Search results for: vision transformer

Authors: Aberdam Edith, Sangari Linda, Petit Isabelle, Aberdam Daniel

Abstract:

Background and Rationale: Transparent avascularised cornea is essential for normal vision and depends on limbal stem cells (LSC) that reside between the cornea and the conjunctiva. Ocular burns or injuries may destroy the limbus, causing limbal stem cell deficiency (LSCD). The cornea becomes vascularised by invaded conjunctival cells, the stroma is scarring, resulting in corneal opacity and loss of vision. Grafted autologous limbus or cultivated autologous LCS can restore the vision, unless the two eyes are affected. Alternative cellular sources have been tested in the last decades, including oral mucosa or hair follicle epithelial cells. However, only partial success has been achieved by the use of these cells since they were not able to uniformly commit into corneal epithelial cells. Human pluripotent stem cells (iPSC) display both unlimited growth capacity and ability to differentiate into any cell type. Our goal was to design a standardized and reproducible protocol to produce transplantable autologous LSC from patients through cell reprogramming technology. Methodology: First, keratinocyte primary culture was established from a small number of plucked hair follicles of healthy donors. The resulting epithelial cells were reprogrammed into induced pluripotent stem cells (iPSCs) and further differentiate into corneal epithelial cells (CEC), according to a robust protocol that recapitulates the main step of corneal embryonic development. qRT-PCR analysis and immunofluorescent staining during the course of differentiation confirm the expression of stage specific markers of corneal embryonic lineage. First appear ectodermal progenitor-specific cytokeratins K8/K18, followed at day 7 by limbal-specific PAX6, TP63 and cytokeratins K5/K14. At day 15, K3/K12+-corneal cells are present. To amplify the iPSC-derived LSC (named COiPSC), intact small epithelial colonies were detached and cultivated in limbal cell-specific medium. In that culture conditions, the COiPSC can be frozen and thaw at any passage, while retaining their corneal characteristics for at least eight passages. To evaluate the potential of COiPSC as an alternative ocular toxicity model, COiPSC were treated at passage P0 to P4 with increasing amounts of SDS and Benzalkonium. Cell proliferation and apoptosis of treated cells was compared to LSC and the SV40-immortalized human corneal epithelial cell line (HCE) routinely used by cosmetological industrials. Of note, HCE are more resistant to toxicity than LSC. At P0, COiPSC were systematically more resistant to chemical toxicity than LSC and even to HCE. Remarkably, this behavior changed with passage since COiPSC at P2 became identical to LSC and thus closer to physiology than HCE. Comparative transcriptome analysis confirmed that COiPSC from P2 are similar to a mixture of LSC and CEC. Finally, by organotypic reconstitution assay, we demonstrated the ability of COiPSC to produce a 3D corneal epithelium on a stromal equivalent made of keratocytes. Conclusion: COiPSC could become valuable for two main applications: (1) an alternative robust tool to perform, in a reproducible and physiological manner, toxicity assays for cosmetic products and pharmacological tests of drugs. (2). COiPSC could become an alternative autologous source for cornea transplantation for LSCD.

Keywords: Limbal stem cell deficiency, iPSC, cornea, limbal stem cells

Procedia PDF Downloads 413

Authors: Kofi Quartey, Kola Ijasan

Abstract:

Achieving densification around transit-oriented development projects has proven the most ideal way of facilitating urban sprawl whilst increasing the mobility of the majority of the urban populations, making parts of the city that were inaccessible, accessible. Johannesburg has undertaken TOD vision, which was initially called the corridors of freedom. The TOD, in line with the Sustainable Development Goal 11, seeks to establish inclusive, sustainable cities and, in line with the Joburg Growth Development Strategy, aims to create an equitable world-class African city. Equity and inclusivity should occur from the onset of planning and implementation of TOD projects through meaningful community participation. Stakeholder engagement literature from various disciplinary backgrounds has documented dissatisfaction of communities regarding the lack of meaningful participation in government-led development initiatives. The views of other project stakeholders such as project policy planners and project implementors and their challenges in undertaking community participation are, however, not taken into account in such instances, leaving room for a biased perspective. Document analysis was undertaken to determine what is expected of the Project stakeholders according to policy and whether they carried out their duties) seven interviews were also conducted with city entities and community representatives to determine their experiences and challenges with community participation in the various TOD projects attributed to the CoF vision. The findings of the study indicated that stakeholder engagement processes were best described as an ‘educative process’; where local communities were limited to being informed from the onset rather than having an active involvement in the planning processes. Most community members felt they were being informed and educated as to what was going to happen in spite of having their views and opinions collected – primarily due to project deadlines and budget constraints, as was confirmed by professional stakeholders. Some community members exhibited reluctance to change due to feelings of having projects being imposed on them, and the implications of the projects on their properties and lifestyles. It is recommended that community participation should remain a participatory and engaging process that creates an exchange of knowledge and understanding in the form of a dialogue between communities and project stakeholders until a consensus is reached.

