Search

Satellite images mainly utilized in the events of a natural disaster management, identifying geographical information, viz land cover classes namely, buildings, roads, vegetation, water, agriculture land, crop types, plants, bare ground, cities, atmosphere conditions. Machine Learning (ML) approaches have been utilized effectively to develop a model for classification, detection, and segmentation tasks. Therefore, Satellite image processing and analysis purpose, ML techniques plays vital role and remotely sensed data become essential while training the model. The aim of this study is to investigate the various of ML techniques in satellite image analysis. However, to predict the various events in advance across the globe, it is necessary to focus more on remote sensed data and data processing techniques for accurate classification. Even though remote sensing quality has been increased and artificial intelligence solutions are equally increased. This paper addressed various types of advanced ML techniques utilized in the classification and assessment of satellite images and used to track the earthquakes, faulting, landslides, floodings, wildfire, and hazards associated with the stated activities. Still there is a gap and interference in the approaches and it is important to fill the gap by thorough review of recent classification approaches. In this connection it is necessary to look in depth to the state-of-the-art ML techniques of satellite image processing. [ABSTRACT FROM AUTHOR]

Food production and distribution in the food industry require efficient organization. However, manual grading of agricultural produce faces challenges of subjectivity, inefficiency, and inconsistency. To address these issues, this paper proposes a food identification algorithm that utilizes image processing for agricultural technology to identify the type and grades of food products. The methodology involved capturing images of food produced using an ESP32-CAM camera module connected to an Arduino MEGA 2560 board. The images were processed through algorithms to identify the food produced and grade them based on size. The color of the image was extracted by calculating the Euclidean distance of RGB values in the image. The type of food product was identified using the perimeter and the Euclidean distance of RGB values. Then, the algorithm extracted information about the area from the image and provided the value of the area, which determined the grade of the product. The results and discussions present the test outcomes of the food identification and grading algorithms. The food identification algorithm successfully recognized tomatoes and mangoes from the captured images, achieving high accuracy. The food grading algorithm effectively categorized the produce based on the food product's area into Grade A (Big Food Product) if the food product's area is more than 10,000 pixels, and Grade B (Small Food Product) if it is less than 10,000 pixels. In conclusion, the proposed image processing-based food identification and grading system demonstrated potential in automating the food grading process. It provided reliable results in identifying and categorizing an agricultural produce based on visual attributes. [ABSTRACT FROM AUTHOR]

Drowsiness and Fatigue of motorists and drivers cause more than 30% of road accidents. Every time, they increase the quantities of deaths, losses, and injuries encyclopedically. Discovery of drowsiness is needed so that it cautions the motorist and the drivers which might save their lives. It's being done using the CNN algorithm in image processing. In this discovery fashion, a motorist is continuously monitored through a webcam. In this technique, the driver's face is taken and it is fed to Deep Learning so that the blinking of the eye is predicted. We use an algorithm, if a motorist's eye seems to be closed for a particular quantum of time, the system cautions the motorist with a sound. To perform the segmentation of the images, ways similar as Image recognition and Image Processing. [ABSTRACT FROM AUTHOR]

To gauge head growth and identify foetal anomalies, it is crucial to examine head shape during the foetal period. The most popular imaging technique for carrying out this assessment is prenatal ultrasonography. On the other hand, manual interpretation of these images is difficult, hence methods for image processing have been put out in the literature to help with this task. The goal of this essay is to review these cutting-edge techniques. This study aims to evaluate and categorize the various image processing methods applied to ultrasound imaging to evaluate the foetal head. For foetal head examination, numerous image processing techniques have been put forth. In conclusion, methods from several categories demonstrated their potential to enhance clinical practice. However, more study is required to improve the existing techniques, particularly for 3D image acquisition and analysis for detection of abnormality. [ABSTRACT FROM AUTHOR]

The application of image processing and Neural Network for controlling an actuator or robot was developed and evaluated. The main objective of the system is to establish or create a system for the disability to use facial expressions to control an actuator to assist their daily activities which the group of people easily struggling with. By making use and capturing of the open and closed eyes and mouth motion, the data able to "train" Convolution Neural Network (CNN) models and Mouth Aspect Ratio (MAR) value. All the mentioned parameters had been evaluated together with a new proposed technique, which is named as "quantity" method. The result showed that the developed system is able to capture the actual and true expression of an individual accurately. [ABSTRACT FROM AUTHOR]

The presence of atypical lymphocyte cells in the blood smear of a patient suspected of dengue is a specific sign that the person concerned is positive for an attack of dengue fever. In general, the counting of atypic lymphocyte cells is done manually under a microscope, but this method has consequences for counting errors and a longer time. In digital image processing, we can use the improved counting morphology algoritm to count atypic lymphocytes faster and more accurately than manual calculations. The improved counting morphology algorithm uses the concept of adjacency 8 as the basis for calculating atypical lymphocytes by memorizing the neighbor next to it, if the color next to it is the same color as the reference point, it is said that the pixel points are neighbors or cognate with the reference pixel point and are counted as the same object. But on the contrary, if a pixel point is not related to its neighbors, it means that the pixel point is a separator or a different object. The results of this tool have an accuracy of 100%. [ABSTRACT FROM AUTHOR]

Agricultural productivity is problematic while the plant attacks several micro-organisms, viruses, and bacterial infections. Earlier disease identification is a unity regarding the challenging solutions for increasing plant production. The signs from these attacks are generally identified in the leaves, fruit, and stems inspection. Nowadays, the need for computational diagnosis of plant diseases increases to gain efficient plant productivity. Here, the process identifies and analyzes the Citrus leaf and Fruits disease literally from the infected images by adopting image-processing procedures to distinguish leaf infections from digital-images. The suggested hybrid algorithm includes pre-processing and RGB HSI conversion. The initial step practices CLAHE segments the features of affected areas applying K-means clustering and statistical GLCM. The SVM classifier has recognized the diseased image and performs those methods in citrus disease detection. Finally, the Fuzzy-based estimation has been convoluted to measure the disease grade severity. [ABSTRACT FROM AUTHOR]

