Computer Image Processing Techniques Research Articles

A semi-analytical prediction model for fluid elastic instability in the streamwise direction of a normal triangular tube array

Fluid elastic instability (FEI) is widely regarded as the most destructive mechanism of flow-induced vibration. Numerous failures of heat exchanger tubes have been attributed to this phenomenon. This paper focuses on a normal triangular tube array and develops a semi-analytical time-domain solving model to address FEI in the streamwise direction. Utilizing computational fluid dynamics simulations and image processing techniques, a comprehensive set of tube-in-channel model parameters is derived, including the area perturbation amplitude, phase difference, mean area along the flow channel, and steady-state term of flow velocity. Tube motion in the streamwise direction generates a disturbance distinct from that caused by transverse flow, leading to revisions in the expressions of relevant model parameters. Furthermore, variations in these parameters across different positions and velocities are quantitatively analyzed through function fitting, yielding specific mathematical formulas. Ultimately, the derived tube-in-channel model parameters applicable to streamwise FEI are validated against experimental data, affirming the model's accuracy.

Application of computer vision techniques to damage detection in underwater concrete structures

Underwater concrete structure crack detection and structural health condition assessment based on image processing is a challenging task. The complex underwater environment and severe image degradation seriously affect the accuracy of crack detection. To solve these problems, a monocular vision and image-enhanced fractal-based fractal science based on computer vision and image processing techniques are proposed to carry out a non-contact detection study of underwater concrete cracks. In this study, a four-level structural health condition was established to assist in underwater crack measurement and safety assessment. The box-counting method was used as a practical tool to calculate the fractal dimension. To verify the effective distance of the algorithm, three distances of 0.5 m,0.8 m, and 1.2 m were set. The results show that the method proposed in this study can effectively detect cracks in submerged concrete members within 0.6 m and help managers correctly determine the health of the structure using the fractal dimension.

Identification and classification of Prakriti of human using facial features

<p>Changes in the lifestyle of individual, several of disease has emerged due to imbalance of doshas components. Ayurveda practitioners could identify the imbalance of dosha and relate the root cause of imbalance of doshas. Analysis of dosha varies from practitioner to practitioner and it requires well practiced practitioner to identify dosha. To overcome, the darshana method was adopted to automatically identify predominant dosha using facial features such as face, eyes, nose, and mouth and skin color. Computer vision and Image processing techniques were made attempted in Ayurveda domain, for identification of predominate prakriti, age, and gender of the subject. Eye Aspect Ratio (EAR), Nose Aspect Ratio (NAR), Mouth Aspect Ratio (MAR) and skin color was computed based on Euclidean distance to identify on-live predomaint prakriti of an individual. The values of MAR ≤ 0.5,EAR ≤ 0.1,NAR ≤ 0.8 as identified as vata , 0.5 ≥ MAR ≤ 0.6, 0.1 ≥ EAR ≤ 0.2, 0.8 ≥ NAR ≤ 1 as identified as pitta and MAR ≥ 0.6,EAR ≥ 0.2 and NAR ≥ 1 as identified as kapha dosha. With the features MAR, EAR and NAR classification of predominant prakriti was carried out with an accuracy of 87.5% with support vector classifier.</p>

Enhancing Brain Tumour Identification through Convolutional Neural Network

A tumor is carried on by rapid and uncontrolled cell proliferation in the brain. If it is not treated in the early stages, it could prove fatal. Despite multiple significant efforts and encouraging outcomes, accurate segmentation and classification still pose a difficulty. Detection of brain tumors is extremely complicated by the distinctions in tumor position, structure, and proportions. Using computational intelligence and statistical image processing techniques, proposed in our project provide multiple approaches to detect brain cancer and tumors. Also shows an evaluation matrix for a specific system using particular systems and type of dataset. Also explains the morphology of brain tumors, accessible data sets, augmentation techniques, component extraction, and classification of Deep Learning (DL), Transfer Learning (TL), and Machine Learning (ML) models. Key Words: brain tumour, image classification, image sementation, convolutional neural networks.

Quantitative study of solid-state mechanical mixing by high-pressure torsion based on image processing

Mechanical alloying through the process of severe plastic deformation can significantly enhance the mechanical properties of dissimilar metals. However, uneven mixing of these materials can adversely affect their mechanical performance and may even lead to material failure. To quantitatively assess the degree of mixing between dissimilar materials, a mechanical mixing efficiency index (η) has been introduced. This index is calculated accurately by combining an improved Non-dominated Sorting Genetic Algorithm II (NSGA-II) with computer image processing techniques. The study focuses on the Al-Zn mechanical mixture, establishing a parametric correlation between the mechanical mixing efficiency index and the microstructural evolution of mechanical mixing. The findings indicate that when the η value falls below 0.27865, the material transitions from the mechanical mixing stage to the chemical fusion stage. This research facilitates the controllable preparation of mechanical alloying, providing theoretical support and practical evidence for the process.

