Image Thresholding Research Articles

Improving pipe perforation estimates from ultrasonic imaging using subpixel machine learning trained on optical data

During well completion, well pipes are perforated to gain access the reservoir. The size and shape of the perforated holes can be found from these holes’ outlines, which are generally estimated by optical or ultrasonic image logging. While optical imaging typically has a much higher spatial resolution than ultrasonic imaging and thus allows more precise outline estimates, optical imaging requires transparent liquids inside the pipe. Ultrasonic imaging, on the other hand, can be performed in a wider variety of liquids and can provide further information about the well state from the same measurement. One strategy is therefore to combine both types of measurement on the same toolstring. Thus, ultrasonic imaging can take over the job of estimating hole outlines from optical imaging when the liquid is no longer sufficiently transparent. One issue with this strategy is that the agreement between these two imaging techniques currently leaves much to be desired. This work addresses this issue by training a machine learning (ML)-based subpixel segmentation algorithm to take ultrasonic images of perforations and reproduce perforation outline estimates made from optical images. This approach assists the algorithm in drawing out information from the ultrasonic data which is not easily accessible using traditional image processing techniques. We use a dataset of 390 perforations, measured by both an optical and an ultrasonic tool, to train and test the machine learning algorithm. For comparison, we use a baseline algorithm based on interpolation and image thresholding. We evaluate the algorithms’ performance according to their estimated outlines’ match with the optical outlines. The outlines’ overlap is quantified via the intersection over union metric (baseline: 50.8%, ML: 74.2%; higher is better), and their area match is quantified via mean relative area error and compared to the results of another study from the literature (baseline: 54.4%, ML: 18.7%, other study: 54.5%; lower is better).

An efficient multilevel thresholding image segmentation through improved elephant herding optimization

An efficient multilevel thresholding image segmentation through improved elephant herding optimization

A multi-level thresholding image segmentation algorithm based on equilibrium optimizer

Multi-level thresholding for image segmentation is one of the key techniques in image processing. Although numerous methods have been introduced, it remains challenging to achieve stable and satisfactory thresholds when segmenting images with various unknown properties. This paper proposes an equilibrium optimizer algorithm to find the optimal multi-level thresholds for grayscale images. The proposed algorithm AEO (advanced equilibrium optimizer) uses two sub-populations to balance exploration and exploitation during the multi-level threshold search process. Two mutation schemes are proposed for the sub-populations to prevent them from being trapped in local optima. AEO offers a repair function to avoid generating duplicate thresholds. The performance of AEO is evaluated on multiple benchmark images. Experimental results demonstrate that AEO has an outstanding ability for multi-level threshold image segmentation in terms of cross-entropy, signal-to-noise ratio (PSNR), structural similarity index measure (SSIM), and feature similarity index (FSIM).

Adaptive Mesh Generation and Numerical Verification for Complex Rock Structures Based on Optimization and Iteration Algorithms

ABSTRACTThe modelling of rock structure is of great significance in characterizing rock characteristics and studying the failure laws of rock samples. In order to construct a high‐fidelity model of the rock structure efficiently, this paper proposes an adaptive mesh dissection algorithm based on the Voronoi structure. Image processing techniques, including greyscale, threshold segmentation and edge detection, are applied to simplify the original rock image into a feature edge image. Then, a probability density diagram of the feature image is generated, which provides a probabilistic basis for the subsequent spreading of mesh seed points. Moreover, the concept of polygonal representation rate and the mesh quality evaluation system of four‐dimensional metrics are established to suggest values for the seed point parameters of the initial mesh. The initial mesh is continuously optimized and iterated by barycentric iteration and gradient descent optimization methods to form mesh structural models with high representational performance efficiently. The model tests on particle, fracture and multi‐phase rock images show that the optimized mesh model is highly similar to the original image in terms of similarity and edge fit, and the algorithm significantly reduces the short‐edge rate and improves the shape regularity of the mesh structure. Finally, numerical tests of uniaxial compression are carried out based on the optimized mesh model. The results show that the model has computational potential in numerical calculations. This method builds a procedural structure from digital images to numerical models, which can provide a reliable model basis for simulating the physico‐mechanical behaviour of heterogeneous rocks.

