Traditional Visual Inspection Method Research Articles

Modal Identification Techniques for Concrete Dams: A Comprehensive Review and Application

Throughout history, the implementation of structural health monitoring systems has played a crucial role in evaluating the responses of dams to environmental and human-induced threats. By continuously monitoring structural integrity and analyzing dynamic characteristics, these systems offer a robust alternative to traditional visual inspection methods, ensuring the long-term safety of dams. This paper delves into the intricate process of operational modal analysis applied to dams, encompassing data collection, preprocessing, and the utilization of diverse modal identification techniques across both time and frequency domains. Moreover, it explores innovative approaches aimed at overcoming challenges encountered in previous methodologies. Also, the evolution of automated modal identification techniques and their application in dams are investigated. It explores the advancements in this field and their implications for enhancing the efficiency and accuracy of modal analysis processes. Furthermore, this paper evaluates the effectiveness of damage detection methods in dams based on operational modal identification.

An image Analysis Algorithm for Detecting Modified Content on the Internet Against Cybercrime. A Technique for Estimating the Probability of Modification

Aim: The aim of this study is to present a cutting-edge image analysis algorithm designed to estimate the probability of modifications in digital images, a critical component in cybersecurity for detecting altered content on the Internet. This technique enhances the accuracy of change detection by distinguishing real inconsistencies from accidental changes. As the problem of sophisticated image processing software and fake images continues to rise on the Internet, the issue of image authenticity becomes increasingly crucial in many areas of science and society. Traditional visual inspection methods are inadequate, as research shows human perception is limited in recognizing the authenticity of images. This study aims to improve the accuracy of detecting modifications by distinguishing authentic inconsistencies from random anomalies Project and methods: This project involves developing an algorithm that leverages noise analysis and a statistical validation step to detect image modifications. The algorithm employs the False Positive Rate Index (FPRI) to manage false positives, providing a reliable confidence level for each detection. The method includes visual exploration to interpret detections and compares the algorithm’s performance with other state-of-the-art techniques in scenarios such as retouching, coloring, and merging. The algorithm analyzes noise in images, considering changes in its statistical properties due to image processing like noise reduction, demosaicing, chromatic aberration correction, color matching, gamma correction, and image compression. It compares global noise curves with local ones to identify potential changes, with results visualized using heat maps. Results: The algorithm demonstrates promising performance across various test scenarios, successfully identifying true modifications among random events. However, it shows limitations in detecting certain types of forgeries, such as internal copy transfer and merging in high-noise areas. Tests conducted on various data sets, including scenarios of retouching, coloring, and merging, confirm the algorithm’s effectiveness in detecting modifications. The results are compared with other noise analysis methods such as Splicebuster, Noiseprint, and Mahdian, showing superior performance in most cases. Conclusions: The proposed method represents a significant advance in image forgery detection within the field of cybersecurity, offering rigorous statistical validation and a measurable level of detection confidence. Despite some challenges in detecting specific types of manipulation, the algorithm is a valuable tool in digital image analysis and forensic research, enhancing the reliability of image alteration detection. As the use of digital technologies increases, so does the importance of image interpretation in areas like fire investigations and other incidents involving records from both classical and thermographic imaging systems. This method is particularly valuable in forensic investigations, where objective evidence of image authenticity is crucial. Future challenges include the use of neural networks to replicate the characteristic noise of original images to hide alterations, underscoring the need for continued development of advanced forgery detection techniques. Keywords: wykrywanie modyfikacji obrazu, cyberbezpieczeństwo, analiza obrazu, wykrywanie fałszywych obrazów