Image enhancement-based detection of concrete cracks under turbid water bodies

ABSTRACT Underwater concrete structure crack detection and structural health condition assessment based on image processing is challenging. The complex underwater environment and severe image degradation seriously affect the accuracy of crack detection. To solve these problems, a monocular vision and image-enhanced fractal-based fractal science based on computer vision and image processing techniques are proposed to conduct a non-contact detection study of underwater concrete cracks. This study established a four-level structural health condition to assist in underwater crack measurement and safety assessment. The box-counting method was used as a practical tool to calculate the fractal dimension. Three distances of 0.5, 0.8, and 1.2 m were set to verify the effective distance of the algorithm. The results show that the method proposed in this study can effectively detect cracks in submerged concrete members within 0.6 m and help managers correctly determine the structure's health using the fractal dimension.

IoT Based Traffic Congestion Management and Accident Detection System

This review proposes a traffic congestion management and accident detection system to reduce congestion at junctions and to provide emergency assistance during accidents. The proposed system employs advanced computer vision and image processing techniques like You Only Look Once (YOLO) to monitor and analyze real-time traffic conditions and accidents. The pivotal feature of this system lies in its adaptive decision-making capability, automatically adjusting traffic signal timings based on observed density patterns and updating and reporting about the congestions and accidents for which Internet of Things (IOT) technologies are used. Nowadays many junctions and roads are equipped with surveillance cameras connected to traffic management systems. Therefore, techniques like YOLO can be viable tools for automatic accident detection and traffic control.

Automated construction of multi-leaf walls with natural stones

This paper introduces an innovative strategy for the construction of multi-leaf walls utilizing raw stones, a significant step forward in sustainable building practices. Our approach leverages advanced computational algorithms and image processing techniques to automatically build walls with irregular stones. By focusing on the physical assembly of these materials, the objective is to optimize their use in creating structurally robust walls. The core of our methodology lies in the efficient arrangement of irregular objects based on image-convolution techniques. The proposed packing algorithm is proven to be efficient and versatile in stacking stones without restriction in the shape and size of units. Evaluations of constructed walls based on geometric indices show that the stone layout meets masons’ rules-of-art. This work presents a scalable wall construction solution that could revolutionize the use of natural stones in construction.

Advancements in Roundness Measurement Parts for Industrial Automation Using Internet of Things Architecture-Based Computer Vision and Image Processing Techniques

In the era of Industry 4.0, the digital capture of products has become a critical aspect, which prompts the need for reliable inspection methods. In the current technological landscape, the Internet of Things (IoT) holds significant value, especially for industrial devices that require seamless communication with local and cloud computing servers. This research focuses on the advancements made in roundness measurement techniques for industrial automation by leveraging an IoT architecture, computer vision, and image processing. The interconnectedness enables the efficient collection of feedback information, meeting the demands of closed-loop manufacturing. The accuracy and performance of assemblies heavily rely on the roundness of specific workpiece components. In order to address this problem, automated inspection methods are needed. A new method of computer vision for measuring and inspecting roundness is proposed in this paper. This method uses a non-contact method that takes into account all points on the contours of measured objects, making it more accurate and practical than conventional methods. The system developed by AMMC Laboratory captures Delrin work images and analyzes them using a specially designed 3SMVI system based on Open CV with Python script language. The system can measure and inspect several rounded components in the same part, including external frames and internal holes. It is calibrated to accommodate various units of measurement and has been tested using sample holes within the surface feature of the workpiece. According to the results of both techniques, there is a noticeable difference ranging from 2.9 µm to 11.6 µm. However, the accuracy of the measurements can be enhanced by utilizing a high-resolution camera with proper lighting. The results were compared to those obtained using a computer measurement machine (CMM), with a maximum difference of 8.7%.