Most people who get sick around the world have some kind of skin condition. Despite its frequency, it is very challenging to identify and requires substantial domain experience. This study explains how to identify various types of these disorders. We use a two-stage strategy that efficiently integrates image processing and deep learning based on molecular diagnostic histological features to reliably detect the disease. During the initial stage of the process, the image of the skin condition is processed in a variety of different ways before "feature extraction" is performed. In the second stage of this process, Machine Learning (ML) algorithms are used to diagnose diseases based on histological characteristics discovered through skin examination. These characteristics can be found by analysing the skin. After being trained and evaluated, the system could correctly name up to 95% of the six diseases. [ABSTRACT FROM AUTHOR]

Anxiety is an element of existence it is a disagreeable condition of expressive stimulation that public understanding in circumstances similar to operational for extended hours in facade of electronic gadgets. Electronic gadgets like laptops, mobiles and computers have grow to be a approach of existence, to a great extent living is worn-out on the Electronic gadgets moreover consequently we are as a result additional exaggerated through the ups and downs that they reason us. One cannot immediately totally circumvent their toil on Electronic gadgets although one knows how to be as a minimum authority his/her handling while being worried regarding him being stressed out at one position of instance. Scrutinize the expressive condition of a human being who is functioning in frontage of a Electronic gadgets for extended period is fundamental for the wellbeing of a human being. In this toil a instantaneous non-intrusive videocassette records are taken by capturing the community, which notice the expressive condition of a person by investigating the facial appearance. We identify a human being feeling in every recorded video capture structure and the conclusion on the strain intensity is completed in chronological times of the videocassette imprisoned. We utilize a method that permits us to educate a representation and investigate differentiations in forecasting these characteristics. The package Theano is a python programming construction which intends at improving both the completing moment in addition to improvement point of the Linear Regression representation that is utilized in the perspective as a DL algorithm. The investigational outcomes demonstrate that the enlarged organization is well on information with the standard representation of every era. [ABSTRACT FROM AUTHOR]

Glaucoma is the 2d main purpose of everlasting blindness worldwide. Early detection of glaucoma can restrict the disorder development. This paper directs the exceptional picture processing strategies for glaucoma detection based totally on CDR. After calculating the effects are in contrast the usage of sensitivity and specificity. Glaucoma is a criticism prompted due to loss of feature of neurons of the eye which leads to loss of eye sight. The photo processing strategies are Multi level thresholding approach, model of Active Contour Method, Region Based Segmentation. It can be calculated by using gazing Intra optical pressure (IOP), view, optical slice appearance, Glaucoma influences the supplemental imaginative and prescient loss by using including Mug to disc ratio. [ABSTRACT FROM AUTHOR]

The increased car density has a significant negative impact on the city's quality of life. As a result, parking in densely populated areas with many cars often results in more driving time in search of free space, which can contribute to pollution. Although there are numerous new smart parking systems, the majority of them are designed for close parking. At the moment, no sensors or technologies are used for open parking spaces, such as those found in commercial complexes. Smart parking sensors and technology can help drivers find parking more easily and efficiently. The inefficiency of parking management for open parking is discussed in this paper. From the literature search it is concluded that currently, an image processing technique was used to solve the problem of open parking space is the best solution. This is because many cars are driving around unnecessarily in search of a free parking space. This can be mitigated by providing drivers with information about available parking spaces in an open parking area. The image processing technique is used to guide the driver in determining which area has a vacant parking slot. To determine the number of available parking spaces, several images of open parking areas were captured and analyzed using image processing techniques. The discovery reveals that as the value of the overall parking space increases, so does the value of marker sizes and font size. The equation of a straight line is derived from the graphs to represent the real-life equation of each graph. [ABSTRACT FROM AUTHOR]

An algorithm for detecting active objects from real-time is presented. Security cameras are common devices that especially use motion sensing technology. Nowadays, cameras with motion detectors are preferred for better protection. Motion detection technology is widely used for security purposes. Here with python programming motion detection software is used that detects human, cat and dog movement in front of the camera. Caffe model is used for image processing based motion sensing that are pretrained. The detected movement is notified to the user via the Telegram chatbot along with detected video. Images and videos can also be retrieved in it. [ABSTRACT FROM AUTHOR]

Advancements in deep learning and computer vision have resulted in a surplus of counterfeit facial content that is incredibly realistic in appearance, which is controlled by artificial intelligence., for example, Deep Fake or Face2Face that control facial personalities or on the other hand articulations. Counterfeit appearances made utilizing Generative Adversarial Networks (GANs) are becoming increasingly sophisticated and difficult to distinguish. The synthetic faces were designed primarily for amusement purposes, but their abuse has created emotional harm. For example, a few celebrities have fallen victim to Deep Fake's false erotic entertainment. Concerns are also growing regarding false political debate recordings generated by Face2Face. Building models that can identify fake faces in the media is essential for maintaining personal, societal, political, and international security. The purpose of this paper is to suggest a novel way to identify these counterfeits on the internet using Convolutional Neural Networks (CNN) and image processing techniques. We first used Contrast Limited Adaptive Histogram Equalization (CLAHE) and Complex Shearlet-based edge detection on the dataset and then used the resultant images obtained to train a proposed CNN to categorize fake and real personalities and separate them. [ABSTRACT FROM AUTHOR]

