Employs Image Processing Techniques Research Articles

Diabetic Retinopathy Severity Level Detection Using Convolution Neural Network

Diabetic retinopathy is a common complication of diabetes mellitus, leading to damage and blockage of retinal blood vessels. Early and accurate detection of diabetic retinopathy severity levels is crucial for timely treatment and prevention of blindness. Diagnostic methods rely on manual examination and human interpretation, resulting in slower and less efficient treatment processes. As a branch of artificial intelligence, computer vision offers a potential solution to analyze retinal images quickly and accurately. The developed system employs image processing techniques and a CNN-based classification model to detect and classify the severity levels of diabetic retinopathy. By providing an automated and efficient approach, the system aims to assist doctors and optometrists in making informed decisions and reducing subjectivity in diagnosis. Early detection through this system can facilitate prompt treatment and improve patient outcomes. The developed system achieves promising results through experimentation and testing with various datasets, with accuracy ranging from 80% to 97%. This project's integration of artificial intelligence, machine learning, and image processing technologies demonstrates their potential in healthcare applications, particularly in diabetic retinopathy diagnosis.

Development of a real-time cattle lameness detection system using a single side-view camera

Recent advancements in machine learning and deep learning have revolutionized various computer vision applications, including object detection, tracking, and classification. This research investigates the application of deep learning for cattle lameness detection in dairy farming. Our study employs image processing techniques and deep learning methods for cattle detection, tracking, and lameness classification. We utilize two powerful object detection algorithms: Mask-RCNN from Detectron2 and the popular YOLOv8. Their performance is compared to identify the most effective approach for this application. Bounding boxes are drawn around detected cattle to assign unique local IDs, enabling individual tracking and isolation throughout the video sequence. Additionally, mask regions generated by the chosen detection algorithm provide valuable data for feature extraction, which is crucial for subsequent lameness classification. The extracted cattle mask region values serve as the basis for feature extraction, capturing relevant information indicative of lameness. These features, combined with the local IDs assigned during tracking, are used to compute a lameness score for each cattle. We explore the efficacy of various established machine learning algorithms, such as Support Vector Machines (SVM), AdaBoost and so on, in analyzing the extracted lameness features. Evaluation of the proposed system was conducted across three key domains: detection, tracking, and lameness classification. Notably, the detection module employing Detectron2 achieved an impressive accuracy of 98.98%. Similarly, the tracking module attained a high accuracy of 99.50%. In lameness classification, AdaBoost emerged as the most effective algorithm, yielding the highest overall average accuracy (77.9%). Other established machine learning algorithms, including Decision Trees (DT), Support Vector Machines (SVM), and Random Forests, also demonstrated promising performance (DT: 75.32%, SVM: 75.20%, Random Forest: 74.9%). The presented approach demonstrates the successful implementation for cattle lameness detection. The proposed system has the potential to revolutionize dairy farm management by enabling early lameness detection and facilitating effective monitoring of cattle health. Our findings contribute valuable insights into the application of advanced computer vision methods for livestock health management.

Validating Synthetic Data for Perception in Autonomous Airport Navigation Tasks

Autonomous navigation within airport environments presents significant challenges, mostly due to the scarcity of accessible and labeled data for training autonomous systems. This study introduces an innovative approach to assess the performance of vision-based models trained on synthetic datasets, with the goal of determining whether simulated data can train and validate navigation operations in complex airport environments. The methodology includes a comparative analysis employing image processing techniques and object detection algorithms. A comparative analysis of two different datasets was conducted: a synthetic dataset that mirrors real airport scenarios, generated using the Microsoft Flight Simulator 2020®video game, and a real-world dataset. The results indicate that models trained on a combination of both real and synthetic images perform much better in terms of adaptability and accuracy compared to those trained only on one type of dataset. This analysis makes a significant contribution to the field of autonomous airport navigation and offers a cost-effective and practical solution to overcome the challenges of dataset acquisition and algorithm validation. It is thus believed that this study lays the groundwork for future advancements in the field.