Development of Deep Learning Algorithms for Improved Facial Recognition in Security Applications

This research aims to develop artificial intelligence (AI) algorithms in the context of facial recognition with a focus on increasing accuracy in difficult environmental conditions. Although facial recognition technology has made great progress, challenges such as poor lighting, variations in facial expressions, and head rotation are still problems that must be overcome. The research methodology involved collecting a wide dataset covering a wide variety of faces under various environmental conditions. This data is then processed and its features are extracted using computer image processing techniques. Furthermore, several deep neural network architectures, such as Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs), were developed, trained, and evaluated for face recognition tasks. The expected result is the development of an AI algorithm that is able to overcome challenges in facial recognition with higher accuracy than existing methods. In particular, significant improvements in facial recognition accuracy are expected especially under low lighting conditions and variations in facial expressions. This research has a major impact in a variety of security applications, such as border surveillance, building access control, and corporate security. With higher facial recognition accuracy, security risks can be significantly reduced, resulting in safer and more efficient security solutions. In conclusion, this research aims to bring innovation in facial recognition technology through advanced AI approaches, with the potential to improve security in various contexts.

Secure Face and Liveness Detection with Criminal Identification for Security Systems

The advancement of computer vision, machine learning, and image processing techniques has opened new avenues for enhancing security systems. In this research work focuses on developing a robust and secure framework for face and liveness detection with criminal identification, specifically designed for security systems. Machine learning algorithms and image processing techniques are employed for accurate face detection and liveness verification. Advanced facial recognition methods are utilized for criminal identification. The framework incorporates ML technology to ensure data integrity and identification techniques for security system. Experimental evaluations demonstrate the system's effectiveness in detecting faces, verifying liveness, and identifying potential criminals. The proposed framework has the potential to enhance security systems, providing reliable and secure face and liveness detection for improved safety and security.
 The accuracy of the algorithm is 94.30 percent. The accuracy of the model is satisfactory even after the results are acquired by combining our rules inwritten by humans with conventional machine learning classification algorithms. Still, there is scope for improving and accurately classifying the attack precisely.

A Study on Intelligent Traffic Monitoring System

Urban traffic has become increasingly congested in recent years due to the prevalence of personal vehicles. As a result, one of the major issues in major cities around the world is traffic. Congestion, accidents that take a lot of time and result in property damage, and environmental pollution are some of the traffic- related issues. In this research paper, an automated and intelligent traffic control system that makes use of computer vision and image processing techniques is designed and put into operation. The algorithm counts the number of vehicles on each route and provides an optimal waiting period based on the number of vehicles loaded onto each road. A dynamically managed traffic system can take the place of the traditional fixed-time, present traffic system using this completely automated approach. Low-cost image processing and traffic load balancing are the main topics of this study.

ED algorithm of inscription picture combining fuzzy logic rules

The expansive growth of information on the Internet has led to new developments in computer vision technology and image processing techniques. Since stone inscriptions are subject to erosion and polishing by the external environment for years, it is difficult to extract image and text information. In this study, the fuzzy control theory is combined with edge detection technology for image edge detection. Firstly, a suitable fuzzy rule and affiliation function are set, then a fuzzy control system is used to extract and detect the image edge information, and then a fuzzy logic rule-based edge detection algorithm is proposed to detect the inscription images. To test the performance of the algorithm, the detection effect of the image is first analyzed from a subjective perspective. The experimental results show that the proposed algorithm has better edge detection for both inscription and lena images, with better noise suppression without excessive distortion, and clearer inscription images. The proposed algorithm has the lowest MSE value of 41.26 when the detection object is the lena image b, and the highest PSNR value of 33.84 when the detection object is the lena image a. The proposed algorithm has the lowest MSE value of 41.26 when the detection object is the lena image b, and the highest PSNR value of 41.26 when the detection object is the lena image b. The proposed algorithm has the highest PSNR value of 33.84 when the detection object is the lena image b. In summary, the analysis of both subjective and objective indicators shows that the inscription image processing algorithm used in this paper has better processing effect, and the processed images become clearer with less distortion, which is helpful for both inscription image and text extraction.

Analysis of Colored Pottery Decoration using Hidden Markov Model Directional Clustering Classification with Deep Learning

Colored pottery decoration is an important cultural artifact that carries significant imagery, symbols, and cultural connotations. This paper presented an in-depth analysis of colored pottery decoration by employing a novel approach, Hidden Markov Model Directional Clustering Classification (HMMDCC), combined with deep learning techniques. The evaluated data comprehensive dataset of colored pottery designs, representing different historical periods and cultural contexts. The imagery, symbols, and cultural connotations embedded in the designs are extracted through a combination of computer vision and image processing techniques. The HMMDCC model is then utilized to perform directional clustering, which identifies spatial relationships and patterns within the decoration elements. To enhance classification accuracy and capture intricate patterns, deep learning techniques are incorporated into the HMMDCC model. The deep learning model is trained on the dataset, enabling it to recognize and classify the imagery, symbols, and cultural connotations present in colored pottery decoration. The findings of this study shed light on the hidden meanings and cultural significance associated with colored pottery decoration. The application of the HMMDCC model with deep learning showcases its effectiveness in analyzing and interpreting complex visual data. The results contribute to a deeper understanding of the historical and cultural contexts in which colored pottery decoration emerged, providing valuable insights for archaeologists, historians, and art enthusiasts.