Keywords: stakeholder engagement, transit oriented development, community participation, Johannesburg

Procedia PDF Downloads 127

Authors: Victoria H. Hamilton

Abstract:

Charles Bonnet Syndrome (CBS) is when a person experiences pseudohallucinations that fill in visual information from any type of sight loss. CBS arises from an epiphenomenal process, with the physical actions of sight resulting in the mental formations of images. These pseudohallucinations—referred to as visions by the CBS community—manifest in a wide range of forms, from complex scenes to simple geometric shapes. To share these unique visual experiences, a remote co-creation website was created where CBS participants communicated their lived experiences. This created a reflexive process, and we worked to produce true representations of these interesting and little-known phenomena. Digital reconstruction of the visions is utilised as it echoes the vivid, experiential movie-like nature of what is being perceived. This paper critically analyses co-creation as a method for making digital assets. The implications of the participants' vision impairments and the application of ethical safeguards are examined in this context. Important to note, this research is of a medical syndrome for a non-medical, practice-based design. CBS research to date is primarily conducted by the ophthalmic, neurological, and psychiatric fields and approached with the primary concerns of these specialties. This research contributes a distinct approach incorporating practice-based digital design, autoethnography, and phenomenology. Autoethnography and phenomenology combine as a foundation, with the first bringing understanding and insights, balanced by the second philosophical, bigger picture, and established approach. With further refining, it is anticipated that the research may be applied to other conditions. Conditions where articulating internal experiences proves challenging and the use of digital methods could aid communication. Both the research and CBS communities will benefit from the insights regarding the relationship between cognitive perceptions and the vision process. This research combines the digital visualising of visions with interest in the link between metaphor, embodied cognition, and image. The argument for a link between CBS visions and metaphor may appear evident due to the cross-category mapping of images that is necessary for comprehension. They both are— CBS visions and metaphors—the experience of picturing images, often with lateral connections and imaginative associations.

Keywords: Charles Bonnet Syndrome, digital design, visual hallucinations, visual perception

Procedia PDF Downloads 44

Authors: Ersumo Tessema, Bogale Girmaye Tamrat, Mohammed Burka

Abstract:

Background: Hyperthyroidism is a common thyroid disorder especially in women and characterized by increased thyroid hormone synthesis and secretion. The disorder manifests predominantly as Graves’ disease in iodine-sufficient areas and has increasing prevalence in iodine-deficient countries in patients with nodular thyroid disease and following iodine fortification. In Ethiopia, the magnitude of the disorder is unknown and, in Africa, due to scarcity of resources, its management remains suboptimal. Objective: The aim of this study was to analyze the pattern and management of patients with hyperthyroidism at the United Vision Medical Services Center, Addis Ababa between August 30, 2013, and February 1, 2018. Patients and methods: The study was a retrospective analysis of medical records of all patients with hyperthyroidism at the United Vision Private Medical Services Center, Addis Ababa. A questionnaire was filled out; the collected data entered into a computer and statistically analyzed using the SPSS package. The results were tabulated and discussed with literature review. Results: A total of 589 patients were included in this study. The median age was 40 years, and the male to female ratio was 1.0:7.9. Most patients (93%) presented with goiter and the associated features of toxic goiter except weight loss, sweating and tachycardia were uncommon. Majority of patients presented more than two years after the onset of their presenting symptoms. The most common physical finding (91%), as well as diagnosis, was toxic nodular goiter. The most frequent (83%) derangement in the thyroid function tests was a low thyroid-stimulating hormone, and the most commonly (94%) used antithyroid drug was a propylthiouracil. The most common (96%) surgical procedure in 213 patients was a near-total thyroidectomy with a postoperative course without incident in 92% of all the patients. Conclusion: The incidence and prevalence of hyperthyroidism are apparently on the increase in Addis Ababa, which may be related to the existing severe iodine-deficiency and or the salt iodation program (iodine-induced hyperthyroidism). Hyperthyroidism predominantly affects women and, in surgical services, toxic nodular goiter is more common than diffuse goiter, and the treatment of choice in experienced hands is a near-total thyroidectomy.

Keywords: Ethiopia, grave’s disease, hyperthyroidism, toxic nodular goiter

Procedia PDF Downloads 176

Authors: R. Schurig, T. Désesquelles, A. Dumont, E. Lefranc, A. Lux

Abstract:

A cost-efficient alternative is proposed to the use of a single drone carrying multiple cameras in order to take stereoscopic images and videos during its flight. Such drone has to be particularly large enough to take off with its equipment, and stable enough in order to make valid measurements. Corresponding performance for a single aircraft usually comes with a large cost. Proposed solution consists in using multiple smaller and cheaper aircrafts carrying one camera each instead of a single expensive one. To give a proof of concept, AR drones, quad-rotor UAVs from Parrot Inc., are experimentally used.

Keywords: drone squadron, flight control, rotorcraft, Unmanned Aerial Vehicle (UAV), AR drone, stereoscopic vision

Procedia PDF Downloads 472

Authors: Vaidehi Murarka, Sneha Mehta, Dishant Upadhyay

Abstract:

Our project topic is “Alphabet Recognition using pixel probability distribution”. The project uses techniques of Image Processing and Machine Learning in Computer Vision. Alphabet recognition is the mechanical or electronic translation of scanned images of handwritten, typewritten or printed text into machine-encoded text. It is widely used to convert books and documents into electronic files etc. Alphabet Recognition based OCR application is sometimes used in signature recognition which is used in bank and other high security buildings. One of the popular mobile applications includes reading a visiting card and directly storing it to the contacts. OCR's are known to be used in radar systems for reading speeders license plates and lots of other things. The implementation of our project has been done using Visual Studio and Open CV (Open Source Computer Vision). Our algorithm is based on Neural Networks (machine learning). The project was implemented in three modules: (1) Training: This module aims “Database Generation”. Database was generated using two methods: (a) Run-time generation included database generation at compilation time using inbuilt fonts of OpenCV library. Human intervention is not necessary for generating this database. (b) Contour–detection: ‘jpeg’ template containing different fonts of an alphabet is converted to the weighted matrix using specialized functions (contour detection and blob detection) of OpenCV. The main advantage of this type of database generation is that the algorithm becomes self-learning and the final database requires little memory to be stored (119kb precisely). (2) Preprocessing: Input image is pre-processed using image processing concepts such as adaptive thresholding, binarizing, dilating etc. and is made ready for segmentation. “Segmentation” includes extraction of lines, words, and letters from the processed text image. (3) Testing and prediction: The extracted letters are classified and predicted using the neural networks algorithm. The algorithm recognizes an alphabet based on certain mathematical parameters calculated using the database and weight matrix of the segmented image.