Medical imaging has seen substantial improvements not only in image procurement but also in terms of interpretational methodologies. The practice of using technologies such as image processing and artificial intelligence has shown remarkable results in the detection, classification, and localization of human bone fractures. Breakage or cracks in the bone occur when an external force is applied to a bone that is significantly larger than the bone's tolerance, which may lead to dislocation requiring extensive treatment. However, manual examination by experts is a complex and arduous process with a high chance of false detection. Many approaches under Artificial Intelligence and its subsets are proposed to facilitate fracture diagnosis and prevent misinterpretation, including Deep Learning and Convolutional Neural Networks (CNN). This paper is a comprehensive study of the various techniques and models designed using image processing approaches and the advances in machine learning and Deep Learning that have demonstrated improved performance in the precise detection of fractures in human bones. [ABSTRACT FROM AUTHOR]

Xilinx System Generator (XSG) integrates with MATLAB Simulink to build a complex system through Xilinx and Simulink blocks. This paper presents a hardware implementation of various image processing techniques to modify and extract the pixel information from various input images in run time. Complex coding strategies are eradicated with integrated system. The intricacy in structural design is reduced to a huge extent using Xilinx System generator. The reconfigurable attribute of Field Programmable Gate Array (FPGA) bags the massive advantages in image processing like co simulating with selective pixel and verified on FPGA. Image processing algorithms explored in methodical manner by performing diverse operations like edge detection, and color enhancement, complement, contrast stretching etc. The proposed image processing techniques (IPT) intended for various applications. Dissimilar image processing techniques are processed by FPGA and displayed on Video Graphics Array (VGA). Image processing system is designed in XSG, hardware description language (HDL) like VHDL & Verilog code is generated by processing system by system generator token. Design is implemented on Artix-7 and targeted on Basys3 reconfigurable hardware (XC7A35T-1-CPG236C). [ABSTRACT FROM AUTHOR]

Image processing is the process of manipulating images by visualizing, recognizing, sharpening of image features. Digital image processing deals with enhancing quality of image by using image processing techniques. Real time object tracking methods are studied for effective object identification. The algorithms for single and multiple object detection, color detection and face detection of an image are discussed. This article provides a comparative study of various object detection algorithms for video surveillance. The advantages and disadvantages of the system are compared and contrasted. From the results, it is evident that the object tracking algorithms provide high accuracy and precision by capturing the video frame and processing it frame-by-frame. It is used in various applications such as health care, forensics and security systems. [ABSTRACT FROM AUTHOR]

Embedded applications have begun to reap the benefits of embedded vision in recent years. The proliferation of high-performance computers, the maturing of deep learning algorithms, and vision is being added to electronic devices like integrated medical applications due to image processing advances. However, due to the high amount of CPU resources necessary to process even a single image to recognize an object, it is especially challenging to integrate image processing in embedded systems. Deep learning approaches, when deployed and validated on data sets matched to the specific task at hand, may produce better results. We propose and investigate the use of deep learning to improve the image-processing architecture of embedded vision systems in this study. Deep learning algorithm testing on embedded vision yielded encouraging results. [ABSTRACT FROM AUTHOR]

When the Viola-Jones machine learning algorithm is used to forecast a driver's sleepiness behaviour, it may help save people from nodding off behind the wheel and causing accidents that can leave them seriously injured or dead. To determine the predictive accuracy of the Viola-Jones machine learning system for Blink frequency, the sample size was varied from 100 times. The sensitivity of Viola-Jones is much higher. Viols Jones has a higher blink frequency percentage than HOG (86.14 percent) (75.84 percent). The correlation between Voila-Jones and HOGt was significant (p=0.000). The Viola-Jones model improves the sensitivity % for forecasting driver fatigue. If the algorithm determines that the driver is, in fact, sleepy, an alert will sound. [ABSTRACT FROM AUTHOR]

Positioning within indoor environments is one of the major challenges facing many different applications. It has become necessary to find algorithms and systems capable of locating a specific object within an environment that contains many objects. In this paper, the researchers have made a detailed study on many methods of positioning within different environments. Some methods rely on digital images and others rely on identification systems by radio waves and other methods. Each system has a number of advantages that qualify it to be the best choice in certain applications. On the other hand, there are drawbacks to each system that prevent it from working within certain environments. Clear comparisons are made in this paper showing the different systems with the pros and cons of each one. [ABSTRACT FROM AUTHOR]

Image compression techniques have become a hot topic due to the increasing growth of multimedia applications. This paper presents a ground-breaking full quadtree range block partition-based search-less fractal encoding algorithm that can efficiently encode range blocks or even individual pixels. This method, which may also be applied to high-quality image compression, solves both issues. Experimental results show that the technique outperforms traditional search-based and search-less fractal algorithms at low and high compression rates, thus improving both compression ratio and reconstructed image quality. Another advantage of the approach is its speedy encoding speed because it does not require a search process. When compared to traditional fractal algorithms, this method encrypts data much faster. [ABSTRACT FROM AUTHOR]

Corn consumption in Indonesia can reach around 14.37 million tons and continues to increase every year. This increase is directly proportional to the demand from corn processing industries/factories/warehouses to farmers for the supply of corn ingredients. However, the industry often finds corn with quality that is not suitable for production from farmers/collectors, this causing losses for industries. This problem is very important to overcome so that losses can be reduced and the results of corn processing production can also have good quality. This research proposes an image processing solution for industrial corn quality classification using the Convolutional Neural Network (CNN) method. This research trains the CNN to be able to recognize the quality of corn by utilizing 600 images/images of corn as a dataset consisting of 300 images of good quality corn and 300 images of poor quality corn. The CNN method is then used as an algorithm to recognize good quality corn and which is not. The research results so far show an accuracy rate of 99.1% with 10 epoch training. [ABSTRACT FROM AUTHOR]

This paper consists of the comparison on image processing using Minimal spanning tree and thresholding methods. [ABSTRACT FROM AUTHOR]

