40.68% ± 15.48%), p = 0.007; in contrast, for digits 4 and 5, prevalence was greater on the radial (54.3% ± 10.99%) than the ulnar side (45.7% ± 10.99%), p = 0.007. No significant differences were noted in osteoarthritis distribution between the radial and ulnar aspects of the metacarpophalangeal and distal interphalangeal joints. ConclusionsThis study identified distinct patterns of osteoarthritis distribution in long-finger joints, with greater prevalence in the proximal interphalangeal joints. Although there were differences between stable (digits 2 and 3) and mobile (digits 4 and 5) fingers, further research is necessary to conclusively determine the role of biomechanical forces in the development of osteoarthritis. These findings lay the groundwork for future studies of the pathophysiology of osteoarthritis in the hand, and could guide the development of preventive and therapeutic interventions.

Microbes Breaking Down Plastic: Insights for Sustainable Waste Management

This research investigates the microbial degradation of low-density polyethylene (LDPE) and high-density polyethylene (HDPE) plastics by Bacillus sp., Proteus sp., Pseudomonas sp., and Salmonella sp. The study employs a systematic approach, isolating microorganisms from plastic-contaminated soil and subjecting them to a series of biochemical tests for identification. The research evaluates the weight loss of LDPE and HDPE over two months, revealing varying degrees of degradation among the bacterial strains. Results suggest a potential greater susceptibility of HDPE to microbial degradation. Scanning Electron Microscopy (SEM) analysis provides high-resolution images of the plastic surface, indicating structural changes and biofilm formation during degradation. The findings highlight the unique enzymatic capabilities of each strain and underscore the significance of SEM in elucidating microbial interactions with plastics. The study prompts discussions on optimization, synergistic effects, and the identification of key enzymes in plastic degradation, emphasizing the importance of microbial strategies for waste management. Overall, this research contributes valuable insights into the potential of bacterial strains for addressing plastic pollution challenges.

Comparing 3D imaging devices for the measurement of cutaneous neurofibromas in patients with Neurofibromatosis Type 1.

Cutaneous neurofibromas (cNFs) are a major cause of disfigurement in patients with Neurofibromatosis Type 1 (NF1). However, clinical trials investigating cNF treatments lack standardised outcome measures to objectively evaluate changes in cNF size and appearance. 3D imaging has been proposed as an objective standardised outcome measure however various systems exist with different features that affect useability in clinical settings. The aim of this study was to compare the accuracy, precision, feasibility, reliability and accessibility of three imaging systems. We compared the Vectra-H1, LifeViz-Micro and Cherry-Imaging systems. A total of 58 cNFs from 13 participants with NF1 were selected for imaging and analysis. The primary endpoint was accuracy as measured by comparison of measurements between imaging systems. Secondary endpoints included reliability between two operators, precision as measured with the average coefficient of variation, feasibility as determined by time to capture and analyse an image and accessibility as determined by cost. There was no significant difference in accuracy between the three devices for length or surface area measurements (p>0.05), and reliability and precision were similar. Volume measurements demonstrated the most variability compared to other measurements; LifeViz-Micro demonstrated the least measurement variability for surface area and image capture and analysis were fastest with LifeViz-Micro. LifeViz-Micro was better for imaging smaller number of cNFs (1-3), Vectra-H1 better for larger areas and Cherry for uneven surfaces. All systems demonstrated excellent reliability but possess distinct advantages and limitations. Surface area is the most consistent and reliable parameter for measuring cNF size in clinical trials.

Postprocessing steps for endocardial surface imaging.

Postprocessing steps for endocardial surface imaging.

Improved 3D characterization of in-situ soil desiccation cracking by multi-source data integration

Desiccation cracking is a common and natural phenomenon under a drought climate. The geometric and morphologic characteristics of the crack pattern are critical to understanding the response of soil mechanical and hydraulic properties to drought climate. It is always a big challenge to obtain the refined geometric structure of the in-situ soil desiccation crack network. This study proposes an integrated method named ERT+ for determining the three-dimensional geometry of in-situ soil desiccation crack networks by integrating multi-source data from electrical resistivity tomography (ERT), surface image analysis, and depth investigation. The proposed method was applied to three in-situ expansive soil sites with different crack geometries and soil properties. The results showed that the joint application of ERT with other investigations reduced the ambiguity of interpreting each technique independently. The crack network model characterized the three-dimensional geometric structure of the desiccation crack accurately and quantitatively, verifying the effectiveness and feasibility of the ERT+ method. Field and laboratory experiments showed that heterogeneity in soil properties resulted in different cracking morphologies (width, depth, and width-depth ratio). The crack geometric data suggested that the width-depth ratio of the crack network was related to the cracking modes and the soil properties, indicating the non-homology of different crack networks. In addition, the correction gradients calculated from the ERT+ method also varied with the cracking modes and soil properties, further suggesting the reliability and prospect of the proposed method.