Precision Ice Detection on Power Transmission Lines: A Novel Approach with Multi-Scale Retinex and Advanced Morphological Edge Detection Monitoring.

Line icings on the power transmission lines are dangerous risks that may lead to situations like structural damage or power outages. The current techniques used for identifying ice have certain drawbacks, particularly when used in complex environments. This paper aims to detect lines on the top and bottom in PTLI with low illumination and complex backgrounds. The proposed method integrates multistage image processing techniques, including image enhancement, filtering, thresholding, object isolation, edge detection, and line identification. A binocular camera is used to capture images of PTLI. The effectiveness of the method is evaluated through a series of metrics, including accuracy, sensitivity, specificity, and precision, and compared with existing methods. It is observed that the proposed method significantly outperforms the existing methods of ice detection and thickness measurement. This paper uses average accuracy of detection and isolation of ice formations under various conditions at a percentage of 98.35, sensitivity at 91.63%, specificity at 99.42%, and precision of 96.03%. Furthermore, the accuracy of the ice thickness based on the thickness measurements is shown with a much smaller RMSE of 1.20 mm, MAE of 1.10 mm, and R-squared of 0.95. The proposed scheme for ice detection provides a more accurate and reliable method for monitoring ice formation on power transmission lines.

Intelligent inspection and quality control of table olives

Quality is a key factor of product-marketing in the field of agriculture and food industry. A number of researches have suggested the use of vision computer systems and machine learning (ML) techniques essentially to inspect imperfection surfaces and defects in fruit products. Following the trend, this paper still relies on the use of vision computer system by assuring an intelligent detection of table olive defect, with a certainty of its quality control. The specifically exploited computer vision system consists of automatically extracting the texture, colour, and shape features mainly from the global thresholding segmented image. With reference to the extracted features, the newly introduced system has the capacities of distinguishing between the defected and the healthy olive fruits rapidly and effectively at the same time. Subsequently, it is capable of identifying and specifying the diseased table olive. This system is also helpful for estimating the table olive size automatically. The experimental findings are indicative of the high accuracy of the binary classification algorithm, reaching 99, 32%, the average processing time for just one olive is about 0.4 s, which could meet the requirements of the real-time applications, and the error in estimating the size of the table olives does not exceed 10%.

MRI Image Enhancement Using Multilevel Image Thresholds Based on Contrast-limited Adaptive Histogram Equalization

Magnetic resonance imaging (MRI) creates detailed images by combining a powerful magnetic field with high-frequency radio waves. MRI images need to increase brightness and contrast to facilitate distinguishing diseases. This study aims to improve the MRI images depending on the suggested algorithm, including three main stages. The first is to divide the image into three main areas using Multilevel image thresholds and then improve using CLAHE for each area separately. Finally, these images combine to form one image and improve it again. By analyzing the results and calculating non-referenced quality measures, the proposed method obtained the best quality results compared to the rest of the methods with quality rates of EN(6.14), AG(6.79), and CEM(0.84), which indicates excellent success in increasing clarity in those images.

Mosaicking based optimal threshold image enhancement for violence detection with deep quadratic attention mechanism

Violence is a prevalent societal issue that poses significant threats to individuals and communities. To address this challenge, researchers have developed various machine learning models to detect and prevent violent incidents. The violence detection problem statement is a critical challenge that requires innovative and collaborative solutions. By leveraging advancements in artificial intelligence, machine learning, and data analytics, systems are developed to effectively detect violence, prevent its occurrence, and protect vulnerable individuals and communities. Therefore proposed framework utilizes Differential Evolution Algorithm (DEA) along with mosaicking process which helps in enhancing the quality of the image in order to detect the presence of violence. Hockey fight dataset consist of 40 video clips out of which 20 are labelled as fight videos and remaining 20 is labelled as non-fight videos. The image is split as 30:20 ratio and the image will undergo either brightness enhancement or contrast enhancement according to the threshold value obtained by the fitness function. In mosaicking process, Ruled KNN based SURF algorithm is employed in the proposed model as it works faster and helps in providing high accuracy which assist in detecting the violence present in the images. The classification technique employed in the proposed method is Deep Quadratic Attention Mechanism (DQAM). It helps in providing better results for image classification, reducing the convergence rate and helps in improving the normalization. The performance of the proposed method is evaluated using different performance metrics which includes PSNR, SSIM and proposed method is compared with previous phase model in comparative analysis for comparing the efficiency of the proposed and existing model.