Keywords: contour-detection, neural networks, pre-processing, recognition coefficient, runtime-template generation, segmentation, weight matrix

Procedia PDF Downloads 389

Authors: Dor Oppenheim, Yael Edan, Guy Shani

Abstract:

This paper presents an image analysis algorithm to detect and count yellow tomato flowers in a greenhouse with uneven illumination conditions, complex growth conditions and different flower sizes. The algorithm is designed to be employed on a drone that flies in greenhouses to accomplish several tasks such as pollination and yield estimation. Detecting the flowers can provide useful information for the farmer, such as the number of flowers in a row, and the number of flowers that were pollinated since the last visit to the row. The developed algorithm is designed to handle the real world difficulties in a greenhouse which include varying lighting conditions, shadowing, and occlusion, while considering the computational limitations of the simple processor in the drone. The algorithm identifies flowers using an adaptive global threshold, segmentation over the HSV color space, and morphological cues. The adaptive threshold divides the images into darker and lighter images. Then, segmentation on the hue, saturation and volume is performed accordingly, and classification is done according to size and location of the flowers. 1069 images of greenhouse tomato flowers were acquired in a commercial greenhouse in Israel, using two different RGB Cameras – an LG G4 smartphone and a Canon PowerShot A590. The images were acquired from multiple angles and distances and were sampled manually at various periods along the day to obtain varying lighting conditions. Ground truth was created by manually tagging approximately 25,000 individual flowers in the images. Sensitivity analyses on the acquisition angle of the images, periods throughout the day, different cameras and thresholding types were performed. Precision, recall and their derived F1 score were calculated. Results indicate better performance for the view angle facing the flowers than any other angle. Acquiring images in the afternoon resulted with the best precision and recall results. Applying a global adaptive threshold improved the median F1 score by 3%. Results showed no difference between the two cameras used. Using hue values of 0.12-0.18 in the segmentation process provided the best results in precision and recall, and the best F1 score. The precision and recall average for all the images when using these values was 74% and 75% respectively with an F1 score of 0.73. Further analysis showed a 5% increase in precision and recall when analyzing images acquired in the afternoon and from the front viewpoint.

Keywords: agricultural engineering, image processing, computer vision, flower detection

Procedia PDF Downloads 329

Authors: Yang-Keun Ahn, Kwang-Soon Choi, Young-Choong Park, Kwang-Mo Jung

Abstract:

Vision based user interfaces to control TVs and PCs have the advantage of being able to perform natural control without being limited to a specific device. Accordingly, various studies on hand gesture recognition using RGB cameras or depth cameras have been conducted. However, such cameras have the disadvantage of lacking in accuracy or the construction cost being large. The proposed method uses a low cost IR camera to accurately differentiate between the hand and the background. Also, complicated learning and template matching methodologies are not used, and the correlation between the fingertips extracted through curvatures is utilized to recognize Click and Move gestures.

Keywords: recognition, hand gestures, infrared camera, RGB cameras

Procedia PDF Downloads 406

Authors: Yahoel Van Essen, Gordon Ho, Brett Russell, Hans-Georg Worms, Xiao Lin Long, Edward David Cooper, Avner Bachar

Abstract:

Weeds reduce farm and forest productivity, invade crops, smother pastures and some can harm livestock. Farmers need to spend a significant amount of money to control weeds by means of biological, chemical, cultural, and physical methods. To solve the global agricultural labor shortage and remove poisonous chemicals, a fully autonomous, eco-friendly, and sustainable weeding technology is developed. This takes the form of a weeding robot, ‘Roboweeder’. Roboweeder includes a four-wheel-drive self-driving vehicle, a 4-DOF robotic arm which is mounted on top of the vehicle, an electromagnetic wave generator (magnetron) which is mounted on the “wrist” of the robotic arm, 48V battery packs, and a control/communication system. Cameras are mounted on the front and two sides of the vehicle. Using image processing and recognition, distinguish types of weeds are detected before being eliminated. The electromagnetic wave technology is applied to heat the individual weeds and clusters dielectrically causing them to wilt and die. The 4-DOF robotic arm was modeled mathematically based on its structure/mechanics, each joint’s load, brushless DC motor and worm gear’ characteristics, forward kinematics, and inverse kinematics. The Proportional-Integral-Differential control algorithm is used to control the robotic arm’s motion to ensure the waveguide aperture pointing to the detected weeds. GPS and machine vision are used to traverse the farm and avoid obstacles without the need of supervision. A Roboweeder prototype has been built. Multiple test trials show that Roboweeder is able to detect, point, and kill the pre-defined weeds successfully although further improvements are needed, such as reducing the “weeds killing” time and developing a new waveguide with a smaller waveguide aperture to avoid killing crops surrounded. This technology changes the tedious, time consuming and expensive weeding processes, and allows farmers to grow more, go organic, and eliminate operational headaches. A patent of this technology is pending.

Keywords: autonomous navigation, machine vision, precision heating, sustainable and eco-friendly

Procedia PDF Downloads 252

Authors: Arpit Rai

Abstract:

In this work, we propose PatchMix, a semi-supervised method for pre-training visual representations. PatchMix mixes patches of two images and then solves an auxiliary task of predicting the label of each patch in the mixed image. Our experiments on the CIFAR-10, 100 and the SVHN dataset show that the representations learned by this method encodes useful information for transfer to new tasks and outperform the baseline Residual Network encoders by on CIFAR 10 by 12% on ResNet 101 and 2% on ResNet-56, by 4% on CIFAR-100 on ResNet101 and by 6% on SVHN dataset on the ResNet-101 baseline model.