Cervical cancer is caused by the growth of abnormal cells in the cervix's lining. Human papillomavirus (HPV), a sexually transmitted infection, plays a role in most cervical cancers. Therefore, cervical cancer can be avoided by having regular screenings and being vaccinated against HPV infection. The term pap-smear refers to human cell samples stained using the Papanicolaou method. The Papanicolaou method is used to detect precancerous cell changes before they become invasive cancer. The smear cell image is composed of a nucleus and cytoplasm. Cancer prevalence is determined by the shape and structure of the nucleus. Therefore, the segmentation of the nucleus is an essential step in detecting cancer. However, overlapping, poor contrast, uneven staining, and other factors make cervical nucleus segmentation difficult. This paper proposes a new segmentation method for the cervical nucleus using digital image processing. Our proposed method used a median filter to remove noise and a non-linear contrast stretching to enhance the Pap smear images. Then, we used Bradley thresholding for the segmented cervical nucleus. The main impact of this paper will assist doctors in diagnosing cervical cancer based on Pap smear images and increase the accuracy percentages compared to the conventional method. [ABSTRACT FROM AUTHOR]

QR codes is the type of matrix bar code. It was invented by Denso Wave in japan.QR code,Due to its quick readability and bigger storage capacity than traditional UPC barcodes. This essay explores the fundamentals of QR codes as well as how they used in daily life. Shopping centres are expanding in size and variety,carrying a bigger variety of goods. To reduce the complexity of shopping frameworks,and various shopping guide mailings, and To make Shopping for customers more convenient,the system has developed.he creation of an easy, quick, and convenient shopping system is a concern for both retailers and customers. Additionally, new approach corrects a number of flaws in the old system. This system generates QR codes for each product using a QR code generator. The test's findings are implemented in a Python software that uses a camera to read QR codes and create a bill for purchases. [ABSTRACT FROM AUTHOR]

To take out precise ROIs and improve images for precise night-time vehicle number plate detection, we combine a brand-new area of interest (ROI) extraction technique based on improved multi-scale retinex with a method for improving night time images that combines object recommendations with vehicle light detection. (MSR). A suggested score-level feature fusion unifies five complementing traits. The approach we provide can find moving objects that are blurry or partially obscured, as well as moving objects of different shapes, sizes, numbers, positions, and backdrops. The use of a licence plate detection (LPD) system is essential in a variety of traffic-related applications. This project aims to create a sophisticated detecting system that excels in challenging situations. It suggests a reliable preprocessing improvement technique for correctly identifying licence plates in challenging vehicle photos. The suggested approach combines a Gaussian filter with a contrast-limited adaptive histogram equalisation methodology and an enhanced cumulative histogram equalisation method, the local binary pattern and the histogram of an orientated gradient. We identify names using the licence plate. Eventually, we see outcomes. Using the voice command, DON'T PARK IN THIS AREA to send an SMS to the number that is saved in the database. This object has a penalty of this exact amount if it is not obeyed. [ABSTRACT FROM AUTHOR]

The purpose of this research is to identify and classify leaf diseases in corn plants. In this research, the Convolutional Neural Network (CNN) method is used to classify and identify objects in real time. This research describes an automated method for detecting corn leaf diseases that is based on three steps: pre-processing, feature extraction, and classification. Experimental findings show that the proposed method can efficiently and accurately detect leaf diseases of corn plants. The test results prove that the classification of corn plant diseases through leaf images using deep learning with the convolutional neural network (CNN) method in the experiment obtained an accuracy rate of 95%. This high level of accuracy is expected to help farmers control the spread of corn diseases as early as possible to minimize greater losses. [ABSTRACT FROM AUTHOR]

Increasing the speed of an analog and mixed mode signal circuit without compromising power consumption is a difficulty in very large scale integrated (VLSI) design. This study work is carried out to construct a 12-bit Pipeline A/D converter (ADC) with 400MS/s sample rate to fulfil the high computational needs. In order to accommodate various applications, the design is concentrated on determining high speed and resolution in pipeline ADC. The pipeline method's key benefits include ease of implementation, increased speed, and ease of layout design. A pipeline ADC architecture is designed using a proposed method that multiplies the D/A converter, comparator, sample and hold circuit (S/H), and operational transconductance amplifier (OTA). A switching capacitor integrator module is used by OTA to convert a differential input voltage into a current. The insertion of a S/H circuit in the pipeline ADC's initial stage eliminates the need for a separate S/H amplifier. The obtained results indicate that the integral and differential non-linear values are, respectively, +0.61/-0.75LSB and +0.48/-0.55LSB. A value of 64.2 dB of Signal to Noise Dynamic Range (SNDR) and 81.8 dB of Spurious Free Dynamic Range (SFDR) is derived through simulation and plotted. The suggested design for the 45nm gpdk library is designed, simulated, and verified using an EDA platform with a cadence tool. [ABSTRACT FROM AUTHOR]

This study investigates the statistical characteristics of liquid film thickness fluctuations during countercurrent flow limitation on a complex geometry containing a horizontal and inclined pipe which are connected by an elbow. The fluctuation data are obtained by image processing technique which was applied towards high-quality images captured by a high-speed video recording. The result indicates that the thickening of the liquid triggers the liquid blockage that initiates flooding at the upper edge of the hydraulic jump. Moreover, the stochastic analysis reveals that the obtained data imply consistence trends across whole water discharge. Furthermore, it also shows a good agreement to the visual observations. Meanwhile, the wavelet analysis can distinguish energy fluctuations under different water discharges. [ABSTRACT FROM AUTHOR]

