Binarization Technique Research Articles

MACHINE LEARNING-DRIVEN PREDICTIVE MAINTENANCE: OPTIMIZING NAIVE BAYES WITH SYNTHETIC DATA AND CLASS IMBALANCE TECHNIQUES

Integrating machine learning into predictive maintenance has steadily gained traction across industries, revolutionizing maintenance strategies. While numerous studies have investigated this topic, and many sectors have successfully implemented machine learning models, some industries rely on inefficient and costly traditional methods. This project aims to bridge that gap by advancing machine learning-driven predictive maintenance research utilizing a synthetic dataset. Specifically, the study explores a novel approach by combining binarisation techniques with the Naive Bayes algorithm—an area largely underexplored in existing literature. Additionally, Naive Bayes is enhanced with Bagging and Boosting techniques to improve performance. At the same time, SMOTE (Synthetic Minority Over-Sampling Technique) and under-sampling are applied to address the class imbalance in predicting machine failure. Six models were developed: Naive Bayes [SMOTE], Naive Bayes with Bagging [SMOTE], Naive Bayes with Boosting [SMOTE], Naive Bayes [Under sampling], Naive Bayes with Bagging [Under sampling], and Naive Bayes with Boosting [Under sampling]. Among these, the Naive Bayes [SMOTE] model achieved outstanding results, with an accuracy of 0.999 and a precision of 1.0, outperforming previous studies and setting a new benchmark in predictive maintenance research. These findings highlight the potential of advanced machine learning techniques in significantly improving predictive maintenance accuracy and efficiency across industries.

Towards the development of an algorithm for automatic analysis and detection of shock waves in the schlieren visualization technique

This paper presents the development of an algorithm for identifying shock waves formed during the supersonic combustion process, detonation, in a pulsed detonation chamber. The algorithm is based on image binarization techniques, followed by the identification of regions of interest in the images, the detection and isolation of the edges of these regions, and the determination of the wave speed between two consecutive frames. The advantage of this automation process is the efficiency in analysing the repeated detonation cycles of a pulsed detonation chamber. Additionally, it can provide a more accurate evaluation of the data, making it a valuable tool for researchers and engineers in various scientific fields and shock wave applications.

Methods of medical image segmentation analysis to improve the effectiveness of diagnosing lung diseases

Methods of medical image segmentation analysis to improve the effectiveness of diagnosing lung diseases. The article aims to review selected methods for segmenting lung medical images based on computed tomography (CT) image data analysis. Included in the analysis are 11 different approaches, including edge algorithms such as Canny, Prewitt, Roberts, Sobel and Log, clustering-based algorithms such as Fuzzy C-means, genetic K-means, k-nearest neighbours, and neural networks such as perceptron, neural and fuzzy sets, and edge binarization techniques, among others. Each method was evaluated for its ability to segment lung structures on CT images accurately. Finally, the contour of the heart from the CT image was determined using the maximum entropy thresholding method. The juxtaposition of different approaches to medical image segmentation is an important contribution to developing medical diagnostic techniques, which can help improve the efficiency of diagnosing lung diseases.

A novel method of ultrasonic tomographic imaging of defects in the coating layer by image fusion and binarization techniques

A novel method of ultrasonic tomographic imaging of defects in the coating layer by image fusion and binarization techniques

Performance of Binarization Algorithms on Tamizhi Inscription Images: An Analysis

Binarization of Tamizhi (Tamil-Brahmi) inscription images are highly challenging, as it is captured from very old stone inscriptions that exists around 3rd century BCE in India. The difficulty is due to the degradation of these inscriptions by environmental factors and human negligence over ages. Though many works have been carried out in the binarization of inscription images, very little research was performed for inscription images and no work has been reported for binarization of inscriptions inscribed on irregular medium. The findings of the analysis hold true to all writings that are carved in irregular background. This article reviews the performance of various binarization techniques on Tamizhi inscription images. Since no previous work was performed, we have applied the existing binarization algorithms on Tamizhi inscription images and analyzed the performance of these algorithms with proper reasoning. In the future, we believe that this reasoning on the results will help a new researcher to adapt or combine or devise new binarization techniques.