Feature extraction and fusion using deep convolutional neural networks for Covid-19 detection using CT and X-RAY images

COVID-19 represents a novel variant of the coronavirus disease, having rapidly disseminated across the globe. In recent studies pertaining to computer vision, image processing, and classification techniques, numerous methodologies have been introduced employing chest X-ray images and computerized tomography (CT) images. This investigation introduces novel fused deep features founded on CT and X-ray image classification, with the aim of detecting COVID-19 disease. The proposed approach encompasses three key phases: deep feature generation, iterative feature selection, and classification. During the feature generation phase, well-known pre-trained deep convolutional neural networks, namely DenseNet201, MobileNetV2, ResNet18, ResNet50, ResNet101, VGG16, and VGG19, were leveraged. Each network model generated 1000 features, which were subsequently fused, culminating in the acquisition of a final feature vector comprising 7000 dimensions. In order to distill the most pertinent information, the ReliefF and iterative maximum relevance minimum redundancy (RFImRMR) feature selection techniques were employed in the generation of the ultimate feature vector. To evaluate the performance of the proposed method, publicly available datasets of CT images and X-ray images were employed. Notably, the suggested deep learning approach attained accuracy rates of 99.33% and 93.10% for COVID-19 detection using CT and X-ray images, respectively. These achieved results serve as compelling evidence substantiating the efficacy of the proposed fused deep features and RFImRMR-based COVID-19 detection.

Optimized Placement of Frost-Measuring Sensors in Heat Exchangers via Image Processing of Frost Formation Pattern.

Heat exchangers (HXs) play a critical role in maintaining human thermal comfort and ensuring product safety and quality in various industries. However, the formation of frost on HX surfaces during cooling operations can significantly impact their performance and energy efficiency. Traditional defrosting methods primarily rely on time-based control of heaters or HX operation, overlooking the actual frost formation pattern across the surface. This pattern is influenced by ambient air conditions (humidity and temperature) and surface temperature variations. To address this issue, frost formation sensors can be strategically placed within the HX. However, the non-uniform frost pattern poses challenges in sensor placement. This study proposes an optimized sensor placement approach using computer vision and image processing techniques to analyze the frost formation pattern. Through creating a frost formation map and evaluating various sensor locations, frost detection can be optimized to control defrosting operations with higher accuracy, thereby enhancing the thermal performance and energy efficiency of HXs. The results demonstrate the effectiveness of the proposed method in accurately detecting and monitoring frost formation, providing valuable insights for sensor placement optimization. This approach presents significant potential in enhancing the overall performance and sustainability of the operation of HXs.

Convolutional Neural Networks for Defect Detection on LV cables

A convolutional neural network (CNN) is a machine learning algorithm that is particularly well-suited for tasks such as object recognition, image captioning, and speech recognition. CNNs are particularly effective at detecting features in images that are not easily observable by other machine learning algorithms, such as defects in manufacturing. By analyzing large collections of images, CNNs are able to find patterns that are indicative of defects. Power cables are an important part of manufacturing, as they allow machines to be operated and communicated with. Recently, great importance has been offered in making electricity generation, transmissions, distribution and storage smart. However, the shift to smart grid should include intelligent methods of detecting the reliability of electrical connections. Several types of electrical cables are used to transmission and distribution of electrical energy. Due to excellent electrical and mechanical properties, cross-linked polyethylene (XLPE) cables are widely used in power systems. Poor manufacturing techniques in the production and installation of cable joints will cause insulation defects. Some suggest, the use of interdigital capacitive (IDC) for online monitoring on XLPE cables. Others suggest The use of a continuous wave (CW) terahertz (THz) imaging technology could help display and detect interior faults in cross-linked polyethylene (XLPE) plates used for power line insulation. In this paper, I developed models which predominantly use locally collected custom dataset to forecast individual power cable physical safety status. The model is aimed at replacing the physical inspection with computer vision and image processing techniques to classify defective power cable from non-defective ones. The project is implemented using the Python programming language, the Tensorflow library, and a Convolutional neural network.The Convolutional Neural Network (CNN)-based method is purposefully chosen and applied in this project for power cable defect classification. The project culminates by recommending the use of same or additional datasets and provide modalities to detect power cable defect from live video.