Enhancing lung lesion localization in CT-scans: a novel approach using FE_CXY and statistical analysis

Intelligence algorithm systems rely on a large dataset to effectively extract significant features that can recognize patterns for classification purposes and extensively utilized to assist the physicians in diagnosis of lung cancer. Extracting valuable features from the available dataset is crucial, especially in cases where additional real data may not be readily accessible. In this context, we propose a novel method called feature extraction based on centroid (FE_CXY) for lesion localization, utilizing a statistical approach. The approach begins with a segmentation process that employs image processing techniques to extract features of interest which is data centroid. This extracted data is then used to compute statistical measurements, revealing hidden patterns that contribute to distinguishing between lesion and non-lesion locations. The method’s efficiency is reflected in the development of robust models with improved performance in localizing lung lesions. The study’s statistical findings strongly indicate that FE_CXY plays a crucial role as an important feature for detecting lesion localization supported by a student’s t-test, which identifies a statistically significant difference in the patterns between lesion and non-lesion localization (p<0.05). By incorporating this method into lung cancer detection systems, we anticipate improved accuracy and efficacy, thereby benefiting early diagnosis and treatment planning.

Gesture Identification Model In Traditional Indian Performing Arts By Employing Image Processing Techniques

Classical dance forms are an integral part of the Indian culture and heritage. Kathakali is an Indian classical dance composed of complex hand gestures, body moments, facial expressions and background music. Kathakali mudras are difficult to understand common peoples. There are 24 classes of hand gestures. The images of hand mudras of kathakali dance are collected from the dataset. The proposed work explores the possibilities of recognizing classical dance mudras in kathakali dance forms in india. This system has achieved an accuracy of 84% with Convolutional Neural Network for classifying the mudras.

Hybrid Convolutional Neural Network with Whale Optimization Algorithm (HCNNWO) Based Plant Leaf Diseases Detection

Plant diseases appear to be posing a serious danger to the production and availability of food globally. The main factor affecting the quality and productivity of agricultural products is the health of the plants. In this paper, we describe a modified plant disease detection using deep convolutional neural networks in real time. By employing image processing techniques to enlarge the plant illness photos, the plant disease sets of data were initially produced. To recognise plant illnesses, a system called Convolutional Neural Network combined with Wolf Optimisation algorithm (CNN-WO) was used. Finally, the Whale Optimization algorithm (WO) is used to maximise and optimizes getting input. And it is given to CNN's learning rate for classification process. This paper presents an image segmentation and classification technique to automatically identify plant leaf diseases. The suggested strategy increased accuracy, sensitivity, precision, F1 measure, and specificity of plant disease detection. According to this study, HCNNWO real detectors have improved, which would require deep learning. It would be an effective method for determining plant illnesses and other diseases within plants. According to the evaluation report, the suggested method offers good reliability. To evaluate how well the suggested algorithm performs in comparison to cutting-edge techniques such as SVM, BPNN and CNN, experiments are conducted on datasets that are openly accessible.

Advancement in Integrated Crop Management System for Sustainable Agriculture

The Crop Management System (CMS) is an innovative web application designed to revolutionize agricultural practices by integrating advanced technologies. This project encompasses four essential modules: Crop Prediction, Disease Detection, Marketing, and Government Forum Dashboard. Leveraging the power of Python, the CMS aims to provide a comprehensive solution for modern farming. The Crop Prediction module employs machine learning algorithms to forecast optimal crops based on factors such as soil type, climate, and historical data. This feature empowers farmers to make informed decisions, enhancing crop yield and profitability. The Disease Detection module employs image processing techniques to identify and diagnose diseases affecting crops, allowing for timely intervention and reducing yield loss. The Marketing module serves as a platform for farmers to streamline the selling process. Through features like price tracking and market trends analysis, farmers can optimize their sales strategies. The Government Forum Dashboard acts as a central hub for stakeholders to exchange information, policies, and best practices, fostering a collaborative ecosystem. Implemented as a web application, the CMS ensures accessibility across devices, providing a user-friendly interface for farmers and stakeholders. The backend is built using Python, leveraging its versatility and robust libraries for data processing, machine learning, and web development. In conclusion, the Crop Management System addresses critical aspects of modern agriculture, ranging from crop selection to disease management, marketing, and policy advocacy. By harnessing the power of Python and cutting-edge technologies, this project stands as a pivotal tool for advancing agricultural practices, ultimately contributing to sustainable and efficient farming practices

Hybrid features extraction for the online mineral grades determination in the flotation froth using Deep Learning

The control of the froth flotation process in the mineral industry is a challenging task due to its multiple impacting parameters. Accurate and convenient examination of the concentrate grade is a crucial step in realizing effective and real-time control of the flotation process. The goal of this study is to employ image processing techniques and CNN-based features extraction combined with machine learning and deep learning to predict the elemental composition of minerals in the flotation froth. A real world dataset has been collected and preprocessed from a differential flotation circuit at the industrial flotation site based in Guemassa, Morocco. Using image-processing algorithms, the extracted features from the flotation froth include: the texture, the bubble size, the velocity and the color distribution. To predict the mineral concentrate grades, our study includes several supervised machine learning algorithms (ML), artificial neural networks (ANN) and convolutional neural networks (CNN). The industrial experimental evaluations revealed relevant performances with an accuracy up to 0.94. Furthermore, our proposed Hybrid method was evaluated in a real flotation process for the Zn, Pb, Fe and Cu concentrate grades, with an error of precision lesser than 4.53. These results demonstrate the significant potential of our proposed online analyzer as an artificial intelligence application in the field of complex polymetallic flotation circuits (Pb, Fe, Cu, Zn).