Focal point technology: Controlling treatment depth and pattern of skin injury by a novel highly focused laser

Selective photothermolysis has limitations in efficacy and safety for dermal targets. We describe a novel concept using scanned focused laser microbeams for precise control of dermal depth and pattern of injury, using a 1550nm laser that generates an array of conical thermal zones while minimizing injury to the epidermis. To characterize the conical thermal zones invivo and determine safe starting parameters to transition to a second phase to explore potential clinical indications. A focused toroidal (ring) laser beam was delivered through a cold sapphire window, sparing epidermal injury in a central zone. Pulse energy, lesion depth, density, and energy delivery were titrated in exvivo human skin and subsequently on the backs of 21 human subjects. Histology showed microscale patterns of thermal injury, which varied predictably with laser parameters. Time-course healing through histology and skin surface imaging demonstrated the ability of the device to deliver high energies without sequelae. Clinical data are currently being collected to further explore the safety and efficacy of the device. The 1550nm laser with focal point technology enables precise control of lesion depth while simultaneously sparing a large portion of the epidermis, lowering the risk of adverse effects.

Prediction model for intraoperative implant volume using the 3D surface imaging system (VECTRA XT 3D) in direct-to-implant breast reconstructions.

In direct-to-implant breast reconstruction, accurate preoperative breast volume estimation is crucial for surgeons to select the appropriate implant volume, considering the cosmetic outcomes during surgery. We proposed the prediction model for intraoperative implant volume based on the preoperative estimated volume of the contralateral breast obtained through a three-dimensional surface imaging system (3DSI) as surgeons usually choose the implant volume on the breast which should be reconstructed considering symmetricity with the contralateral breast. We enrolled 97 patients from our single institution who underwent unilateral mastectomy with immediate breast reconstruction using smooth silicone implants between October 2021 and January 2023. Preoperatively, plastic surgeons measured the volume of the contralateral breast using the VECTRA XT 3D imaging system. Data on implant volume and the types of acellular dermal matrix used during surgery, determined by a single surgeon to ensure symmetry, were also collected. Linear regression analysis was utilized to construct the predictive model. In the multiple linear regression analysis with preoperative contralateral breast volume, age, and body mass index as variables, the coefficient of determination of the model expressed as R squared (R2) was 0.554, and except for age, the other variables were statistically significant. When replaced by mastectomy volume instead of age, R2 increased to 0.723 and all variables were significant. 3DSI can be helpful for preoperative surgical planning and postoperative outcome simulation. With our multiple linear regression model, we can predict the intraoperative implant volume using preoperative contralateral breast volume measured by the 3D scans.

Four-dimensional endocardial surface imaging with dynamic virtual reality rendering: a technical note.

Open heart surgery requires a proper understanding of the endocardial surface of the heart and vascular structures. While modern four-dimensional (4D) imaging enables excellent dynamic visualization of the blood pool, endocardial surface anatomy has not routinely been assessed. 4D image data were post-processed using commercially available virtual reality (VR) software. Using thresholding, the blood pool was segmented dynamically across the imaging volume. The segmented blood pool was further edited for correction of errors due to artifacts or inhomogeneous signal intensity. Then, a surface shell of an even thickness was added to the edited blood pool. When the cardiac valve leaflets and chordae were visualized, they were segmented separately using a different range of signal intensity for thresholding. Using an interactive cutting plane, the endocardial surface anatomy was reviewed from multiple perspectives by interactively applying a cutting plane, rotating and moving the model. In conclusions, dynamic three-dimensional (3D) endocardial surface imaging is feasible and provides realistic simulated views of the intraoperative scenes at open heart surgery. As VR is based on the use of all fingers of both hands, the efficiency and speed of postprocessing are markedly enhanced. Although it is limited, visualization of the cardiac valve leaflets and chordae is also possible.

Concrete Crack Detection and Segregation: A Feature Fusion, Crack Isolation, and Explainable AI-Based Approach.