Hand Gesture Recognition for Disabled People Using Bayesian Optimization with Transfer Learning

Sign language recognition can be treated as one of the efficient solutions for disabled people to communicate with others. It helps them to convey the required data by the use of sign language with no issues. The latest developments in computer vision and image processing techniques can be accurately utilized for the sign recognition process by disabled people. American Sign Language (ASL) detection was challenging because of the enhancing intraclass similarity and higher complexity. This article develops a new Bayesian Optimization with Deep Learning-Driven Hand Gesture Recognition Based Sign Language Communication (BODL-HGRSLC) for Disabled People. The BODL-HGRSLC technique aims to recognize the hand gestures for disabled people’s communication. The presented BODL-HGRSLC technique integrates the concepts of computer vision (CV) and DL models. In the presented BODL-HGRSLC technique, a deep convolutional neural network-based residual network (ResNet) model is applied for feature extraction. Besides, the presented BODL-HGRSLC model uses Bayesian optimization for the hyperparameter tuning process. At last, a bidirectional gated recurrent unit (BiGRU) model is exploited for the HGR procedure. A wide range of experiments was conducted to demonstrate the enhanced performance of the presented BODL-HGRSLC model. The comprehensive comparison study reported the improvements of the BODL-HGRSLC model over other DL models with maximum accuracy of 99.75%.

Brain Tumor Detection and Classification Using Intelligence Techniques: An Overview

A tumor is carried on by rapid and uncontrolled cell growth in the brain. If it is not treated in the initial phases, it could prove fatal. Despite numerous significant efforts and encouraging outcomes, accurate segmentation and classification continue to be a challenge. Detection of brain tumors is significantly complicated by the distinctions in tumor position, structure, and proportions. The main disinterest of this study stays to offer investigators, comprehensive literature on Magnetic Resonance (MR) imaging’s ability to identify brain tumors. Using computational intelligence and statistical image processing techniques, this research paper proposed several ways to detect brain cancer and tumors. This study also shows an assessment matrix for a specific system using particular systems and dataset types. This paper also explains the morphology of brain tumors, accessible data sets, augmentation methods, component extraction, and categorization among Deep Learning (DL), Transfer Learning (TL), and Machine Learning (ML) models. Finally, our study compiles all relevant material for the identification of understanding tumors, including their benefits, drawbacks, advancements, and upcoming trends.

Computer vision system for research in the area of defectoscopy for materials and products

In many cases, visual and optical methods can be used in defectoscopy for different materials and products. With the development of microprocessor components and significant expansion of usage of computer technologies and image processing and analysis techniques in different areas, the use of visual and optical methods in defectoscopy for production and research purposes is rapidly developing. In this paper, the author proposes a computer vision system for experiments and research in the area of studying defects of materials and products. The system uses modern methods of image processing and object identification based on their images. The system allows to install the object so that it can be rotated horizontally, take high-quality images of the object using a digital video camera, pre- process images to enhance image quality using a local computing module, transfer images to the main computing module to identify defects and make decisions about rejection of the material or product. To install and rotate the material or product, the author uses the stepper motor 17HS4401 and a horizontal platform fixed on the vertical axis. The stepper motor is controlled using Microstep Driver TB6600 and a local computing module based on a microcontroller with an ARM Cortex-M7 core. The video stream is recorded using a USB microscope video camera which provides sufficiently high image resolution allowing to find defects on the object surface of size 50 micron and larger. Rotation speed can be controlled using a local computing module. The input data for the local computing module can be provided in the form of a video stream or a sequence of images. The local computing module has an LCD screen based on the ВС1602А indicator, programmable LEDs, a keyboard to select operating modes for the stepper motor, a USB port to connect the microscope video camera and an SWD port to program the Flash memory and debug the firmware in real time. Original images or the images after quality enhancement are passed to the main computing module using the SPI interface. The author has developed software for the local computing module to control the stepper motor, record a video stream or series of images of the object area with possible defects, quality enhancement and passing the video stream or images to the main computing module for further processing and analysis. The results can be used in scientific research and in development of automated systems for non-destructive defectoscopy for materials and end products.