Keywords: self-supervised learning, representation learning, computer vision, generalization

Procedia PDF Downloads 89

Authors: M. Ananthakrishnan, Sunil K Patil, Koti Naveen, Inuganti Hemanth Kumar

Abstract:

State estimation output of EMS forms the base case for all other advanced applications used in real time by a power system operator. Ensuring tuning of state estimator is a repeated process and cannot be left once a good solution is obtained. This paper attempts to demonstrate methods to improve state estimator solution by identifying incorrect modelling and telemetry inputs to the application. In this work, identification of database topology modelling error by plotting static network using node-to-node connection details is demonstrated with examples. Analytical methods to identify wrong transmission parameters, incorrect limits and mistakes in pseudo load and generator modelling are explained with various cases observed. Further, methods used for active and reactive power tuning using bus summation display, reactive power absorption summary, and transformer tap correction are also described. In a large power system, verifying all network static data and modelling parameter on regular basis is difficult .The proposed tuning methods can be easily used by operators to quickly identify errors to obtain the best possible state estimation performance. This, in turn, can lead to improved decision-support capabilities, ultimately enhancing the safety and reliability of the power grid.

Keywords: active power tuning, database modelling, reactive power, state estimator

Procedia PDF Downloads 7

Authors: Snehal G. Teli, R. J. Shelke

Abstract:

CNN and MultiUNet models are the framework for the proposed method for enhancing and reconstructing underwater images. Multiscale merging of features and regeneration are both performed by the MultiUNet. CNN collects relevant features. Extensive tests on benchmark datasets show that the proposed strategy performs better than the latest methods. As a result of this work, underwater images can be represented and interpreted in a number of underwater applications with greater clarity. This strategy will advance underwater exploration and marine research by enhancing real-time underwater image processing systems, underwater robotic vision, and underwater surveillance.

Keywords: convolutional neural network, image enhancement, machine learning, multiunet, underwater images

Procedia PDF Downloads 75

Authors: Pruthvi Nath Palleti, Sarmada Madhulika Kone

Abstract:

Communities are formed based on the commonalities that exist in a set of individuals and when the group comes together on identifying those commonalities and to achieve their common goals. Thus, community planning with its vision to strengthen the community mostly involves with making or remaking of commons, which results in making or remaking of communities. This paper looks into different practices of planning around the world and tried to establish a link between commoning (the act of exercising the rights over commons by commoners) and participatory approach to community planning.

Keywords: commoners, commoning, community, participatory planning, urban local bodies

Procedia PDF Downloads 373

Authors: Medjram Sofiane, Babahenini Mohamed Chaouki, Mohamed Benali Yamina

Abstract:

Skin detection is an important task for computer vision systems. A good method for skin detection means a good and successful result of the system. The colour is a good descriptor that allows us to detect skin colour in the images, but because of lightings effects and objects that have a similar colour skin, skin detection becomes difficult. In this paper, we proposed a method using the YCbCr colour space for skin detection and lighting effects elimination, then we use the information of texture to eliminate the false regions detected by the YCbCr colour skin model.

Keywords: skin detection, YCbCr, GLCM, texture, human skin

Procedia PDF Downloads 459

Authors: R. M. Kalayappan, N. Kathiravan

Abstract:

In automobile industries, Controller Area Network (CAN) is widely used to reduce the system complexity and inter-task communication. Therefore, this paper proposes the hardware implementation of data frame communication between one controller to other. The CAN data frames and protocols will be explained deeply, here. The data frames are transferred without any collision or corruption. The simulation is made in the KEIL vision software to display the data transfer between transmitter and receiver in CAN. ARM7 micro-controller is used to transfer data’s between the controllers in real time. Data transfer is verified using the CRO.

Keywords: control area network (CAN), automotive electronic control unit, CAN 2.0, industry

Procedia PDF Downloads 398

Authors: John Allard, Alex Rich, Iris Aguilar, Zachary Dodds

Abstract:

With this paper we present several experiences deploying novel, low-cost resources for computing with 3D spatial models. Certainly, computing with 3D models undergirds some of our field’s most important contributions to the human experience. Most often, those are contrived artifacts. This work extends that tradition by focusing on novel resources that deliver uncontrived models of a system’s current surroundings. Atop this new capability, we present several projects investigating the student-accessibility of the computational tools for reasoning about the 3D space around us. We conclude that, with current scaffolding, real-world 3D models are now an accessible and viable foundation for creative computational work.

Keywords: 3D vision, matterport model, real-world 3D models, mathematical and computational methods

Procedia PDF Downloads 536

Authors: Salah Al-Enezi, Rashed Al-Zufairi, Naseer Ahmad

Abstract:

A Thermo-mechanical technique was developed to determine softening point temperature/glass transition temperature (Tg) of polystyrene exposed to high pressures. The design utilizes the ability of carbon dioxide to lower the glass transition temperature of polymers and acts as plasticizer. In this apparatus, the sorption of carbon dioxide to induce softening of polymers as a function of temperature/pressure is performed and the extent of softening is measured in three-point-flexural-bending mode. The polymer strip was placed in the cell in contact with the linear variable differential transformer (LVDT). CO₂ was pumped into the cell from a supply cylinder to reach high pressure. The results clearly showed that full softening point of the samples, accompanied by a large deformation on the polymer strip. The deflection curves are initially relatively flat and then undergo a dramatic increase as the temperature is elevated. It was found that increasing the pressure of CO₂ causes the temperature curves to shift from higher to lower by increment of about 45 K, over the pressure range of 0-120 bars. The obtained experimental Tg values were validated with the values reported in the literature. Finally, it is concluded that the defection model fits consistently to the generated experimental results, which attempts to describe in more detail how the central deflection of a thin polymer strip affected by the CO₂ diffusions in the polymeric samples.