ARTIFICIAL INTELLIGENCE BASED SUSTAINABLE COTTON CROP PRODUCTION USING REMOTE SENSING DATASET

The decline in cotton production in Pakistan has directly impacted the country's foreign exchange earnings, plummeting from twelve million bales in 1991, the number decreased to four million bales in 2020. Considering a substantial portion of Pakistan's exports are linked to the textiles area, with 40% of jobs tied to production, this decline is concerning. Factors such as climate change and changes in biotic and abiotic stresses have contributed to substantial losses in cotton production. In Pakistan, where 90% of farmers face resource constraints, there is a pressing need for more accurate mechanisms to understand the challenges in crop production. Timely acknowledgment and adoption of more flexible approaches to irrigation, fertilizer application, and pest and disease control are crucial. Excessive use of pesticides has exacerbated the situation, leading to the development of resistance among pests and diseases. Traditional visual inspection methods alone are insufficient for predicting crop health effectively, resulting in deprived crop yields and financial losses for agrarians. Enhancing crop health condition requires leveraging information gathered from remote sensing and terrestrial sensors, along with truth field surveys. By processing this data through tools like the Decision Support System for Agrotechnology Transfer (DSSAT) and employing machine learning techniques such as artificial neural networks, more accurate local and field-level advice can be provided to farmers. This approach aims to optimize crop management practices, reduce losses, and ultimately improve farmers' incomes.

A bearing surface defect detection method based on multi-attention mechanism Yolov8

Surface defects in bearings not only affect the appearance but also impact the service life and performance. Therefore, it is imperative for bearing manufacturers to conduct quality inspections before bearings leave the factory. However, traditional visual inspection methods exhibit shortcomings such as high omission rates, insufficient feature fusion and oversized models when dealing with multiple target defects in bearings. To address these challenges, this paper proposes a surface defect detection method for bearings based on an improved Yolov8 algorithm (G-Yolov8). Firstly, a C3Ghost convolutional module based on the Ghost module is constructed in YOLOv8 to simplify model computational costs. Secondly, a global attention mechanism module is designed at the end of the backbone network to increase sensitivity to implicit small target area features and optimize feature extraction efficiency. Subsequently, a deep deformable convolution feature pyramid network is constructed by introducing the deformable convolutional networks version 2 (DCNv2) and the lightweight content-aware reassembly of features upsampling operator to reduce sampling information loss and improve the fusion of multi-scale target defects. Finally, different attention mechanisms are embedded in the detection network to construct a multi-attention detection head to replace the decoupled head, refining classification and localization tasks, reducing feature confusion, and improving the model’s detection accuracy. Experimental results demonstrate that the improved algorithm achieves a 3.5% increase in mean average precision on a self-made small-scale train bearing surface defect dataset, with a 17.3% reduction in model size. This improvement not only enhances accuracy but also addresses the requirement for lightweight deployment in subsequent stages.

Deep learning techniques to detect rail indications from ultrasonic data for automated rail monitoring and maintenance

The increasing number of passengers and services using railways and the corresponding increase in rail use has caused the acceleration of rail wear and surface defects which makes rail defect identification an important issue for rail maintenance and monitoring to ensure safe and efficient operation. Traditional visual inspection methods for identifying rail defects are time-consuming, less accurate, and associated with human errors. Deep learning has been used to improve railway maintenance and monitoring tasks. This study aims to develop a structured model for detecting railway artifacts and defects by comparing different deep-learning models using ultrasonic image data. This research showed whether it is practical to identify rail indications using image classification and object detection techniques from ultrasonic data and which model performs better among the above-mentioned methods. The methodology includes data processing, labeling, and using different conventional neural networks to develop the model for both image classification and object detection. The results of CNNs for image classification, and YOLOv5 for object detection show 98%, and 99% accuracy respectively. These models can identify rail artifacts efficiently and accurately in real-life scenarios, which can improve automated railway infrastructure monitoring and maintenance.

Detecting vineyard plants stress in situ using deep learning

Diseases and nutritional deficiencies have the potential to seriously impact the production yield and proper development of perennial species such as grapevine. The distinction between changes resulting from normal growth stages and plant alterations caused by biotic and abiotic stress is often drawn through visual inspection, where the observer’s subjectivity could introduce human errors, despite the presence of experience and technical knowledge. This document presents an assessment of CNNs for detecting plant stress in grapevine RGB images captured in situ, under conditions that could include variations in light, shadows, insects, or the presence of scrubs. We evaluated five architectures for their ability to discriminate plants with stress symptoms in images captured through the annual grapevine cycle in field conditions. The best model exhibited a 97.2% accuracy, 0.996 ROC AUC, and 0.958 AP using the EfficientNetB3 architecture. Our methodology aims to support winegrowers in their decision-making by enhancing the information they collect through traditional visual inspection methods.