Chronic Lower Respiratory Diseases detection based on Deep Recursive Convolutional Neural Network

Recently, symptoms of Chronic Obstructive Pulmonary Disease (COPD) have been identified concerning long-term continuous treatment. Furthermore, predicting the life probability of patients with COPD is crucial for formative ensuing treatment and conduct plans. Additionally, it plays a vital role in providing complementary solutions using technologies such as Deep Learning (DL) to address experiments in the medical field. Early and timely analysis of clinical images can improve prognostic accuracy. These include COPD, pneumonia, asthma, tuberculosis and fibrosis. Conventional methods of diagnosing COPD often rely on physical exams and tests such as spirometers, chest and genetic analysis. However, respiratory diseases pose an enormous comprehensive health burden for many patients. Thus these methods are not always accurate or obtainable. However, succeeding in their accuracy involves a nonspecific diagnosis rate, time-consuming manual procedures, and extensive clinical imaging knowledge of the radiologist. To solve this problem, we use a Deep Recursive Convolutional Neural Network (DRCNN) method to detect chronic lower respiratory disease. Initially, we collected the images from the Kaggle repository, and evaluate the result based on the following stage. The first stage is pre-processing using a Gaussian filter to reduce noise and detect the edges. The second stage is segmentation used on Image Threshold Based Segmentation (ITBS), used for counting the binary image and separating the regions. In the third stage, we use the chi-square test to select the best features and evaluate the image values for each feature and threshold. Finally, classification using DRCNN detects CLRD classifying better than the previous method. In synthesis, CLRD can be detected by many staging measures, such as sensitivity, specificity, accuracy, precision, and Recall

Improved snow ablation optimization for multilevel threshold image segmentation

Improved snow ablation optimization for multilevel threshold image segmentation

A Deep Learning Biomimetic Milky Way Compass.

Moving in straight lines is a behaviour that enables organisms to search for food, move away from threats, and ultimately seek suitable environments in which to survive and reproduce. This study explores a vision-based technique for detecting a change in heading direction using the Milky Way (MW), one of the navigational cues that are known to be used by night-active insects. An algorithm is proposed that combines the YOLOv8m-seg model and normalised second central moments to calculate the MW orientation angle. This method addresses many likely scenarios where segmentation of the MW from the background by image thresholding or edge detection is not applicable, such as when the moon is substantial or when anthropogenic light is present. The proposed YOLOv8m-seg model achieves a segment mAP@0.5 of 84.7% on the validation dataset using our own training dataset of MW images. To explore its potential role in autonomous system applications, we compare night sky imagery and GPS heading data from a field trial in rural South Australia. The comparison results show that for short-term navigation, the segmented MW image can be used as a reliable orientation cue. There is a difference of roughly 5-10° between the proposed method and GT as the path involves left or right 90° turns at certain locations.

Halvorsen chaotic system based microwave absorber modelling for fighter jet stealth technologies