Investigating and Classifying the Level of Thermal Runaway-Induced Failure Energy Storage Systems in Photovoltaics

In a photovoltaic system, the electricity generated by a solar panel can depend on the amount of sunlight available at any given time. Due to the nonlinear behavior of this energy, the storage component of the system is critical. As a result, many solar panel systems are equipped with batteries. However, changes in environmental and other conditions can cause damage to these batteries. This study explores novel technologies for detecting thermal runaway failures in lithium-ion batteries. Specifically, the study employs image processing techniques to detect structural failures and applies deep learning techniques for automatic classification. Thermal damage to a battery can result in irreparable harm, making design and construction considerations crucial. Through image analysis, any internal changes in the battery can be transformed into a measurable variable, providing a reliable indication of potential failure. The study compares the current series with previous ones to highlight the structural differences. Results confirm that the proposed approach has significant potential for detecting and estimating internal variations during production. Overall, the proposed method can serve as a valuable tool for drafting and implementing a comprehensive plan to address early problems in the battery.

Classification of Cotton Leaf Diseases using Whales Optimization Algorithm based Deep Neural Network

In the realm of agriculture, a current focal point of research revolves around the identification of plant diseases through leaf imagery. Employing image processing techniques for the recognition of agricultural plant diseases holds promise for reducing farmers' reliance on safeguarding their crops. This paper introduces a novel approach to classify cotton leaf diseases, utilizing a Deep Neural Network enhanced by the Whales Optimization Algorithm (WOA). The dataset comprises 10,000 cotton images sourced from Kaggle.com, directly captured from farm fields, covering healthy leaves, bacterial blight, Anthracnose, Cercospora leaf spot, and Alternaria diseases. Preprocessing involves the application of a median filter to eliminate image noise, and for segmenting diseased and normal regions, the Gustaffson-Kessel (G-K) fuzzy clustering method is employed. The WOA-augmented DNN demonstrates its effectiveness in classifying cotton images.

An Investigation on Filler Particle Size and Its Distribution in Glass-Epoxy Hybrid Composites Using Image Processing

Analysis of filler particle size and distribution plays a vital role in achieving a good balance of polymer composites’ electrical, thermal, and mechanical properties. Such assessments help understand the relationship between microstructure and the physical nature of polymer composites reinforced with fillers. In this investigation, glass epoxy composites with different nano and micro-sized fillers, namely SiO <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> , Al <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> O <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sub> , MoS <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> , graphite and cenosphere, are fabricated by hand-lay technique with bagging. Particle analysis is carried out by employing image processing techniques using ImageJ. To understand the nature of hybrid filler size distribution, a comprehensive assessment of the micro and nanofiller particle distribution and the average shortest and longest separation distance between the nano and micron particles is estimated. From the particle size distribution analysis, it is observed that the micron particles are more evenly distributed than nanoparticles. The average distance between micron and nanofiller particles is observed depending on the type of filler used. The microparticles tend to surround the nanoparticles, which is desirable for decreasing the free space distances between the filler particles. Also, statistical analysis is carried out to study the experimental and estimated micron and nanofiller particle sizes used in the composites.

Driver Drowsiness Detection System Using Machine Learning

Abstract: Drowsy driving is one of the major causes of road accidents and death. Hence, detection of driver's fatigue and its indication is an active research area. Most of the conventional methods are either vehicle based, or behavioral based or physiological based. Few methods are intrusive and distract the driver, some require expensive sensors and data handling. Therefore, in my literature survey, a low cost, real time driver's drowsiness detection system is developed with acceptable accuracy. The proposed work mainly focus on a webcam records the video and driver's face is detected in each frame employing image processing techniques. Facial landmarks on the detected face are pointed and subsequently the eye aspect ratio is computed and depending on their values, drowsiness is detected based on developed adaptive thresholding. Machine learning algorithms have been implements as well in an offline manner