Applications of Terrestrial Laser Scanner in Detecting Pavement Surface Defects

An entire roadway system represents a crucial element in the sustainable urban transportation planning process. Pavement surfaces are at continual risk of accumulating serious deteriorations and defects throughout their service life due to traffic loading and environmental impact. Since roadway networks are growing rapidly, relying on visual pavement inspection is not always feasible. Therefore, this paper proposes an effective assessment method for evaluating flexible pavement surface distresses using a terrestrial laser scanner (TLS) and calculating the pavement condition index (PCI). The proposed terrestrial laser scanner method results in road condition assessments becoming faster, safer, and more systematic. It also aims to determine the geometric characteristics of the investigated roads. A major road in Egypt was selected to test the proposed technique and compare it with the traditional visual inspection method. The evaluation was carried out to assess different types of pavement distress, such as cracking, rutting, potholes, and raveling distresses. Every pavement distress was defined in terms of surface area, the width of the crack, and intensity, and the data from TLS were then processed by MAGNET COLLAGE software. A MATLAB program was developed to match the TLS observational data to plane equations. PAVER software was also used to determine the PCI values for each TLS position. The revealed distresses for the investigated road using TLS observations reveal a significant improvement in determining flexible pavement distresses and geometric characteristics.

Potato Blight Classification Android Application using Deep learning

Farmers who grow potatoes suffer from significant financial losses each year due several diseases that affect potato plants. The most common diseases are Early and Late Blight , caused by fungus and specific microorganisms, respectively. Early detection and appropriate treatment can save a lot of waste and prevent economic losses. However, traditional visual inspection methods are time-consuming and prone to errors. To address this challenge, we propose a Convolutional Neural Network ( CNN ) approach for plant disease diagnosis. CNNs are a type of deep learning algorithm widely used for image classification tasks. They can automatically learn features from input image data, making them well-suited for plant disease diagnosis. Our customized CNN has fewer trainable parameters, reducing computation time and minimizing information loss. We used several convolutional and pooling layers, followed by fully connected layers with the ReLU activation function. We also applied dropout regularization to prevent overfitting. In conclusion, accurate and efficient plant disease diagnosis is essential for preventing economic losses. Our customized CNN for plant disease diagnosis has the potential to be an effective tool for farmers. It can help them diagnose plant diseases quickly and accurately, leading to timely treatment and reduced loss.

Wind Turbine Actual Defects Detection Based on Visible and Infrared Image Fusion

It is of paramount importance to conduct accurate inspections of wind turbine blades to identify and address any defects. However, traditional visual inspection methods are often lacking in intelligence, have high rates of false detections, and are relatively inefficient. Conventional image-based detection methods are also not capable of distinguishing between actual defects, such as coatings falling off, and false positives, such as dust, urine, or feces, in the abnormal areas of images. In this study, we propose a novel approach for distinguishing actual defects in wind turbine blades through the implementation of an efficient data processing method and a feature fusion module for identifying potential actual defects. By utilizing multiple forms of feature fusion, we are able to effectively eliminate incorrectly characterized defects. Our proposed “Regression Crop” data processing method enables the automatic selection and cropping of relevant areas of wind turbine blades in the original images, while our adaptive feature fusion module for RGB and infrared images improves classification and localization accuracy for actual defects. Experimental results indicate that our approach achieves optimal results, with the “Regression Crop” data processing method resulting in a significant improvement in detection accuracy and the adaptive feature fusion module increasing the precision of actual defects to 99%. Furthermore, the adaptive feature fusion module is easily integrated into advanced object detectors such as YOLOv7 to improve their accuracy.

Research on the detection of early caries based on hyperspectral imaging

Objective: We applied hyperspectral imaging (HSI) system to distinguish early caries from sound and pigmented areas. It will provide a theoretical basis and technical support, for research and development of an instrument that could be used for screening and detection of early dental caries. Methods: Eighteen extracted human teeth (molars and premolars), with varying degrees of natural pathology and no degree of decay involving dentin were obtained. HSI system with a wavelength range from 400 to 1000[Formula: see text]nm was used to obtain images of all 18 teeth containing sound, carious and pigmented areas. We compared the spectra of the wavebands at both 500[Formula: see text]nm and 780[Formula: see text]nm from the different tooth states, and the reflectance difference between sound versus carious lesions and sound versus pigmented areas, respectively. Results: There was a slight difference in reflectance between carious areas and pigmented areas at 500[Formula: see text]nm. A substantial difference was additionally noted in reflectance between carious areas and pigmented areas at 780[Formula: see text]nm. Conclusion: The results have shown that the interference of tooth surface pigment can be eliminated in the near-infrared (NIR) waveband, and the caries can be effectively identified from the pigmented areas. Thus, it could be used to detect carious areas of teeth in place of the traditional visual inspection method or white light endoscopy. Clinical significance: The NIR diffused light signal enables the identification of early caries from pigment and other interference, providing a reasonable detection tool for early detection and early treatment of teeth diseases.