This study focuses on the development and detailed analysis of a broadband microwave absorber utilizing Halvorsen chaotic dynamics, that focuses at enhancing stealth capabilities for fighter jets. The investigation begins by exploring the mathematical formulation of the Halvorsen chaotic system and conducting a parametric sweep of its control parameters to generate distinct two-dimensional and three-dimensional chaotic attractor plots. These plots are then post-processed using Julia set theory to develop intricate fractal patterns, which serve as the foundation for the absorber design. Image processing techniques, including filtering and thresholding, are employed to refine the patterns by removing artifacts and noise, ensuring they are suitable for practical implementation. The refined fractal patterns are then imported into a computational electromagnetic simulation environment where they are patterned onto a 0.035 mm thick copper sheet. Moreover, the Magtrex 555 substrate with a thickness of 1.52 mm, is selected for its high permittivity and low-loss characteristics. A comprehensive series of parametric studies are conducted to evaluate the influence of various design parameters such as side length, unit cell geometry, and substrate thickness on the absorber’s electromagnetic performance. Important parameters include the effects of chaotic control parameter optimization and the electromagnetic boundary conditions applied during the simulations. Extensive simulations are performed across the 2–20 GHz frequency range to evaluate absorption efficiency, focusing on key metrics like absorptivity, surface current distribution, and electric field distribution. The final design achieves over 90 % absorption efficiency within the target frequency band when the chaotic control parameters are optimized. Finally, comparative analysis using different commercially available substrates, including FR-4 and Rogers RO3003, reveals that Magtrex 555 offers superior absorption performance. The study concludes with a detailed presentation of the final absorber design, alongside an indepth discussion of the frequency range, parametric variations, and the impact of chaotic system dynamics on the absorption properties. This research provides crucial insights into the design and optimization of chaotic-system-based microwave absorbers, that advances the development of stealth technology in military applications.

Detection and isolation of brain tumors in cancer patients using neural network techniques in MRI images

MRI imaging primarily focuses on the soft tissues of the human body, typically performed prior to a patient's transfer to the surgical suite for a medical procedure. However, utilizing MRI images for tumor diagnosis is a time-consuming process. To address these challenges, a new method for automatic brain tumor diagnosis was developed, employing a combination of image segmentation, feature extraction, and classification techniques to isolate the specific region of interest in an MRI image corresponding to a brain tumor. The proposed method in this study comprises five distinct steps. Firstly, image pre-processing is conducted, utilizing various filters to enhance image quality. Subsequently, image thresholding is applied to facilitate segmentation. Following segmentation, feature extraction is performed, analyzing morphological and structural properties of the images. Then, feature selection is carried out using principal component analysis (PCA). Finally, classification is performed using an artificial neural network (ANN). In total, 74 unique features were extracted from each image, resulting in a dataset of 144 observations. Principal component analysis was employed to select the top 8 most effective features. Artificial Neural Networks (ANNs) leverage comprehensive data and selective knowledge. Consequently, the proposed approach was evaluated and compared with alternative methods, resulting in significant improvements in precision, accuracy, and F1 score. The proposed method demonstrated notable increases in accuracy, with improvements of 99.3%, 97.3%, and 98.5% in accuracy, Sensitivity and F1 score. These findings highlight the efficiency of this approach in accurately segmenting and classifying MRI images.

Imbalanced image classification algorithm based on fine-grained analysis

Fine-grained attribute analysis and data imbalance have always been research hotspots in the field of computer vision. Due to the complexity and diversity of fine-grained attribute images, traditional image classification methods have shortcomings in paying attention to fine-grained attributes of images and perform poorly when dealing with imbalanced data sets. To overcome these problems, this study proposes a fine-grained image threshold classification algorithm based on deep metric learning. By introducing a metric learning method, the focus on fine-grained attributes of images is enhanced. At the same time, by applying pairwise loss and proxy loss, the classification accuracy of the model is improved and the model convergence speed is accelerated. In order to deal with the problem of data imbalance, a classifier based on threshold analysis is designed. The classifier uses threshold analysis technology to achieve multi-level classification of fine-grained images, thereby improving the problem of low classification accuracy of a few categories in imbalanced data sets. Experimental results show that the proposed fine-grained image threshold classification algorithm based on deep metric learning is significantly superior to other methods in terms of classification accuracy.

Multi-mode high resolution TFM imaging of microdefects based on laser ultrasonic full matrix capture