Keywords: softening, high-pressure, polystyrene, CO₂ diffusions

Procedia PDF Downloads 127

Authors: Reza Nematirad, Anil Pahwa, Balasubramaniam Natarajan

Abstract:

Time series data consist of successive data points collected over a period of time. Accurate prediction of future values is essential for informed decision-making in several real-world applications, including electricity load demand forecasting, lifetime estimation of industrial machinery, traffic planning, weather prediction, and the stock market. Due to their critical relevance and wide application, there has been considerable interest in time series forecasting in recent years. However, the proliferation of sensors and IoT devices, real-time monitoring systems, and high-frequency trading data introduce significant intricate temporal variations, rapid changes, noise, and non-linearities, making time series forecasting more challenging. Classical methods such as Autoregressive integrated moving average (ARIMA) and Exponential Smoothing aim to extract pre-defined temporal variations, such as trends and seasonality. While these methods are effective for capturing well-defined seasonal patterns and trends, they often struggle with more complex, non-linear patterns present in real-world time series data. In recent years, deep learning has made significant contributions to time series forecasting. Recurrent Neural Networks (RNNs) and their variants, such as Long short-term memory (LSTMs) and Gated Recurrent Units (GRUs), have been widely adopted for modeling sequential data. However, they often suffer from the locality, making it difficult to capture local trends and rapid fluctuations. Convolutional Neural Networks (CNNs), particularly Temporal Convolutional Networks (TCNs), leverage convolutional layers to capture temporal dependencies by applying convolutional filters along the temporal dimension. Despite their advantages, TCNs struggle with capturing relationships between distant time points due to the locality of one-dimensional convolution kernels. Transformers have revolutionized time series forecasting with their powerful attention mechanisms, effectively capturing long-term dependencies and relationships between distant time points. However, the attention mechanism may struggle to discern dependencies directly from scattered time points due to intricate temporal patterns. Lastly, Multi-Layer Perceptrons (MLPs) have also been employed, with models like N-BEATS and LightTS demonstrating success. Despite this, MLPs often face high volatility and computational complexity challenges in long-horizon forecasting. To address intricate temporal variations in time series data, this study introduces Times2D, a novel framework that parallelly integrates 2D spectrogram and derivative heatmap techniques. The spectrogram focuses on the frequency domain, capturing periodicity, while the derivative patterns emphasize the time domain, highlighting sharp fluctuations and turning points. This 2D transformation enables the utilization of powerful computer vision techniques to capture various intricate temporal variations. To evaluate the performance of Times2D, extensive experiments were conducted on standard time series datasets and compared with various state-of-the-art algorithms, including DLinear (2023), TimesNet (2023), Non-stationary Transformer (2022), PatchTST (2023), N-HiTS (2023), Crossformer (2023), MICN (2023), LightTS (2022), FEDformer (2022), FiLM (2022), SCINet (2022a), Autoformer (2021), and Informer (2021) under the same modeling conditions. The initial results demonstrated that Times2D achieves consistent state-of-the-art performance in both short-term and long-term forecasting tasks. Furthermore, the generality of the Times2D framework allows it to be applied to various tasks such as time series imputation, clustering, classification, and anomaly detection, offering potential benefits in any domain that involves sequential data analysis.

Keywords: derivative patterns, spectrogram, time series forecasting, times2D, 2D representation

Procedia PDF Downloads 42

Authors: Kenza Benali

Abstract:

Over the past decade, a number of remarkable projects in urban brownfield reconversion emerged across Canada, including the reconversion of former military bases owned by the Canada Lands Company (CLC) into sustainable communities. However, unlike other developments, the regeneration project of the former Canadian Forces Base Rockcliffe in Ottawa – which was announced as one of the most ambitious Smart growth projects in Canada – faced serious obstacles in terms of social acceptance by the local community, particularly urban minorities composed of Francophones, Indigenous and vulnerable groups who live near or on the Base. This turn of events led to the project being postponed and even reconsidered. Through an analysis of its press coverage, this research aims to understand the causes of this urban conflict which lasted for nearly ten years. The findings reveal that the conflict is not limited to the “standard” issues common to most conflicts related to urban mega-projects in the world – e.g., proximity issues (threads to the quality of the surrounding neighbourhoods; noise, traffic, pollution, New-build gentrification) often associated with NIMBY phenomena. In this case, the local actors questioned the purpose of the project (for whom and for what types of uses is it conceived?), its local implementation (to what extent are the local history and existing environment taken into account?), and the degree of implication of the local population in the decision-making process (with whom is the project built?). Moreover, the interests of the local actors have “jumped scales” and transcend the micro-territorial level of their daily life to take on a national and even international dimension. They defined an alternative view of how this project, considered strategic by his location in the nation’s capital, should be a reference as well as an international showcase of Canadian ambition and achievement in terms of urban sustainability. This vision promoted, actually, a territorial and national identity approach - in which some cultural values are highly significant (respect of social justice, inclusivity, ethnical diversity, cultural heritage, etc.)- as a counterweight to planners’ vision which is criticized as a normative/ universalist logic that ignore the territorial peculiarities.