Evaluation of Gonad Reproductive Condition Using Non-invasive Ultrasonography in Red Abalone (Haliotis rufescens)

The reproductive status in abalone (Haliotis spp.) can be difficult to discern using traditional visual inspection methods and may therefore rely on lethal histological analyses for a definitive determination. With endangered species of abalone, lethal histological examinations are strictly prohibited due to their imperiled status. This research investigated the effectiveness of using non-invasive ultrasound technology as an alternate means of identifying the reproductive status of abalone rather than through invasive biopsies or lethal histological methods. Twelve abalone were randomly selected from cultured red abalone broodstock cohort (n = 200) produced at The Cultured Abalone Farm (Goleta, CA) to initially test these methods. Following initial gonad ultrasound determination, a subset of another cohort of 5 year-old (n = 62) red abalone produced at the UC Davis Bodega Marine Laboratory were monitored using ultrasonography to detect seasonal changes in gonad size for 7 weeks. Following the 7-week assessment period, the full cohort of 5-year old abalone (n = 122) were then spawned and gonads were also evaluated using ultrasound technology before and immediately after spawning. Ultrasound imaging technology accurately differentiated between digestive and reproductive tissues, allowing for the characterization of an ultrasound gonad index score for rapid assessment. Mean gonad thickness determined via ultrasound imaging was categorized on a scale of 1–5, with an index of 1 being the lowest (gonad tissue thin or absent) and an index of 5 being the highest (gonad margin has increased and is compressing the digestive gland) for red abalone. Ultrasound imaging technology was successful in rendering a gonad index score that can be used to track gonad maturation over time in cultured or captive species. Tracking ultrasound gonad scores over time is a useful tool for improving abalone culture production, and increasing animal welfare by reducing handling stress associated with gonad assessments. Individual abalone that are empirically determined ready to be spawned can then be selected for broodstock. Non-invasive ultrasound technology has the potential for broad applications within abalone aquaculture to enhance both food and conservation aquaculture breeding programs.

Condition assessment of rigid pavement using terrestrial laser scanner observations

ABSTRACT The main task of this study is to introduce an applied methodology using terrestrial laser scanner for facilitating the process of detecting, measuring defects and evaluating the condition assessment of rigid pavement instead of traditional visual inspection methods which cause waste of time, high cost and work to disrupt traffic. Terrestrial laser scanner (TLS) was used to create a three-dimensional model for rigid pavement of Gas station in Kafr El-sheikh city in Egypt to evaluate the condition assessment and defect detecting. The evaluation was carried out on rigid pavement slabs with different distress types such as transverse cracking, joint spalling, polished aggregate, blow up, punch out failures, and surface defects. Pavement distresses were identified in terms of defected area, crack width, and intensity. Geometric properties of each rigid pavements slabs are also determined from terrestrial laser scanner observations such as dimensions, settlement of each rigid slab using plane equation technique, rotation and tilting angles for each slab. MAGNET COLLAGE software program was applied to process TLS data and surface cracks measuring. Designed MATLAB software program are also created to fit TLS observations to plane equations. Based on the results, it is deduced that the distortion of rigid slabs ranged from 10 to 46 mm as maximum values and from 1 to 5 mm as minimum values, the angles of the rotation of studied rigid pavement slabs ranged from 0° to 2.3° but the angles of tilting of studied rigid pavement slabs are ranged from −1.6° to 1.9°. The revealed defects in the studied Gas station using terrestrial laser scanner observations provide a significant improvement for detecting rigid pavements characteristics and distresses.