A Review of Document Binarization: Main Techniques, New Challenges, and Trends

Document image binarization is a challenging task, especially when it comes to text segmentation in degraded document images. The binarization, as a pre-processing step of Optical Character Recognition (OCR), is one of the most fundamental and commonly used segmentation methods. It separates the foreground text from the background of the document image to facilitate subsequent image processing. In view of the different degradation degrees of document images, researchers have proposed a variety of solutions. In this paper, we have summarized some challenges and difficulties in the field of document image binarization. Approximately 60 methods documenting image binarization techniques are mentioned, including traditional algorithms and deep learning-based algorithms. Here, we evaluated the performance of 25 image binarization techniques on the H-DIBCO2016 dataset to provide some help for future research.

Challenging the Limits of Binarization: A New Scheme Selection Policy Using Reinforcement Learning Techniques for Binary Combinatorial Problem Solving.

In this study, we introduce an innovative policy in the field of reinforcement learning, specifically designed as an action selection mechanism, and applied herein as a selector for binarization schemes. These schemes enable continuous metaheuristics to be applied to binary problems, thereby paving new paths in combinatorial optimization. To evaluate its efficacy, we implemented this policy within our BSS framework, which integrates a variety of reinforcement learning and metaheuristic techniques. Upon resolving 45 instances of the Set Covering Problem, our results demonstrate that reinforcement learning can play a crucial role in enhancing the binarization techniques employed. This policy not only significantly outperformed traditional methods in terms of precision and efficiency, but also proved to be extensible and adaptable to other techniques and similar problems. The approach proposed in this article is capable of significantly surpassing traditional methods in precision and efficiency, which could have important implications for a wide range of real-world applications. This study underscores the philosophy behind our approach: utilizing reinforcement learning not as an end in itself, but as a powerful tool for solving binary combinatorial problems, emphasizing its practical applicability and potential to transform the way we address complex challenges across various fields.

Conical shell X-ray beam tomosynthesis and micro-computed tomography for microarchitectural characterisation

Bone quality is commonly used to diagnose bone diseases such as osteoporosis, with many studies focusing on microarchitecture for fracture prediction. In this study a bovine distal femur was imaged using both micro-computed tomography (µCT) and tomosynthesis using focal construct geometry (FCG) for comparison of microarchitectural parameters. Six regions of interest (ROIs) were compared between the two imaging modalities, with both global and adaptive methods used to binarize the images. FCG images were downsampled to the same pixel size as the µCT images. Bone morphometrics were determined using BoneJ, for each imaging modality, binarization technique and ROI. Bone area/total area was found to have few significant differences between FCG and µCT (p < 0.05 for two of six ROIs). Fractal Dimension had only one significant difference (p < 0.05 for one of six ROIs) between µCT and downsampled FCG (where pixel size was equalized). Trabecular thickness and trabecular spacing were observed to follow trends as observed for the corresponding µCT images, although many absolute values were significantly different (p < 0.05 for between one and six ROIs depending on image types used). This study demonstrates the utility of tomosynthesis for measurement of microarchitectural morphometrics.

Effect of internal energy depletion on active nematic texture and dynamics

ABSTRACTActive nematics represent a relatively new class of materials with liquid crystalline properties that rely on an energy source to maintain their structure and dynamics. In this paper, we focus on the well-known microtubule-kinesin-based active nematic, powered by chemical energy as ATP (Adenosine triphosphate), and report on how ATP depletion impacts active nematic defect dynamics and texture. Using fluorescence microscope video imaging, we measure the time averaged mean defect separations and root mean square velocities of the active flows as a function of time. We also characterise textural changes and the growth of void space in the network after ATP depletion using an image binarization technique.

MLIBT: A multi-level improvised binarization technique for Tamizhi inscriptions

The Tamizhi inscriptions, one of the earliest ever discovered, is predominantly found on memorial stones and caves which dates 5th century BCE to 3rd century CE. Today’s generations need ways to interpret the script in order to know the historical figures and events because the Tamizhi script evolved into the modern Tamil script over time. Currently, only few epigraphists are available to manually decode the inscriptions into modern Tamil. Hence, there is a need for an alternate way to preserve this cultural heritage. Image processing is one such digital technology that enables binarization on inscription images, and the retrieved text may then be utilized to convert them to the needed target language. Nevertheless, binarization of Tamizhi inscription images are highly complex due to aging, environmental factors, handwritten, similar foreground & background and uneven size and shapes of the stones. Also, due to the small dataset, deep learning techniques are inapplicable. Furthermore, existing approaches produce poor results for Tamizhi inscriptions since they can only be used on flat stone backgrounds and require sufficient light illumination for effective binarization. This research suggests a multi-level improvised binarization solution for Tamizhi inscription images to address these challenges. It achieves this utilizing post-processing with shrink and swell filters together with an improved median filter with modified adaptive thresholding. Outperforming the current binarization techniques, which achieved a maximum accuracy of about 74%, MLIBT produced an accuracy of around 92.19%.