Scientific knowledge of image-based crack detection methods is limited in understanding their performance across diverse crack sizes, types, and environmental conditions. Builders and engineers often face difficulties with image resolution, detecting fine cracks, and differentiating between structural and non-structural issues. Enhanced algorithms and analysis techniques are needed for more accurate assessments. Hence, this research aims to generate an intelligent scheme that can recognize the presence of cracks and visualize the percentage of cracks from an image along with an explanation. The proposed method fuses features from concrete surface images through a ResNet-50 convolutional neural network (CNN) and curvelet transform handcrafted (HC) method, optimized by linear discriminant analysis (LDA), and the eXtreme gradient boosting (XGB) classifier then uses these features to recognize cracks. This study evaluates several CNN models, including VGG-16, VGG-19, Inception-V3, and ResNet-50, and various HC techniques, such as wavelet transform, counterlet transform, and curvelet transform for feature extraction. Principal component analysis (PCA) and LDA are assessed for feature optimization. For classification, XGB, random forest (RF), adaptive boosting (AdaBoost), and category boosting (CatBoost) are tested. To isolate and quantify the crack region, this research combines image thresholding, morphological operations, and contour detection with the convex hulls method and forms a novel algorithm. Two explainable AI (XAI) tools, local interpretable model-agnostic explanations (LIMEs) and gradient-weighted class activation mapping++ (Grad-CAM++) are integrated with the proposed method to enhance result clarity. This research introduces a novel feature fusion approach that enhances crack detection accuracy and interpretability. The method demonstrates superior performance by achieving 99.93% and 99.69% accuracy on two existing datasets, outperforming state-of-the-art methods. Additionally, the development of an algorithm for isolating and quantifying crack regions represents a significant advancement in image processing for structural analysis. The proposed approach provides a robust and reliable tool for real-time crack detection and assessment in concrete structures, facilitating timely maintenance and improving structural safety. By offering detailed explanations of the model's decisions, the research addresses the critical need for transparency in AI applications, thus increasing trust and adoption in engineering practice.

2856: Could surface imaging in breast cancer radiotherapy detect early lymphoedema?: The Calibrate Study.

2856: Could surface imaging in breast cancer radiotherapy detect early lymphoedema?: The Calibrate Study.

Optical Real-Time Castability Evaluation for High-Throughput Glass Melting

A novel optical real-time method for evaluating the castability of glass forming melts for laboratory furnaces is presented. The method is based on the analysis of top view images of the melt surface inside the crucible during melting after being subjected to a small mechanical impulse. In this way, the melt surface is excited to oscillate. The difference in contrast between two images taken in quick succession scales with the viscosity, with a larger diffe­rence occurring at lower viscosities. The method is designed as an instrument for the in-line evaluation of the castability for a high-throughput glass melting system as part of the joint project “GlasDigital” in the framework of the German Platform Material Digital initiative but is applicable to other laboratory furnaces as well.

WD-1D-VGG19-FEA: An Efficient Wood Defect Elastic Modulus Predictive Model.

As a mature non-destructive testing technology, near-infrared (NIR) spectroscopy can effectively identify and distinguish the structural characteristics of wood. The Wood Defect One-Dimensional Visual Geometry Group 19-Finite Element Analysis (WD-1D-VGG19-FEA) algorithm is used in this study. 1D-VGG19 classifies the near-infrared spectroscopy data to determine the knot area, fiber deviation area, transition area, and net wood area of the solid wood board surface and generates a two-dimensional image of the board surface through inversion. Then, the nonlinear three-dimensional model of wood with defects was established by using the inverse image, and the finite element analysis was carried out to predict the elastic modulus of wood. In the experiment, 270 points were selected from each of the four regions of the wood, totaling 1080 sets of near-infrared data, and the 1D-VGG19 model was used for classification. The results showed that the identification accuracy of the knot area was 95.1%, the fiber deviation area was 92.7%, the transition area was 90.2%, the net wood area was 100%, and the average accuracy was 94.5%. The error range of the elastic modulus prediction of the three-dimensional model established by the VGG19 classification model in the finite element analysis is between 2% and 10%, the root mean square error (RMSE) is about 598. 2, and the coefficient of determination (R2) is 0. 91. This study shows that the combination of the VGG19 algorithm and finite element analysis can accurately describe the nonlinear defect morphology of wood, thus establishing a more accurate prediction model of wood mechanical properties to maximize the use of wood mechanical properties.

High-Quality SiC Crystal Growth by Temperature Gradient Control at Initial Growth Stage

A modified process condition has been proposed for the growth of high-quality 6-inch 4H-SiC single crystal. Temperature gradient (dT[°C] = Tbottom-Tupper) was controlled by changing coil position in order to investigate the effect of the temperature gradient on the SiC crystal quality. SiC ingot surface and etch pit density (EPD) of etched SiC wafer were investigated according to different dT conditions at the initial stage of SiC crystal growth. The surface of SiC crystal ingots grown with different dT for 10h were observed by OM and etched SiC wafers were prepared from SiC crystal ingots after main growth step for 100h. Different dT conditions in the initial growth stage resulted in dramatically different surface images and the crystal quality evaluated by EPD.