Keywords: smart growth, brownfield reconversion, sustainable neighborhoods, Canada Lands Company, Canadian Forces Base Rockcliffe, urban conflicts

Procedia PDF Downloads 382

Authors: Evan Lowhorn, Rocio Alba-Flores

Abstract:

Traditional classification Convolutional Neural Networks (CNN) attempt to classify an image in its entirety. This becomes problematic when trying to perform classification with a drone’s camera in real-time due to unpredictable backgrounds. Object detectors with bounding boxes can be used to isolate individuals and other items, but the original backgrounds remain within these boxes. These basic detectors have been regularly used to determine what type of object an item is, such as “person” or “dog.” Recent advancement in computer vision, particularly with human imaging, is keypoint detection. Human keypoint detection goes beyond bounding boxes to fully isolate humans and plot points, or Regions of Interest (ROI), on their bodies within an image. ROIs can include shoulders, elbows, knees, heads, etc. These points can then be related to each other and used in deep learning methods such as pose estimation. For drone control based on human motions, poses, or signals using the onboard camera, it is important to have a simple method for pilot identification among multiple individuals while also giving the pilot fine control options for the drone. To achieve this, the OpenPose keypoint detection network was used with body and hand keypoint detection enabled. OpenPose supports the ability to combine multiple keypoint detection methods in real-time with a single network. Body keypoint detection allows simple poses to act as the pilot identifier. The hand keypoint detection with ROIs for each finger can then offer a greater variety of signal options for the pilot once identified. For this work, the individual must raise their non-control arm to be identified as the operator and send commands with the hand on their other arm. The drone ignores all other individuals in the onboard camera feed until the current operator lowers their non-control arm. When another individual wish to operate the drone, they simply raise their arm once the current operator relinquishes control, and then they can begin controlling the drone with their other hand. This is all performed mid-flight with no landing or script editing required. When using a desktop with a discrete NVIDIA GPU, the drone’s 2.4 GHz Wi-Fi connection combined with OpenPose restrictions to only body and hand allows this control method to perform as intended while maintaining the responsiveness required for practical use.

Keywords: computer vision, drone control, keypoint detection, openpose

Procedia PDF Downloads 184

Authors: Jalal M. Abdallah

Abstract:

There is a great problem in testing and investigations the reliability of different type of transformers insulation materials. It summarized in how to create and simulate the real conditions of working transformer and testing its insulation materials for Partial Discharge PD, typically as in the working mode. A lot of tests may give untrue results as the physical behavior of the insulation material differs under tests from its working condition. In this work, the real working conditions were simulated, and a large number of specimens have been tested. The investigations first stage, begin with choosing samples of different types of insulation materials (papers, pressboards, etc.). The second stage, the samples were dried in ovens at 105 C⁰and 0.01bar for 48 hours, and then impregnated with dried and gasless oil (the water content less than 6 ppm.) at 105 C⁰and 0.01bar for 48 hours, after so specimen cooling at room pressure and temperature for 24 hours. The third stage is investigating PD for the samples using ICM PD measuring device. After that, a continuous test on oil-impregnated insulation materials (paper, pressboards) was developed, and the phase resolved partial discharge pattern of PD signals was measured. The important of this work in providing the industrial sector with trusted high accurate measuring results based on real simulated working conditions. All the PD patterns (results) associated with a discharge produced in well-controlled laboratory condition. They compared with other previous and other laboratory results. In addition, the influence of different temperatures condition on the partial discharge activities was studied.

Keywords: transformers, insulation materials, voids, partial discharge

Procedia PDF Downloads 315

Authors: Laura Simmer, Maria Kalt, Oliver Schauer

Abstract:

Freight transport is growing fast, but many vehicles are empty or just partially loaded. The vision and concepts of the Physical Internet (PI) proposes to eliminate these inefficiencies. Aiming for a radical sustainability improvement, the PI – inspired by the Digital Internet – is a hyperconnected global logistic system, enabling seamless asset sharing and flow consolidation. The implementation of a PI in its full expression will be a huge challenge: the industry needs innovation and implementation support including change management approaches, awareness creation and good practices diffusion, legislative actions to remove antitrust and international commerce barriers, standardization and public incentives policies. In order to take a step closer to this future the project ‘Atropine - Fast Track to the Physical Internet’ funded under the Strategic Economic and Research Program ‘Innovative Upper Austria 2020’ was set up. The two-year research project unites several research partners in this field, but also industrial partners and logistics service providers. With Atropine, the consortium wants to actively shape the mobility landscape in Upper Austria and make an innovative contribution to an energy-efficient, environmentally sound and sustainable development in the transport area. This paper should, on the one hand, clarify the questions what the project Atropine is about and, on the other hand, how a proof of concept will be reached. Awareness building plays an important role in the project as the PI requires a reorganization of the supply chain and the design of completely new forms of inter-company co-operation. New business models have to be developed and should be verified by simulation. After the simulation process one of these business models will be chosen and tested in real life with the partner companies. The developed results - simulation model and demonstrator - are used to determine how the concept of the PI can be applied in Upper Austria. Atropine shall pave the way for a full-scale development of the PI vision in the next few decades and provide the basis for pushing the industry toward a new level of co-operation with more shared resources and increased standardization.