A convolutional neural network approach for counting and geolocating citrus-trees in UAV multispectral imagery

Visual inspection has been a common practice to determine the number of plants in orchards, which is a labor-intensive and time-consuming task. Deep learning algorithms have demonstrated great potential for counting plants on unmanned aerial vehicle (UAV)-borne sensor imagery. This paper presents a convolutional neural network (CNN) approach to address the challenge of estimating the number of citrus trees in highly dense orchards from UAV multispectral images. The method estimates a dense map with the confidence that a plant occurs in each pixel. A flight was conducted over an orchard of Valencia-orange trees planted in linear fashion, using a multispectral camera with four bands in green, red, red-edge and near-infrared. The approach was assessed considering the individual bands and their combinations. A total of 37,353 trees were adopted in point feature to evaluate the method. A variation of σ (0.5; 1.0 and 1.5) was used to generate different ground truth confidence maps. Different stages (T) were also used to refine the confidence map predicted. To evaluate the robustness of our method, we compared it with two state-of-the-art object detection CNN methods (Faster R-CNN and RetinaNet). The results show better performance with the combination of green, red and near-infrared bands, achieving a Mean Absolute Error (MAE), Mean Square Error (MSE), R2 and Normalized Root-Mean-Squared Error (NRMSE) of 2.28, 9.82, 0.96 and 0.05, respectively. This band combination, when adopting σ = 1 and a stage (T = 8), resulted in an R2, MAE, Precision, Recall and F1 of 0.97, 2.05, 0.95, 0.96 and 0.95, respectively. Our method outperforms significantly object detection methods for counting and geolocation. It was concluded that our CNN approach developed to estimate the number and geolocation of citrus trees in high-density orchards is satisfactory and is an effective strategy to replace the traditional visual inspection method to determine the number of plants in orchards trees.

Automatic sleep stages classification using optimize flexible analytic wavelet transform

Sleep stages classification avails the diagnosis and treatment of sleep-related disorders. The traditional visual inspection methods used by sleep-experts are time-consuming and error-prone. This framework proposes, an automatic sleep stages classification method based on optimize flexible analytic wavelet transform (OFAWT) for electroencephalogram (EEG) signals. In OFAWT, the parametric optimization is performed to obtain the most appropriate basis for the representation of EEG signals. The OFAWT parameters are selected by solving inequality constraints problem using the genetic algorithm. OFAWT decomposes EEG signal into band-limited basis or sub-bands (SBs). Time domain measures of SBs are used as features for the sleep stages EEG signals. The statistical significance of extracted features is assessed by multiple-comparison post hoc analysis of Kruskal–Wallis test, which ensures that reported features are statistically significant for the discrimination of sleep stages. The SB-wise features set is tested through the variants of decision tree, discriminant analysis, k-nearest neighbor, and ensemble classifiers for sleep stages classification. The ensemble classification model bagged-tree yields better classification accuracies for the classification of six to two sleep stages 96.03%, 96.39%, 96.48%, 97.56%, and 99.36%, respectively as compared to other existing methods.

An Examination of Measurement Procedures and Characteristics of Baseline Outcome Data in Single-Case Research

There has been growing interest in using statistical methods to analyze data and estimate effect size indices from studies that use single-case designs (SCDs), as a complement to traditional visual inspection methods. The validity of a statistical method rests on whether its assumptions are plausible representations of the process by which the data were collected, yet there is evidence that some assumptions-particularly regarding normality of error distributions-may be inappropriate for single-case data. To develop more appropriate modeling assumptions and statistical methods, researchers must attend to the features of real SCD data. In this study, we examine several features of SCDs with behavioral outcome measures in order to inform development of statistical methods. Drawing on a corpus of over 300 studies, including approximately 1,800 cases, from seven systematic reviews that cover a range of interventions and outcome constructs, we report the distribution of study designs, distribution of outcome measurement procedures, and features of baseline outcome data distributions for the most common types of measurements used in single-case research. We discuss implications for the development of more realistic assumptions regarding outcome distributions in SCD studies, as well as the design of Monte Carlo simulation studies evaluating the performance of statistical analysis techniques for SCD data.