Machine learning-aided scenario-based seismic drift measurement for RC moment frames using visual features of surface damage

This paper presents a novel computer vision-based methodology for assessment of the seismic damage in reinforced concrete moment frames using visual characteristics of surface damage following an earthquake. An extensive collected database comprising 974 images associated with 256 cyclic-loaded damaged beam-column joints, providing a set of cracking and crushing progression with increasing the evolution of damage level, is collected and used for the development and validation of the methodology. Employing image processing techniques, the characteristics of the surface damage, including the cracking length and crushing areas, are measured and used in a scenario-based assessment for the seismic peak drift prediction. Based on the availability of the structural information, four scenarios are proposed using input parameters among cracking length, crushing areas, concrete compressive force, and the aspect ratio of the joint. The machine learning regression method is employed for developing nonlinear regression models for each scenario. The proposed models measure the seismic peak drift ratio during the earthquake excitation based on the visual damage features at the external surface of the components. Finally, the seismic peak drift ratio obtained by the proposed methodology of this paper can be used as an input engineering demand parameter in the existing seismic loss measurement frameworks. An example specimen at various drift ratios is also presented as a case study to evaluate the predicted versus actual experimental drift ratio.

DRIVER DROWSINESS DETECTION USING MACHINE LEARNING

Drowsy driving is one of the major causes of road accidents and death. Drivers must keep a close eye on the road, so they can react to sudden events immediately. Driver fatigue often becomes a direct cause of many traffic accidents. Therefore, there is a need to develop the systems that will detect and notify the driver bad psychophysical condition, which could significantly reduce the number of fatigue related car accidents. However, the Development of such systems encounters many difficulties related to fast and proper recognition of a driver’s fatigue symptoms.. Most of the conventional methods are either vehicle based, or behavioral based or physiological based. Few methods are intrusive and distract the driver, some require expensive sensors and data handling. Therefore, in this study, a low cost, real time driver’s drowsiness detection system is developed with acceptable accuracy. One of the technical possibilities to implement driver drowsiness detection systems is to use the vision-based approach. Here we are detecting the driver drowsiness by estimating vision system of him/her. We will be using Open CV for gathering the images from webcam and feed them into a Deep Learning model which will classify whether the person’s eyes are ‘Open’ or ‘Closed’. Webcam records the video and driver’s face is detected in each frame employing image processing techniques. Facial landmarks on the detected face are pointed and subsequently the eye aspect ratio, mouth opening ratio and nose length ratio are computed and depending on their values, drowsiness is detected based on developed adaptive thresholding. Machine learning algorithms have been implemented as well in an offline manner.

Integrated 3D Building Model and Thermography Analysis to NTD Virtual Reality

A thermal infrared camera can be used to restore the 3D model such that the restored model contains the surface temperature information. The surface temperature information can be employed to identify the defects illustrated in the model. However, the 3D model is limited to the resolutions of the thermal infrared camera. Furthermore, the surface temperature information is supposed to be analyzed by employing image processing techniques to develop the temperature boundaries for defect detection. The study proposes to restore a 3D model by employing a series of photos collected by a drone. In doing so, a 3D model with high accuracy can be built. Then, the surface temperature information can be recorded with a thermal infrared camera which can provide thermal infrared images with high resolutions. Then, image segmentation is applied to the recorded thermographs by introducing the level set functions in an iteration scheme. The initial level set functions will change their locations and shapes during the iteration scheme till the convergence can be reached. Eventually, the thermal infrared images are segmented, and the surface temperature boundaries can be located. The restored 3D model from the drone and the segmented results are integrated by attaching the segmented results to the 3D model. In doing so, a 3D model with high accuracy and the segmented results can be integrated such that the defects can be identified on the model.

Cancer Detection using Deep Learning: A Survey

Cancer refers to a group of diseases that are defined by abnormal cell proliferation and can spread to other parts of the body. Pancreatic cancer originates in the organ behind the lower stomach called the pancreas. The pancreas secretes digestive enzymes as well as hormones that help regulate sugar metabolism. Pancreas cancer is usually discovered late, spreads quickly, and has a poor prognosis. Cancer-specific symptoms may not appear until later stages, and there are no reliable screening techniques to identify high-risk patients. The manual method of checking whether a patient has Pancreatic cancer or not can be time-consuming and it can be tedious because it required experience in Radiologist interpretation; to avoid poor decisions by giving wrong diagnoses to patients. One of the leading causes of cancer death is pancreatic cancer; however, if found early enough, it can be treated. To solve this issue, an Artificial Neural Network (ANN) has been developed, trained, and tested using an abdominal Computed Tomography (CT) scan images dataset to detect pancreatic tumors by many authors. To detect the tumor, it employs image processing techniques of a convolution neural network to detect the tumorous area in the image after it has been pre-processed. It was observed that the technique gives better accuracy in tumor detection.&#x0D;