The oil palm industry heavily relies on foreign labour for harvesting, including the collection of fresh fruit bunches (FFB) and loose fruits (LF). Manual LF collection, prone to inefficiencies and worker injuries, prompted the development of an automated LF collector. This project introduces a comprehensive system, featuring an LF picker with a robot arm, an LF detector using image processing, a GPS-based human-follower vehicle, a back-to-home navigation system with weight detection, and an obstacle avoidance system. The automated LF collector aims to reduce workforce dependency, enhance LF collection productivity, and prevent worker injuries. The study discusses the motivation, challenges, and objectives of developing the system, emphasizing potential economic and societal benefits. Furthermore, the implementation of advanced image processing techniques, such as the Faster Objects More Objects (FOMO) neural network, is detailed for efficient LF detection. The section on Back-to-Home Navigation with Weight Detection introduces the concept of an autonomous collector robot vehicle, highlighting the importance of sensors and weight detection for navigation and load management. The integration of these technologies promises to revolutionize loose fruit collection in oil palm plantations, reducing labour dependence, increasing productivity, and mitigating economic losses. [ABSTRACT FROM AUTHOR]

This study looks into a number of recently developed pre-trained models and assesses how well they perform using the ImageNet Dataset. In the disciplines of image processing, computer vision, and machine learning, image Classification is a well-known issue. In this study, we studied six different models—three transformer-based models and three convolutional neural network-based models. The findings of this research offer insightful information on the perks and drawbacks of using pre-trained models for image classification tasks. The results indicate that while transformer-based models exhibit promising outcomes, convolutional neural network-based models continue to perform better overall and with regard to accuracy. Additionally, the pre-trained models' performance may be greatly enhanced by fine-tuning them on a smaller dataset. This paper provides an assessment of recently developed pre-trained models and their performance on the ImageNet dataset, which adds to the current research in the fields of image processing, computer vision, and machine learning. The knowledge gathered from this work can guide the creation of models for image classification tasks that are more effective and accurate. [ABSTRACT FROM AUTHOR]

Schizophrenia (SZ) is a complex mental disorder that affects millions of people worldwide, and early diagnosis is critical for effective treatment. Structural changes in certain areas of the brain have been consistently observed across many studies of the disorder. However, existing imaging techniques have limitations in identifying structural brain changes in the disorder because the changes begin prior to the onset of clinical symptoms. The changes observed in Grey Matter (GM) and White Matter (WM) are particularly concerned with language processing and communication, and later, they can be detected by progressive ventricular enlargement. The proposed work involves utilizing various algorithms to perform image processing on both patients with onset schizophrenia (SZ) and healthy controls (HC), with the aim of segmenting the Region of Interest (ROI) of the brain using Magnetic Resonance Images (MRI). The resulting segmented MRI images are then employed to generate a 3D visualization of the brain. The segmentation of WM using the Level based Fuzzy C-Means algorithm (LFCM) and K-means algorithm is established. The segmented ROI is further used for analysis. The goal for the segmentation of brain MRI images in schizophrenia is to identify structural pattern changes of regions such as WM of HC and SZ patients. The approach mentioned has the potential to provide a more objective and accurate diagnosis of the WM region in SZ and to identify new targets for treatment by providing detailed information about the location, size, and shape of abnormalities or lesions. The application of these algorithms for image processing tasks in the proposed research showcases the results which can accurately distinguish subjects with SZ from HC. The segmentation of the WM showed the highest similarity with the ground truth (GT) labels, with Structural Similarity Metrics (SSIM) and Feature Similarity Metrics (FSIM) values of 0.82507 and 0.792124, respectively, for HC, and SSIM 0.827187 and FSIM of 0.782535 respectively for SZ disorder. [ABSTRACT FROM AUTHOR]

Line following is one of the most efficient and reliable techniques for mobile robots in indoor environments. Line-following robots can autonomously detect and follow designed paths. It is now commonly applied in transportation and assistance. In related research, ensuring success in complex line following tracks (right angles, sharp curves) and complex optical environments has been a challenge. This paper designs a line following vehicle is designed based on a microcontroller (Raspberry Pi), Pi camera, and image processing by OpenCV. The image processing in line following includes the application of RGB, GRAY, and HSV color models. The vehicle movement is controlled by the PID algorithm. Besides, this system incorporates a template matching module. The template matching module uses binarization, polygon approximation, perspective transformation, Gaussian blur, and similarity comparison. The designed vehicle can complete the line following a designed map and successfully matches the correct templates. A discussion on the effect of light reflection is presented in this paper. This effect is improved by an extra algorithm in line following and expending the threshold of binarization. Also, this paper solves the problem that the vehicle may run out of the track in line following. The experiment shows a satisfying result. This study has potential implications for the design of line-following robots. [ABSTRACT FROM AUTHOR]

Medical image processing requires image segmentation and image registration as important diagnostic tools. Image segmentation aims to split structures of interest into distinct components. Meanwhile, image registration finds an ideal transformation between given images. Despite their interrelationship, both tasks are often performed separately. As a result, it faces several problems when undertaken independently. Therefore, the joint models between segmentation and registration have advantages over the separate tasks, particularly when dealing with noisy images. This paper reviewed and compared three existing variational joint segmentation and registration (JSR) models based on active contour without edges and linear curvature. The models are tested on 2D mono-modal and multi-modal images with the presence of white Gaussian noise. The Dice coefficient metric and the registration performance measure are used to evaluate the performance of the models. Numerical results showed that the JSR model using the modified sum of the squared difference outperformed the other two JSR models for noisy mono-modal images. However, the JSR model using the modified normalized gradient fields performed better than the JSR model based on the modified mutual information for noisy multi-modal images. [ABSTRACT FROM AUTHOR]