Keywords: Atropine, inter-company co-operation, Physical Internet, shared resources, sustainable logistics

Procedia PDF Downloads 223

Authors: Atena Tazikeh Lemeski, Ismail Ozdamar

Abstract:

This study presents a detailed analysis of harmonic distortion in a grid-connected photovoltaic (PV) system using real-time data captured from a solar power plant. Harmonics introduced by inverters in PV systems can degrade power quality and lead to increased Total Harmonic Distortion (THD), which poses challenges such as transformer overheating, increased power losses, and potential grid instability. This research addresses these issues by applying Fast Fourier Transform (FFT) to identify significant harmonic components and employing notch filters to target specific frequencies, particularly the 3rd harmonic (150 Hz), which was identified as the largest contributor to THD. Initial analysis of the unfiltered voltage signal revealed a THD of 21.15%, with prominent harmonic peaks at 150 Hz, 250 Hz and 350 Hz, corresponding to the 3rd, 5th, and 7th harmonics, respectively. After implementing the notch filters, the THD was reduced to 5.72%, demonstrating the effectiveness of this approach in mitigating harmonic distortion without affecting the fundamental frequency. This paper provides practical insights into the application of real-time filtering techniques in PV systems and their role in improving overall grid stability and power quality. The results indicate that targeted harmonic mitigation is crucial for the sustainable integration of renewable energy sources into modern electrical grids.

Keywords: grid-connected photovoltaic systems, fast Fourier transform, harmonic filtering, inverter-induced harmonics

Procedia PDF Downloads 34

Authors: Lawrence Gumbe

Abstract:

Kenya Vision 2030 is the country's programme for transformation covering the period 2008 to 2030. Its objective is to help transform Kenya into a newly industrializing, middle-income, exceeding US$10000, country providing a high quality of life to all its citizens by 2030, in a clean and secure environment. Increased agricultural and production and productivity is crucial for the realization of Vision 2030. Mechanization of agriculture in order to achieve greater yields is the only way to achieve these objectives. There are contending groups and views on the strategy for agricultural mechanization. The first group are those who oppose the widespread adoption of advanced technologies (mostly internal combustion engines and tractors) in agricultural mechanization as entirely inappropriate in most situations in developing countries. This group argues that mechanically powered -agricultural mechanization often leads to displacement of labour and hence increased unemployment, and this results in a host of other socio-economic problems, amongst them, rural-urban migration, inequitable distribution of wealth and in many cases an increase in absolute poverty, balance of payments due to the need to import machinery, fuel and sometimes technical assistance to manage them. The second group comprises of those who view the use of the improved hand tools and animal powered technology as transitional step between the most rudimentary step in technological development (characterized by entire reliance on human muscle power) and the advanced technologies (characterized 'by reliance on tractors and other machinery). The third group comprises those who regard these intermediate technologies (ie. improved hand tools and draught animal technology in agriculture) as a ‘delaying’ tactic and they advocate the use of mechanical technologies as-the most appropriate. This group argues that alternatives to the mechanical technologies do not just exist as a practical matter, or, if they are available, they are inefficient and they cannot be compared to the mechanical technologies in terms of economics and productivity. The fourth group advocates a compromise between groups two and third above. This group views the improved hand tools and draught animal technology as more of an 18th century technology and the modem tractor and combine harvester as too advanced for developing countries. This group has been busy designing an ‘intermediate’, ‘appropriate’, ‘mini’, ‘micro’ tractor for use by farmers in developing countries. This paper analyses and concludes on the different agricultural mechanization strategies available to Kenya and other third world countries

Keywords: agriculture, mechanazation, transformation, industrialization

Procedia PDF Downloads 338

Authors: S. Murr, S. Phillips

Abstract:

One of the biggest challenges entering a market with a carsharing or any other shared mobility (SM) service is sound investment decision-making. To support this process, the authors think that a city index evaluating different criteria is necessary. The goal of such an index is to benchmark cities along a set of external measures to answer the main two challenges: financially viability and the understanding of its specific requirements. The authors have consulted several shared mobility projects and industry experts to create such a Shared Mobility City Index (SMCI). The current proposal of the SMCI consists of 11 individual index measures: general data (demographics, geography, climate and city culture), shared mobility landscape (current SM providers, public transit options, commuting patterns and driving culture) and political vision and goals (vision of the Mayor, sustainability plan, bylaws/tenders supporting SM). To evaluate the suitability of the index, 16 cities on the East Coast of North America were selected and secondary research was conducted. The main sources of this study were census data, organisational records, independent press releases and informational websites. Only non-academic sources where used because the relevant data for the chosen cities is not published in academia. Applying the index measures to the selected cities resulted in three major findings. Firstly, density (city area divided by number of inhabitants) is not an indicator for the number of SM services offered: the city with the lowest density has five bike and carsharing options. Secondly, there is a direct correlation between commuting patterns and how many shared mobility services are offered. New York, Toronto and Washington DC have the highest public transit ridership and the most shared mobility providers. Lastly, except one, all surveyed cities support shared mobility with their sustainability plan. The current version of the shared mobility index is proving a practical tool to evaluate cities, and to understand functional, political, social and environmental considerations. More cities will have to be evaluated to refine the criteria further. However, the current version of the index can be used to assess cities on their suitability for shared mobility services and will assist investors deciding which city is a financially viable market.

Keywords: carsharing, transportation, urban planning, shared mobility city index

Procedia PDF Downloads 303

Authors: Shao-Tzu Huang, Chen-Chien Hsu, Wei-Yen Wang

Abstract:

Matching high dimensional features between images is computationally expensive for exhaustive search approaches in computer vision. Although the dimension of the feature can be degraded by simplifying the prior knowledge of homography, matching accuracy may degrade as a tradeoff. In this paper, we present a feature matching method based on k-means algorithm that reduces the matching cost and matches the features between images instead of using a simplified geometric assumption. Experimental results show that the proposed method outperforms the previous linear exhaustive search approaches in terms of the inlier ratio of matched pairs.