The identification and visualization of submillimeter microdefects have long been a critical focus within the realm of ultrasonic testing. The utilization of high-frequency signals is imperative due to the small acoustic interaction structure. An additional benefit of using lasers is their non-contact nature and ability to utilize multiple imaging modes, effectively reducing false or missed detections and enhancing credibility. This study proposes an advanced optimization algorithm for imaging and quantitative evaluation of submillimeter microdefects using laser ultrasonic full matrix capture (FMC) and total focusing method (TFM). The analysis of acoustic directionality under laser excitation and detection provides a foundation for mode selection. In preprocessing, the multi-scale principal component analysis (MSPCA) method is introduced to suppress coherent noise and blind spots. Moreover, phase coherence methods are implemented to further reduce mode artifacts, background noise, and residual surface acoustic wave (SAW). Additionally, multi-mode imaging and mode fusion are achieved using reflected longitudinal waves, reflected shear waves, converted longitudinal waves, and converted shear waves. Experiments were carried out on two different types of microdefects, namely side drilled holes (SDHs) and blind holes (BHs), resulting in high resolution TFM imaging with a minimum BH size of 0.3 mm. The outcomes demonstrate the efficacy of MSPCA in suppressing coherent noise, with further enhancement in imaging quality after phase coherence weighting. In addition, the −6 dB threshold of multi-mode images also provided the size information. This research is poised to contribute to the resolution of quality assessment challenges in powder metallurgy, welding, and additive manufacturing processes.

The extremely strong non-neutralized electric currents of the unique solar active region NOAA 13664

Context. In May 2024, the extremely complex active region National Oceanic and Atmospheric Administration (NOAA) 13664 produced the strongest geomagnetic storm since 2003. Aims.The aim of this study is to explore the development of the extreme magnetic complexity of NOAA 13664 in terms of its photospheric electric current. Methods. The non-neutralized electric current was derived from photospheric vector magnetograms, provided by the Helioseismic and Magnetic Imager on board the Solar Dynamics Observatory. The calculation method is based on image processing, thresholding, and error analysis. The spatial and temporal evolution of the non-neutralized electric current of the region as well as its constituent subregions was examined. For context, a comparison with other complex, flare-prolific active regions is provided. Results. Active region NOAA 13664 was formed by the emergence and interaction of three subregions, two of which were of notable individual complexity. It consisted of numerous persistent, current-carrying magnetic partitions that exhibited periods of conspicuous motions and strongly increasing electric current at many locations within the region. These periods were followed by intense and repeated flaring. The total unsigned non-neutralized electric currents and average injection rates reached 5.95 ⋅ 1013 A and 1.5 ⋅ 1013 A/day, and are the strongest observed so far, significantly surpassing other super-active regions of Solar Cycle 24 and 25. Conclusions. Active region NOAA 13664 presents a unique case of complexity. Further scrutiny of the spatial and temporal variation of the net electric currents during the emergence and development of super-active regions is paramount to understand the origin of complex regions and adverse space weather.

Automatic thresholding method using single information entropy under product transformation of order difference filter response

To automatically threshold images with unimodal, bimodal, multimodal or non-modal gray level distributions within a unified framework, an automatic thresholding method using single information entropy under the product transformation of order difference filter response is proposed. The proposed method first performs the product transformation of order difference filter response on an input image at different scales to obtain the product transformation image. Critical or non-critical pixels are labelled on each pixel of the binary images corresponding to different thresholds to construct a series of binary label images that are used for distinguishing critical or non-critical regions. A single information entropy is finally used for characterizing the information obtained from the product transformation image with the critical regions of different binary label images, and the threshold corresponding to maximum information entropy is selected as final threshold. The proposed method is compared with seven state-of-the-art segmentation methods. Experimental results on 12 synthetic images and 98 real-world images show that the average Matthews correlation coefficients of the proposed method reached 0.994 and 0.966 for the synthetic images and the real-world images, which outperform the second-best method by 52.4 % and 27.8 %, respectively. The proposed method has more robust segmentation adaptability to test images with different modalities, despite not offering an advantage in terms of computational efficiency.

Heterogeneous biomechanical/mathematical modeling of spatial prediction of glioblastoma progression using magnetic resonance imaging-based finite element method

Background and ObjectiveBrain tumors are one of the most common diseases and causes of death in humans. Since the growth of brain tumors has irreparable risks for the patient, predicting the growth of the tumor and knowing its effect on the brain tissue will increase the efficiency of treatment strategies. MethodsThis study examines brain tumor growth using mathematical modeling based on the Reaction-Diffusion equation and the biomechanical model based on continuum mechanics principles. With the help of the image threshold technique of magnetic resonance images, a heterogeneous and close-to-reality environment of the brain has been modeled and experimental data validated the results to achieve maximum accuracy in predicting growth. ResultsThe obtained results have been compared with the reported conventional models to evaluate the presented model. In addition to incorporating the chemotherapy effects in governing equations, the real-time finite element analysis of the stress tensors of the surrounding tissue of tumor cells and considering its role in changing the shape and growth of the tumor has added to the importance and accuracy of the current model. ConclusionsThe comparison of the obtained results with conventional models shows that the heterogeneous model has higher reliability due to the consideration of the appropriate properties for the different regions of the brain. The presented model can contribute to personalized medicine, aid in understanding the dynamics of tumor growth, optimize treatment regimens, and develop adaptive therapy strategies.