According to the BPS survey, 70% of the rural population of Indonesia is comprised of farmers. This necessitates the availability of vast quantities of fertilizers. Based solely on 2015 data, Indonesia requires fertilizer imports from other countries to reach 6,859.2 tons. In contrast, waste is a worldwide issue that must also be resolved. In actuality, organic waste, particularly leaves, can be used as a component in the production of compost. Compost creation involves a microbial decomposition process. The level of microbial activity (microorganisms) is determined by a number of parameters, including temperature and humidity. This study intends to investigate the effect of homogeneity, temperature, and humidity in the composter on the decomposition rate of compost. Internet of Things (IoT)-based equipment and image processing will be used to monitor the composter box conditions in real time over the internet network in order to improve the monitoring process and the computer's efficiency. The conventional composter was transformed into a smart composter by building a monitoring device with a thermal image sensor so that the mixing of compost fertilizer became better so that it became a new innovation in the environmental management process. [ABSTRACT FROM AUTHOR]

So far, the identification of 16 types of tree damage is still following the guidelines stated in the Forest Health Monitoring method. The types of tree damage can be recognized by human vision, as well as by computers. Computer vision allows computers to identify things that humans can remember. In this case, the study can be realized with computer vision to make work easier. This study aimed to identify 16 types of damage in Forest Health Monitoring using image data or photos with computer vision. The stages of this research are image acquisition (image acquisition), image processing (image preprocessing), and feature extraction (feature extraction). The results showed that computer vision could identify images in JPG/JPEG (Joint Photographic Experts) format, assisted by the Convolutional Neural Network (CNN) algorithm. The percentage value of success using the CNN algorithm reaches 99.06%, with an error detection of 0.94%. However, several classes were incorrectly identified, indicating that the dataset needs improvement. Thus, it can be concluded that identifying 16 types of damage using image data or photos with computer vision has been successfully carried out. [ABSTRACT FROM AUTHOR]

This research paper aims to use UAV photogrammetry in volumetric measurements of different sort of materials that stack in an irregular shape. The research test and evaluate the results delivered from UAV photogrammetry with ground measurements obtained from conventional GNSS and total station techniques in order to show potential of automatic remote sensing techniques in Engineering works. Therefore, we choose an area that contains a stacked bulk materials and remnants of buildings. We used available low-cost hardware for data acquisition, SFM photogrammetric software for processing and analyses to calculate the volume of the stacked materials. Topcon-GR5 multi frequency GNSS receiver used for surveying the position of the base ground truth network precisely. However, mini-UAV drone from DJI Mavic family (DJI mini-2) was used for image acquisition, in addition to total station device to densifying ground points for the reference volume method calculations. The autopilot software used was (drone link) which is a mobile application utilized for flight planning and controlling of the drone. Whereas (Agi-soft) Meta shape photo scan software used for image processing and (AutoCAD) civil 3d for volume calculations. On one side GNSS and total station technologies could perform high accuracy measurements with high level of confidence of the work, but it could really be time consuming and needs high human efforts and remarkable cost for field work data collection. Therefore, this research aims to show the potential of UAV photogrammetry technology to offer a fast, accurate and lower-budget mapping method of large areas with one person managing the process. It considers a powerful volume calculation tool in the field of earthwork quantity surveying in excavation sites, stockpiles, etc. The results show the significant effectiveness of UAV photogrammetry over traditional techniques for estimation, management, and precise calculation of stock and dump materials in engineering surveying works. The resulted volume from UAV technique was 2692.39 m3 with spatial error of ±0.0826 m, while it was 3048.10 m3 from conventional technique with spatial error of ±1.51m. [ABSTRACT FROM AUTHOR]

The most widely used technological advancement today is additive manufacturing, or 3D printing (AM). Currently, additive manufacturing technology is widely developed and used for many different types of tasks, including research and small- and large-scale printing. The heat generated by the 3D printer's base (or bed) and its impact on the printer's output are both crucial study topics. The procedure for gathering data involved using PLA filaments and experimental techniques, which involved heating the nozzle to 190 °C and measuring the printed base's temperature at 30 °C, 50 °C, and 70 °C. Measurements of experimental objects were made to determine the impact of the printed base's heat temperature on their length, width, and height, as well as to determine the pattern's layer-by-layer structure and surface plot or surface roughness. Data was also collected by using macrophoto observations and image processing analysis of the image software image-J with surface roughness(Ra). [ABSTRACT FROM AUTHOR]

An overview of the various methods and algorithms for edge detection in digital images is given in this survey paper. A fundamental issue in image processing is edge detection, which involves in locating the borders which separates various regions in an image. The survey starts with an explanation of edges and their importance in an image analysis, then reviews some of the most widely used edge detection techniques, including the canny edge detector, the Sobel operator, and the Laplacian of gaussian. The following section discusses some of the drawbacks and shortcomings of these techniques. A discussion of current trends and potential future directions in the field of edge detection rounds out the survey. [ABSTRACT FROM AUTHOR]

Cervical cancer is a form of cancer that develops in the cells of the cervix, which links the uterus to the vagina. Cervical cancer can often be prevented by attending the cervical screening, which aims to find and treat changes to cells before they turn into cancer. Pap smear images are a more popular technique for screening cervical cancer. The nucleus and cytoplasm are present in the smear cell images. The spread of cancer is determined by the shape and structure of the nucleus. Thus, nucleus segmentation is an important step in cancer detection. However, overlapping, poor contrast, uneven staining, and other factors make cervical nucleus segmentation difficult. This paper proposes a new segmentation method for the cervical nucleus using digital image processing. In our proposed method, image processing techniques are employed to segment the nucleus. This method mainly compared the performance of using Red Green Blue (RGB) channels and Hue, Saturation, and Value (HSV). The best two channels from each two-colored image representation are compared, where the V channel is the best among H and S channels. On the other hand, G chancel is the best among the G and B channels. Therefore, the best two channels are compared, and the segmentation using the G channel is the best among all, with a sensitivity of 94.3%, specificity of 94.6% and precision of 88.3%. The main impact of this paper will be to assist doctors in diagnosing cervical cancer-based segmentation of the nucleus in Pap smear images. [ABSTRACT FROM AUTHOR]