Choroidal vascular changes in non-alcoholic fatty liver disease.

The most common cause of death in nonalcoholic fatty liver disease (NAFLD) is cardiovascular disease. Choroidal microvascular structure in the eye may be a predictor of systemic vascular disease. We aimed to evaluate the effects of NAFLD on the choroidal microvascular structure using enhanced depth optical coherence tomography (EDI-OCT). This prospective study was conducted by evaluating a total of 96 patients, 52 with steatosis and 44 without steatosis. After anthropometric measurements and ultrasonography were performed in the Gastroenterology Clinic, venous blood samples were taken for biochemical examinations. Then, all patients underwent an eye examination by an ophthalmologist. Subfoveolar choroidal thickness (SFCT) values of the cases were measured with EDI-OCT. Choroid vascular index (CVI) measurements were obtained by dividing the subfoveal choroidal area in the EDI-OCT images into luminal and stromal areas using the image binarization technique (ImageJ). In statistical analysis, the chi-square test was used to compare categorical data, and the independent t-test and Mann-Whitney U test were used to compare quantitative data. The mean age of those with fatty liver was 41±15.7 years, and of those without fatty liver it was 46 ± 10.7 years. There was nostatistically significant difference between the groups in terms of age (p = 0.064). Body mass index (BMI), waist circumference (WC), glucose, uric acid, alanine aminotransferase (ALT), gamma glutamyl transpeptidase (GGT), total cholesterol (TC), ferritin, insulin, and Homestatic Model Assesment - Insuline Restistance (HOMA-IR) were statistically significantly higher in the NAFLD group. On the other hand, there was no statistically significant difference between the groups in terms of low-density lipoprotein (LDL)-cholesterol, high-density lipoprotein (HDL)-cholesterol, triglyceride, and aspartate aminotransferase (AST) values. The mean SFCT was measured as 280.26 ± 23.68 microns in the NAFLD group, and 308.96 ± 18.57 microns in the control group. There was no statistically significant difference in SFCT between the groups (p = 0.077). CVI measurements were 0.63 and 0.65, respectively, and they were significantly lower in the group with NAFLD (p = 0.045). This is the first study in the literature to compare patients with and without ultrasonographic fatty liver in terms of choroidal vascular changes. We found that the choroidal vascular index decreased in NAFLD. This result proves that NAFLD causes changes at the microvascular level and is a multisystemic disease.

Ultrafast quasi-three-dimensional imaging

Understanding laser induced ultrafast processes with complex three-dimensional (3D) geometries and extreme property evolution offers a unique opportunity to explore novel physical phenomena and to overcome the manufacturing limitations. Ultrafast imaging offers exceptional spatiotemporal resolution and thus has been considered an effective tool. However, in conventional single-view imaging techniques, 3D information is projected on a two-dimensional plane, which leads to significant information loss that is detrimental to understanding the full ultrafast process. Here, we propose a quasi-3D imaging method to describe the ultrafast process and further analyze spatial asymmetries of laser induced plasma. Orthogonally polarized laser pulses are adopted to illuminate reflection-transmission views, and binarization techniques are employed to extract contours, forming the corresponding two-dimensional matrix. By rotating and multiplying the two-dimensional contour matrices obtained from the dual views, a quasi-3D image can be reconstructed. This successfully reveals dual-phase transition mechanisms and elucidates the diffraction phenomena occurring outside the plasma. Furthermore, the quasi-3D image confirms the spatial asymmetries of the picosecond plasma, which is difficult to achieve with two-dimensional images. Our findings demonstrate that quasi-3D imaging not only offers a more comprehensive understanding of plasma dynamics than previous imaging methods, but also has wide potential in revealing various complex ultrafast phenomena in related fields including strong-field physics, fluid dynamics, and cutting-edge manufacturing.