Driver Drowsiness Detection

Every year many human beings lose their lives because of fatal road injuries round the arena and drowsy driving is one of the number one causes of avenue accidents and demise. Fatigue and micro sleep at the using controls are often the basis reasons of significant injuries. But preliminary signs of fatigue may be detected earlier than a vital scenario arises and therefore, detection of driving force’s fatigue and its indication is an ongoing research subject matter. Most of the conventional strategies to hit upon drowsiness are based on behavioral aspects while some are intrusive and might distract drivers, at the same time as some require costly sensors. Therefore, the development of a lightweight as well as affordable driver’s drowsiness detection system has become necessary. The major aim of this project is to develop a drowsiness detection system by monitoring face features. It is believed that the symptoms of driver fatigue can be detected early enough to avoid a car accident. In such a case when drowsiness is detected, a warning signal is issued to alert the driver. This detection system provides a noncontact technique for judging different levels of driver alertness. The implemented system records the video and detects the driver’s face in every frame by employing image processing techniques and sends a warning message to the driver about his/her improper driving and raises the alarm (in case of drowsiness). Machine learning algorithms (like SVM and YOLOV5) have been employed to test the efficacy of the carried-out technique. Due to the COVID pandemic, usage of the mask is mandatory. So, the application also includes face mask detection which means this application can keep track of whether the driver is wearing a mask or not. Driver’s drowsiness can be detected using different gestures and postures such as yawning. So, the whole system is comprised to detect driver’s drowsiness as well as mask usage.

Computed tomography scans image processing for nasal symptoms severity prediction

This paper aims to use a new technique of computed tomography (CT) scan image processing to correlate the image analysis with sinonasal symptoms. A retrospective cross-sectional study is conducted by analyzing the digital records of 50 patients who attended the ear, nose and throat (ENT) clinics at King Abdullah University Hospital, Jordan. The coronal plane CT scans are analyzed using our developed software. The purposes of this software are to calculate the surface area of the nasal passage at three different levels visible on coronal plane CT scans: i) the head of the inferior turbinate, ii) the head of the middle turbinate, and iii) the tail of the inferior turbinate. We employ image processing techniques to correlate the narrowing of nasal surface area with sinonasal symptoms. As a consequence, obstruction in the first level is correlated significantly with the symptoms of nasal obstruction while the narrowing in the second level is related to frontal headache. No other significant correlations are found with nasal symptoms at the third level. In our study, we find that image processing techniques can be very useful to predict the severity of common nasal symptoms and they can be used to suggest treatment and to follow up on the case progression.

Multipath Routing in Elastic Optical Networks with Space-Division Multiplexing

This article introduces four multipath routing, core, and spectrum assignment algorithms, all of which employ image processing techniques for fast identification of allocable slots, since this is less computationally complex and adequate for real-time operation. Numerical results show that multipath routing can decrease the probability of blocking by one order of magnitude in relation to its single-path counterpart, but produce acceptable differential delay values.

Canker Detection in Citrus Plants with an Efficient Finite Dissimilar Compatible Histogram Leveling Based Image Pre-processing and SVM Classifier

The massive loss in agricultural yield mainly caused by leaf and fruit diseases. These diseases reduce both quality and quantity of agricultural products. As a major provenience of nutrients like vitamin C, citrus plants such as lemons, mandarins, oranges, tangerines, grapefruits, and limes are commonly grown fruits all over the world. As a result of various plant diseases, citrus producing companies make a huge amount of waste every year whereby 50% of citrus peel is destroyed every year. The disease canker is one of the mentionable leaf and fruit diseases. The main goal of this paper is to recognize and classify the canker disease precisely from the contrived leaf images by employing image processing techniques to detect plant leaf diseases from digital images. We offer a method consists of two phases to enhance the clarity of leaf images. The primary stage uses Finite Dissimilar Compatible Histogram Leveling (FDCHL) in preliminary step which advances the dissimilar level of disease influenced leaf image, segment the region of interest using fuzzy feature selection. The second phase by adopting the Support Vector Machine classifier to find out the canker leaf image and implements these methods in lemon citrus canker disease identification. Experimental results show effective accuracy detection and reduced execution time of canker disease detection.