This work presents an automated and novel system for cricket commentary generation by the introduction of event driven approach and image captioning features. The presented system uses artificial intelligence and machine learning (ML) based methods to investigate real-time match data and generate real-time commentary by Image Captioning technique. For this purpose, deep neural network-based models and digital image processing techniques are used to detect the significant moments in the real-time match and generate commentary based on these real-time match events. The proposed method has been assessed using a dataset of 2 hours live cricket matches of India and England. After processing the match video, it has been observed that the developed model is successfully able to generate high-quality real-time commentary with a significant amount of accuracy. By commissioning leading-edge deep neural network-based model, the developed model determines the fitness to generate subtitles that are not only precise but also contextually appropriate, and efficiently apprehending the essence of the input frames. [ABSTRACT FROM AUTHOR]

Nowadays image processing taking over the internet and plays a major part in human society. We will come across many blurred images in our daily life which may contain information that is very valuable. So, deblurring of an image is the method where we convert low-quality pictures to high-quality pictures. In this process, we deal with noise and crisp to restore the image from the deteriorated image thereby increasing the image quality. This is possible, by applying different types of filters or techniques to theblurred image. The techniques that we worked on for comparison are blind, and non-blind deconvolution methods and the CNN algorithm. In which cnn model gave the better-quality image compared to other methods. We have taken a dataset containing of sharp images from kaggle[9] and generated a dataset of Gaussian blurred images using a Gaussian filter and used this images to build a CNN model, where this CNN model is trained and tested. The low quality picture is given to the CNN model as an input inturn we get the high quality picture as an output. Here the measure psnr ratio is used to determine the quality of the image, where cnn method has shown a higher value compared to other methods. We come across this problem in certain scenarios such as, in criminal perspective, medical field, etc in such cases, this concept is used to solve real-world problems. [ABSTRACT FROM AUTHOR]

Pothole detection is a challenging task as it involves inhomogeneity in intensity, complex topologies, contrast less, and noise rich background. Still, it's a very important task concerning the maintenance of roads. Potholes are being detected using computer vision which generally involves analysis of a two-dimensional image of the road or modeling of the road surface. Provided both these methods are independently used. The accuracy of this method of pothole detection is far from the top point. These bring the usage of digital image processing which develops the detection of potholes with at most accuracy. Digital Image processing provides strong receptivity, usage of huge actions, and broad benefits. In this paper, we propose a system that supports digital image processing techniques of high efficiency to result in consistent and accurate detection of potholes. This project studies an intelligent digital industrial pothole detection mechanism to increase human safety, detection capability and decrease the factor of risk. Firstly, the sources of images from the road are pre-processed to remove noise and other irrelevant features. Then a training model was developed to train the system to classify between path holes and normal cracks using convolution neural networks and MATLAB. Also, an IoT device is used in this system which updates the detected potholes on the road to the respective web server along with the location coordinates, as a part to notify a complaint to check. This system works on providing a better pothole detection technique and also updating the same to the concerned. [ABSTRACT FROM AUTHOR]

FPGAs are well tailored for a student to experiment and FGPA boards help to work on high-end applications like embedded, computer architecture and software/hardware co-design and can be used to implement almost any DSP algorithm. Practical work promotes experiential learning. Implementation of image processing is done using many boards where we aimed at working on zynq boards and attain efficient results. Spartan3E and basys3 are the boards belonging to FPGA family which we used aiming image processing which usually necessitates multiple outcomes involving simple algorithms and sometimes sophisticated transformations. Disparate algorithms lead to many applications where we focused on edge detection in different fields. Many steps involve in image processing where edge detection is one of them which leads to object recognition and using for trace recognition which is reliable in our daily applications. Finding the boundaries to recognize an object is the key point required for image processing. Since there are limitations while going for hardware implementation we would like to experiment on conventional operators. Images used in our project cover conventional lena image using sobel, canny, robert and prewitt techniques. Daily life applications are also covered in sports zone and tracking of objects by taking images of a badminton player and SRA building. Our results show that canny operator works efficiently and show less latency than other techniques. [ABSTRACT FROM AUTHOR]

Curvelet Transform typically offers significantly superior performance in image analysis, multi-resolution and multidi-rectional representation as compared to Wavelet Transform. This paper exploites strong relationship between Wavelet Transform and Curvelet Transform. Also we use mother Wavelet to construct Curvelet Transform as an extension of Wavelet Transform which has broad implications, particularly for the field of signal and image processing. [ABSTRACT FROM AUTHOR]

The oil palm plantation industry heavily relies on foreign labour for harvesting, particularly for collecting loose fruits (LF) alongside fresh fruit bunches (FFB). Manual LF collection, involving bending and repetitive movements, not only diminishes productivity but also poses health risks to workers. This study proposes an automated LF collector to mitigate these challenges. The developed system integrates an LF picker with a robot arm, an LF detector using image processing and a camera, a GPS-based human-follower vehicle, a back-to-home navigation system based on weight detection, and an obstacle avoidance system. The automated LF collector aims to operate autonomously, reducing workforce reliance and enhancing productivity in oil palm plantations. The study discusses the motivation, challenges, and objectives of developing such a system, emphasizing its potential economic and societal benefits. Additionally, the implementation of a novel image processing technique, Faster Objects More Objects (FOMO), using a neural network, is detailed for efficient LF detection. The proposed automated LF collector addresses labour shortages, enhances economic productivity, and reduces the risk of worker injuries in the oil palm industry. [ABSTRACT FROM AUTHOR]