Image to G-Code Conversion using JavaScript for CNC Machine Control

This paper presents a JavaScript-based approach for image-to-G-code conversion in CNC (Computer Numerical Control) machine control. The developed code enables the translation of images and text into machine-readable instructions for precise reproduction using CNC machines. The code implements image loading, preprocessing, binarization, thinning, and G-code generation functionalities. It offers adjustable parameters for CNC and image settings, allowing customization and optimization of the machining process. Experimental evaluations confirm the code's efficiency, accuracy, and usability. The results demonstrate reliable image preprocessing, effective binarization techniques, successful image thinning, and precise G-code generation. The code's accessibility, user-friendly interface, and real-time preview capabilities further enhance its usability. This JavaScript-based approach contributes to the integration of digital workflows into CNC machining, offering a promising solution for accurate and efficient fabrication.

Realtime Detection of Table Objects and Text Areas for OCR Preprocessing

OCR (Optical Character Recognition) is a technology that automatically detects, recognizes, and digitally converts text into images. OCR has a variety of uses, including reducing human error when viewing and typing documents and helping people work more efficiently with documents. It can increase efficiency and save money by eliminating the need to manually type text, especially when scanning documents or digitizing images. OCR is divided into text object detection and text recognition in an image, and preprocessing techniques are used during the original document imaging process to increase the accuracy of OCR results. There are various preprocessing techniques. They are generally classified into image enhancement, binarization techniques, text alignment and correction, and segmentation techniques. In this paper, we propose a special-purpose preprocessing technique and application called Table Area Detection. Recently, table detection using deep learning has been actively researched, and the research results are helping to improve the performance of table recognition technology. Table detection will become an important preprocessing technology for text extraction and analysis in various documents, and it requires a lot of research and accuracy. While many previous studies have focused on improving the accuracy of OCR algorithms through various techniques, this study proposes a method to discover and exclude false positives by introducing a factor called Table Area Detection.

Short time effect of sodium fluorescein on chorodial vascularity index

Purpose This study aimed to investigate the effect of sodium fluorescein on the choroidal vascularity index (CVI). Materials and Methods In this cross-sectional study, a total of 27 eyes of 27 mild nonproliferative diabetic retinopathy patients (without maculopathy and any systemic disease) who performed fluorescein angiography (FA) procedure were included. Choroidal parameters such as choroidal thickness (CT), total choroidal area (TCA), luminal area (LA), stromal area (SA), LA/SA, and CVI were examined with the optical coherence tomography and binarization technique at baseline and 5th, 15th, and 30th minutes after FA. The values of the parameters before and after the procedure were compared. Results At baseline, the mean values of the TCA, LA, SA, LA/SA, and CVI were 0.44 ± 0.14mm2, 0.29 ± 0.09 mm2, 0.15 ± 0.05 mm2, 1.87 ± 0.19. Five minutes later to FA, the mean values of the TCA, LA, SA, LA/SA, and CVI were 0.43 ± 0.13 mm2, 0.28 ± 0.08 mm2, 0.15 ± 0.05 mm2, 1.82 ± 0.20, and 0.64 ± 0.03, respectively. There was a significant decrease in LA and CVI values 5 minutes after FA (p:0.002 and p:0.021, respectively). On the other hand, the mean, nasal, subfoveal, and temporal CT were 279.22 ± 93.40 μm, 289.78 ± 91.17 μm, 267.44 ± 95.71 μm before FA and 270.33 ± 90.34 μm, 279.67 ± 90.01 μm, 261.82 ± 95.82 μm 5 minutes after FA (p = 0.960, p = 0.952, and p = 0.991, respectively). Although there was a reduction in the value of CT, there was not a statistically significant difference between before and after FA. Conclusions This study shows that there was a significant decrease in LA and CVI values 5 minutes after FA in patients with mild nonproliferative diabetic retinopathy.

Harnessing Neural Networks for Enhancing Image Binarization Through Threshold Combination

Threshold-based methods are prevalent across numerous domains, with specific relevance to image binarization, which traditionally employs global and local threshold algorithms. This paper presents a novel approach to image binarization, where the capacity of neural networks is utilized not just for determining optimal thresholds, but also for combining multiple global thresholds sourced from existing binarization techniques. The primary objective of our method is to develop a robust binarization strategy capable of managing a wide array of image conditions. By integrating the strengths of various thresholding techniques, our approach aims to establish a significant connection between traditional thresholding methods and those underpinned by deep learning.