Energy curve based enhanced smell agent optimizer for optimal multilevel threshold selection of thermographic breast image segmentation

Multilevel thresholding image segmentation will subdivide an image into several meaningful regions or objects, which makes the image more informative and easier to analyze. Optimal multilevel thresholding approaches are extensively used for segmentation because they are easy to implement and offer low computational cost. Multilevel thresholding image segmentation is frequently performed using popular methods such as Otsu’s between-class variance and Kapur’s entropy. Numerous researchers have used evolutionary algorithms to identify the best multilevel thresholds based on the above approaches using histogram. This paper uses the Energy Curve (EC) based thresholding method instead of the histogram. Chaotic Bidirectional Smell Agent Optimization with Adaptive Control Strategy (ChBSAOACS), a powerful evolutionary algorithm, is developed and employed in this paper to create and execute an effective method for multilevel thresholding segmentation of breast thermogram images based on energy curves. The proposed algorithm was tested for viability on standard breast thermogram images. All experimental data are examined quantitatively and qualitatively to verify the suggested method’s efficacy.

Zero-shot image segmentation for monitoring thermal conditions of individual cage-free laying hens

Body temperature is a critical indicator of the health and productivity of egg-laying chickens and other domesticated animals. Recent advancements in thermography allow for precise surface temperature measurement without physical contact with animals, reducing animal stress from human handling. Gold standard temperature analysis via thermography requires manual selection of limited points for an object of interest, which could be time-consuming and inadequate for representing the comprehensive thermal profile of a chicken’s body. The objective of this study was to leverage and optimize a zero-shot artificial intelligence technology for the automatic segmentation of individual cage-free laying hens within thermal images, providing insights into their overall thermal conditions. A zero-shot image segmentation model (Segment Anything, “SAM”) was modified by replacing manual selections of target points with automatic selection of the initial point using pre-processing techniques (e.g., thresholding) in each thermal image. The model was also incorporated with post-processing techniques integrated with a machine learning classifier to improve segmentation accuracy. Three versions of modified SAM models (i.e., SAM, FastSAM, and MobileSAM), two common instance segmentation algorithms (i.e., YOLOv8 and Mask R-CNN), and two foundation segmentation models (i.e., U2-Net and ISNet) were comparatively evaluated to determine the optimal one for bird segmentation from thermal images. A total of 1,917 thermal images were collected from cage-free laying hens (Hy-Line W-36) at 77–80 weeks of age. The image dataset exhibited considerable variations such as feathers, bird movement, body gestures, and the specific conditions of cage-free facilities. The experimental results demonstrate that the modified SAM did not only surpass the six zero-shot models—YOLOv8, Mask R-CNN, FastSAM, MobileSAM, U2Net, and ISNet—but also outperformed other modified SAM-based models (Modified FastSAM and Modified MobileSAM) in terms of hen detection performance, achieving a success rate of 84.4 %, and in segmentation performance, with an intersection over union of 85.5 %, recall of 91.0 %, and an F1 score of 92.3 %. The optimal model, modified SAM, was pipelined to extract statistics including the averages (°C) of mean (27.03, 27.04, 28.53, 26.68), median (26.27, 26.84, 28.28, 26.78), 25th percentile (25.33, 25.61, 27.26, 25.53), and 75th percentile (28.04, 27.95, 29.22, 27.55) of surface body temperature of individual laying hens in thermal images for each week. More statistics of hen body surface temperature can be extracted based on the segmentation results. The developed pipeline is a useful tool for automatically evaluating the thermal conditions of individual birds.