Keywords: feature matching, k-means clustering, SIFT, RANSAC

Procedia PDF Downloads 357

Authors: Ahmet Faruk Akyuz, Hasan Sakir Bilge

Abstract:

Depth estimation using stereo images is a challenging problem in computer vision. Many different studies have been carried out to solve this problem. With advancing machine learning, tackling this problem is often done with neural network-based solutions. The images used in these studies are mostly in the visible spectrum. However, the need to use the Infrared (IR) spectrum for depth estimation has emerged because it gives better results than visible spectra in some conditions. At this point, we recommend using thermal-thermal (IR) image pairs for depth estimation. In this study, we used two well-known networks (PSMNet, FADNet) with minor modifications to demonstrate the viability of this idea.

Keywords: thermal stereo matching, deep neural networks, CNN, Depth estimation

Procedia PDF Downloads 279

Authors: Damith Tennakoon, Mojgan Jadidi, Seyedreza Razavialavi

Abstract:

With recent advancements of automation in robotics, from self-driving cars to autonomous 4-legged quadrupeds, one industry that has been stagnant is the construction industry. The methodologies used in a modern-day construction site consist of arduous physical labor and the use of heavy machinery, which has not changed over the past few decades. The dangers of a modern-day construction site affect the health and safety of the workers due to performing tasks such as lifting and moving heavy objects and having to maintain unhealthy posture to complete repetitive tasks such as painting, installing drywall, and laying bricks. Further, training for heavy machinery is costly and requires a lot of time due to their complex control inputs. The main focus of this research is using immersive wearable technology and robotic arms to perform the complex and intricate skills of modern-day construction workers while alleviating the physical labor requirements to perform their day-to-day tasks. The methodology consists of mounting a stereo vision camera, the ZED Mini by Stereolabs, onto the end effector of an industrial grade robotic arm, streaming the video feed into the Virtual Reality (VR) Meta Quest 2 (Quest 2) head-mounted display (HMD). Due to the nature of stereo vision, and the similar field-of-views between the stereo camera and the Quest 2, human-vision can be replicated on the HMD. The main advantage this type of camera provides over a traditional monocular camera is it gives the user wearing the HMD a sense of the depth of the camera scene, specifically, a first-person view of the robotic arm’s end effector. Utilizing the built-in cameras of the Quest 2 HMD, open-source hand-tracking libraries from OpenXR can be implemented to track the user’s hands in real-time. A mixed-reality (XR) Unity application can be developed to localize the operator's physical hand motions with the end-effector of the robotic arm. Implementing gesture controls will enable the user to move the robotic arm and control its end-effector by moving the operator’s arm and providing gesture inputs from a distant location. Given that the end effector of the robotic arm is a gripper tool, gripping and opening the operator’s hand will translate to the gripper of the robot arm grabbing or releasing an object. This human-robot interaction approach provides many benefits within the construction industry. First, the operator’s safety will be increased substantially as they can be away from the site-location while still being able perform complex tasks such as moving heavy objects from place to place or performing repetitive tasks such as painting walls and laying bricks. The immersive interface enables precision robotic arm control and requires minimal training and knowledge of robotic arm manipulation, which lowers the cost for operator training. This human-robot interface can be extended to many applications, such as handling nuclear accident/waste cleanup, underwater repairs, deep space missions, and manufacturing and fabrication within factories. Further, the robotic arm can be mounted onto existing mobile robots to provide access to hazardous environments, including power plants, burning buildings, and high-altitude repair sites.

Keywords: construction automation, human-robot interaction, hand-tracking, mixed reality

Procedia PDF Downloads 80

Authors: Somorjeet Sharma Shamurailatpam, Rabindra Das, A. Suchitra Devi

Abstract:

Computer and video display users might complain about Asthenopia, burning, dry eyes etc. The management of dry eyes is often not in the lines of severity. Following systematic evaluation and grading, dry eye disease is one condition that can be practiced at all levels of ophthalmic care. In the present study, different spectrum causing dry eye and prevalence of dry eye disease in computer users of Manipur, India are determined with 600 individuals (300 cases and 300 control). Individuals between 15 and 50 years who used computers for more than 3 hrs a day for 1 year or more were included. Tear break up time (TBUT) and Schirmer’s test were conducted. It shows that 33 (20.4%) out of 164 males and 47 (30.3%) out of 136 females have dry eye. Possible explanation for the observed result is discussed.

Keywords: asthenopia, computer vision syndrome, dry eyes, Schirmer's test, TBUT

Procedia PDF Downloads 374

Authors: Zeenat Parveen, Ashiq Hussain

Abstract:

This review paper consists of applications of PDMS (polydimethylsiloxane) materials for enhanced performance, optical fiber sensors in acousto-ultrasonic, mechanical measurements, current applications, sensing, measurements and interferometric optical fiber sensors. We will discuss the basic working principle of fiber optic sensing technology, various types of fiber optic and the PDMS as a coating material to increase the performance. Optical fiber sensing methods for detecting dynamic strain signals, including general sound and acoustic signals, high frequency signals i.e. ultrasonic/ultrasound, and other signals such as acoustic emission and impact induced dynamic strain. Optical fiber sensors have Industrial and civil engineering applications in mechanical measurements. Sometimes it requires different configurations and parameters of sensors. Optical fiber current sensors are based on Faraday Effect due to which we obtain better performance as compared to the conventional current transformer. Recent advancement and cost reduction has simulated interest in optical fiber sensing. Optical techniques are also implemented in material measurement. Fiber optic interferometers are used to sense various physical parameters including temperature, pressure and refractive index. There are four types of interferometers i.e. Fabry–perot, Mach-Zehnder, Michelson, and Sagnac. This paper also describes the future work of fiber optic sensors.

Keywords: fiber optic sensing, PDMS materials, acoustic, ultrasound, current sensor, mechanical measurements

Procedia PDF Downloads 388