An Effective Motion-Tracking Scheme for Machine-Learning Applications in Noisy Videos

Detecting and tracking objects of interest in videos is a technology that can be used in various applications. For example, identifying cell movements or mutations through videos obtained in real time can be useful information for decision making in the medical field. However, depending on the situation, the quality of the video may be below the expected level, and in this case, it may be difficult to check necessary information. To overcome this problem, we proposed a technique to effectively track objects by modifying the simplest color balance (SCB) technique. An optimal object detection method was devised by mixing the modified SCB algorithm and a binarization technique. We presented a method of displaying object labels on a per-frame basis to track object movements in a video. Detecting objects and tagging labels through this method can be used to generate object motion-based prediction training data for machine learning. That is, based on the generated training data, it is possible to implement an artificial intelligence model for an expert system based on various object motion measurements. As a result, the main object detection accuracy in noisy videos was more than 95%. This method also reduced the tracking loss rate to less than 10%.

An approach for improving Optical Character Recognition using Contrast enhancement technique

Digital cameras and mobile image capture of documents are two examples of new developments in the fields of optical character recognition and text recognition. Scans of text or text photographic images and even natural photography results can be distorted to the point where OCR digitization is inaccurate. It offers a unique non-parametric unattended approach to correct unwanted document image distortions to achieve optimal OCR accuracy. It applies a highly effective stack of document image enhancement algorithms to restore perfect images distorted by unknown sources of distortion. First, it provides a means of modifying local brightness and contrast in order to better handle different illumination levels and atypical light transmission patterns in the image. Then apply a nifty grayscale conversion method to your photo to give it a new look. Third, it uses unsharp masking techniques to further enhance important details in grayscale images. Finally, we use the best global binarization technique to prepare the final document image for OCR recognition. The proposed technique has the potential to significantly improve the text recognition rate and accuracy of optical character recognition.

An automatic classification of pulmonary nodules for lung cancer diagnosis using novel LLXcepNN classifier.

A critical step to ameliorate diagnosis and extend patient survival is Benign-malignant Pulmonary Nodule (PN) classification at earlier detection. On account of the noise of Computed Tomography (CT) images, the prevailing Lung Nodule (LN) detection techniques exhibit broad variation in accurate prediction. Thus, a novel Nodule Detection along with Classification algorithm for early diagnosis of Lung Cancer (LC) has been proposed. Initially, employing the Adaptive Mode Ostu Binarization (AMOB) technique, the Lung Volumes (LVs) isextortedas of the image together with the extracted lung regions is pre-processed. Then, detection of LNs takes place, and utilizing Geodesic Fuzzy C-Means Clustering (GFCM) Segmentation Algorithm, it is segmented.Next, the vital features are extracted, and the Nodules are classified by utilizing Logarithmic Layer Xception Neural Network (LLXcepNN) Classifier grounded on the extracted feature. The nodules are classified as Benign Nodules (BN) and Malignant Nodules (MN) by the proposed classifier. Lastly, the Lung CT images are scrutinized. Thus, when weighed against the prevailing techniques, the proposed systems' acquired outcomes exhibit that the rate of accuracy of classification is enhanced.

Vehicle Plate Number Localization Using Memetic Algorithms and Convolutional Neural Networks

This paper introduces the third enhanced version of a genetic algorithm-based technique to allow fast and accurate detection of vehicle plate numbers (VPLN) in challenging image datasets. Since binarization of the input image is the most important and difficult step in the detection of VPLN, a hybrid technique is introduced that fuses the outputs of three fast techniques into a pool of connected components objects (CCO) and hence enriches the solution space with more solution candidates. Due to the combination of the outputs of the three binarization techniques, many CCOs are produced into the output pool from which one or more sequences are to be selected as candidate solutions. The pool is filtered and submitted to a new memetic algorithm to select the best fit sequence of CCOs based on an objective distance between the tested sequence and the defined geometrical relationship matrix that represents the layout of the VPLN symbols inside the concerned plate prototype. Using any of the previous versions will give moderate results but with very low speed. Hence, a new local search is added as a memetic operator to increase the fitness of the best chromosomes based on the linear arrangement of the license plate symbols. The memetic operator speeds up the convergence to the best solution and hence compensates for the overhead of the used hybrid binarization techniques and allows for real-time detection especially after using GPUs in implementing most of the used techniques. Also, a deep convolutional network is used to detect false positives to prevent fake detection of non-plate text or similar patterns. Various image samples with a wide range of scale, orientation, and illumination conditions have been experimented with to verify the effect of the new improvements. Encouraging results with 97.55% detection precision have been reported using the recent challenging public Chinese City Parking Dataset (CCPD) outperforming the author of the dataset by 3.05% and the state-of-the-art technique by 1.45%.