Research on Fatigue Crack Growth Detection of M (T) Specimen Based on Image Processing Technology

Detection of the fatigue crack propagation length with high precision is the key to correctly estimating the crack propagation life. In view of the cumbersome testing process, heavy labor intensity and many influences of subjective factors from the operators of the traditional detection methods, the image processing technology based on machine vision is applied to detect the growth state of laboratory standard M (T) specimen with a fatigue crack. First of all, several image de-noising methods are compared to verify that the median filtering method is more suitable for the processing of fatigue crack images in fatigue specimens under laboratory conditions. Secondly, the crack extraction algorithm based on Hough transform and regional growth is determined. Finally, the Zhang thinning algorithm based on the local binary texture feature is improved to refine the image, and the corresponding burr removal algorithm is proposed to refine the images. Compared with the traditional visual inspection method, the image processing technology proposed in this paper can obtain good processing results. The measured crack length error is smaller and smaller with the development of the crack propagation state and can replace the traditional visual inspection method to detect cracks extended status.

Regionally Enhanced Multiphase Segmentation Technique for Damaged Surfaces

AbstractImaging‐based damage detection techniques are increasingly being utilized alongside traditional visual inspection methods to provide owners/operators of infrastructure with an efficient source of quantitative information for ensuring their continued safe and economic operation. However, there exists scope for significant development of improved damage detection algorithms that can characterize features of interest in challenging scenes with credibility. This article presents a new regionally enhanced multiphase segmentation (REMPS) technique that is designed to detect a broad range of damage forms on the surface of civil infrastructure. The technique is successfully applied to a corroding infrastructure component in a harbour facility. REMPS integrates spatial and pixel relationships to identify, classify, and quantify the area of damaged regions to a high degree of accuracy. The image of interest is preprocessed through a contrast enhancement and color reduction scheme. Features in the image are then identified using a Sobel edge detector, followed by subsequent classification using a clustering‐based filtering technique. Finally, support vector machines are used to classify pixels which are locally supplemented onto damaged regions to improve their size and shape characteristics. The performance of REMPS in different color spaces is investigated for best detection on the basis of receiver operating characteristics curves. The superiority of REMPS over existing segmentation approaches is demonstrated, in particular when considering high dynamic range imagery. It is shown that REMPS easily extends beyond the application presented and may be considered an effective and versatile standalone segmentation technique.

Wide band energy analysis of fluidized bed pressure fluctuation signals using a frequency division method

A statistical method based on approximation of the cumulative energy distribution by Student's t-distribution is proposed for the unbiased frequency domain division. The proposed method fixes the number of samples needed to estimate the power spectrum and its corresponding cumulative energy distribution using the Kolmogorov–Smirnov test. The reliability of the method to divide the frequency domain was shown for different fluidization velocities by changing the bed aspect ratio and using different pressure probes. Water-induced defluidization tests were conducted to illustrate the use of wide band energy as a monitoring tool. The Student's t-distribution results are compared with an analysis performed using the traditional visual inspection method. The energy of the power spectrum contained within the frequency regions obtained by the visual method is not able to detect changes in the bed aspect ratio or the start of the rotating distributor. No meaningful differences could be observed in the frequency regions using different quality pressure sensors because the approach using Student's t-distribution focuses on the sharp energy increase produced by the primary frequencies of the power spectrum. The sensitivity exhibited by the proposed frequency division approach for the range of fluidization conditions tested improves the use of the energy contained in these regions as a diagnostic tool in fluidized bed processes.

Clinical applications of spectral analysis and extraction of features from electroencephalograms with slow waves in adult patients

A method is described for extracting from EEGs clinically relevant features, using a small computer. The basic principle is to compute the activity in different frequency bands (delta, theta, alpha, beta), using spectral analysis, and to study some of the relations between these activities. In patients with localized supratentorial brain lesions, the EEG usually presents localized slow waves and also decreased background activity (alpha and beta) in areas nearest to the lesion. A ration of the type (delta + theta)/(alpha + beta) should therefore act as a good indicator of the importance of the abnormality. This ration is improved by using an empirical type of discriminant analysis, applied to a group of normal subjects and to a group of patients. Using a 40 sec sample of the EEG, a concentrated report in the form of a visual, topographically organized display is issued on the computer graphic terminal after a few minutes of computation, reflecting the values of the feature for each of the 16 channels of EEG and some of the relations between channels. The readability of this display to a non-electroencephalographer is a useful feature, especially since it presents in a topographical arrangement clinically significant numerical features which can be apprehended at a glance and are not available to the interpreter of the EEG using the traditional visual inspection method.