Driver drowsiness and fatigue are leading causes of road accidents, particularly due to drivers having to operate vehicles for long hours under challenging physical and adverse weather conditions. Despite extensive research on subjective, physiological, and vehicle-based detection methods, recent reports still highlight drowsiness and fatigue as significant contributors to accidents and injuries on highways. However, existing drowsiness detection models are often intrusive or expensive. To address these limitations, we propose a low-cost, non-intrusive drowsiness detection model based on facial features, leveraging principles of image processing and machine learning. Our research involves a multi-step approach. Initially, image frames are extracted and subjected to image processing techniques. The images are subsequently passed through a Dlib classifier to detect and localize 68 facial landmarks, including the mouth and eyes. From the localized eye landmarks, salient features such as eye aspect ratio (EAR), pupil circularity (PUC), and eyelid closure over the eye pupil are extracted. To classify the driver's state, these collective features are inputted into various machine learning classifiers. We conducted experiments using the National Tsing Hua University Computer Vision Lab dataset, implementing the proposed approach in the Python OpenCV environment. Notably, the KNN algorithm demonstrated the highest accuracy when the test data was evaluated, yielding a paramount accuracy of 78% on the NTHU DDD dataset. In conclusion, the non-intrusive facial features-based driver drowsiness detection model offers a cost-effective solution to address the pervasive problem of drowsiness-related accidents. The promising results obtained through the utilization of machine learning techniques underscore the potential for practical implementation. [ABSTRACT FROM AUTHOR]

There is a correlation between our posture and our physical and emotional wellness. The detection of various human postures has been approached in a variety of ways. Determine a patient's resting position, for example, using posture analysis in the medical industry. Image processing using OpenCV and mediapipe python library for human posture estimation. Analyzing standing and sitting postures is made easier with an image-processing-based technique. The benefits of fitness activities to a person's health are remarkable, but if they are done improperly, they may be ineffective and even detrimental. Exercise blunders happen when someone doesn't adopt the right posture. This suggested application makes use of pose estimation to identify the user's workout posture and then offers specific, tailored advice on what the user can do to correct their posture. A pose estimation module in python called mediapipe helps to detect all major joints of the human body. Then it calculates the angle between the joints and increases the counter for each repetition. Pose estimate uses a picture or video of the subject to determine the precise locations of the body's major joints. This computer vision technology recognizes human posture in films and displays important regions, such as the elbow or shoulder, in the finished product. Another way to work out at home is through interactive games. The proposed application has an interactive boxing game built using pygame, which helps all age groups to work out and take care of their health. Overall the user can track all their details on a workout on this application, like how many calories are burned and the daily goals, and different workout planners for each individual. [ABSTRACT FROM AUTHOR]

Among all foods, rice is the one that people from different cultures eat the most of. When rice is of high grade, its price rises sharply. As it is, a physical evaluation method using the unaided eye is used to evaluate the rice variety & grade. Nevertheless, this approach is laborious, time-consuming, depends on human expertise, and poses risks to the investigator's wellbeing. This paper proposes a technique that employs computerized methods of image processing to automatically detect and categorise rice grains, therefore addressing the aforementioned problems. This image processing method is ideal because it does not involve any physical touch while photographing the rice grains. To evaluate rice quality, a CNN is utilised to pre-process, partition, & retrieve characteristics from the capturand SVM algorithmic classifiers. Our comparative analysis shows that the suggested system categorization outperforms the alternative. [ABSTRACT FROM AUTHOR]

Effective communication is crucial in our daily lives, and it occurs through various channels such as vocal, written, and body language. However, individuals with hearing impairments often rely on sign language as the primary means of communication. The inability to understand sign language can lead to isolation and barriers in communication, hindering the social lives of deaf individuals. To address this need, we propose a marker-free, visual Indian Sign Language identification system that employs image processing, computer vision, and neural network techniques. Our proposed system analyzes video footage captured by a webcam to recognize hand gestures and translate them into text, which is subsequently converted into audio. The system uses a range of image processing techniques to identify the shape of the hand from continuous video frames, including background subtraction, thresholding, and contour detection. The Haar Cascade Classifier algorithm is used to interpret the signs and assign meaning to them based on the recognized patterns. Finally, a speech synthesizer is employed to convert the displayed text into speech. The proposed system is intended to improve the social lives of deaf individuals by facilitating communication with hearing individuals. It is designed to be user-friendly, efficient, and affordable, as it does not require any additional hardware or markers to recognize signs. The proposed system could be integrated into various devices such as smartphones, tablets, or laptops, making it accessible to a wide range of users. The implementation of such a system could potentially break down communication barriers between the deaf and hearing communities, providing deaf individuals with more opportunities to interact with others and participate in society. [ABSTRACT FROM AUTHOR]

As the number of vehicles increases day by day, finding an empty parking space to park a vehicle becomes difficult. Car parking can cause wasted time and interfere with surrounding mobility, as a result the parking area cannot be utilized optimally. The existing parking management system using sensors to detect the available parking spaces is less effective and efficient, for example a system with the use of ultrasonic sensors that must be placed in each parking box will require many sensors in large-scale implementation. The proposed intelligent parking system provides a structured solution by using a parking lot camera available in the campus or office area to observe the used parking area and using image processing to detect the available parking space from the camera in real time. From the results of image processing available parking spaces will be recommended to users through a front-end system based on the closest distance that helps drivers park their vehicles. The proposed system improves the overall effectiveness and efficiency of the current parking system and solves the problem that drivers spend a lot of time in finding suitable parking spaces in crowded campus or office parking areas. The architecture of the intelligent parking system includes three stages: the first stage of the system uses sensors to capture images of the parking area and sends them to the database server in real time; the second stage of the proposed method uses object detection algorithms (ie, Region-based Convolutional Neural Networks) to identify whether parking spaces in the building area are available or not and calculate their utility; the third stage a front-end system was developed for drivers to get real-time parking information by using a monitor placed at the entrance gate of a campus or office parking area. [ABSTRACT FROM AUTHOR]