Search results for: monocular camera

Authors: Nasir A. Al-Geelani, Zulkurnain Abdul-Malek, M. Afendi M. Piah

Abstract:

This research paper investigation is carried out in the laboratory on single units of transmission line glass insulator characterized by different thermal images, which aimed to find out the age of the insulators. The tests were carried out on virgin and aged insulators using the thermography scan. Various samples having different periods of aging 20, 15, and 5 years from a 132 kV transmission line which have exhibited a different degree of corrosion. The second group of insulator samples was relatively mild aged insulators, while the third group was lightly aged; finally, the fourth group was the brand new insulators. The results revealed a strong correlation between the aging and the thermal images captured by the infrared camera. This technique can be used to monitor the aging of high voltage insulators as a precaution to avoid disaster.

Keywords: glass insulator, infrared camera, corona diacharge, transmission lines, thermograpy, surface discharge

Procedia PDF Downloads 160

Authors: Ruquan You, Haiwang Li, Zhi Tao

Abstract:

A time-resolved particle image velocimetry (PIV) system is developed to investigate the boundary layer flow with the effect of rotating Coriolis and buoyancy force. This time-resolved PIV system consists of a 10 Watts continuous laser diode and a high-speed camera. The laser diode is able to provide a less than 1mm thickness sheet light, and the high-speed camera can capture the 6400 frames per second with 1024×1024 pixels. The whole laser and the camera are fixed on the rotating facility with 1 radius meters and up to 500 revolutions per minute, which can measure the boundary flow velocity in the rotating channel with and without ribs directly at rotating conditions. To investigate the effect of buoyancy force, transparent heater glasses are used to provide the constant thermal heat flux, and then the density differences are generated near the channel wall, and the buoyancy force can be simulated when the channel is rotating. Due to the high temporal and spatial resolution of the system, the proper orthogonal decomposition (POD) can be developed to analyze the characteristic of the turbulent boundary layer flow at rotating conditions. With this rotating facility and PIV system, the velocity profile, Reynolds shear stress, spatial and temporal correlation, and the POD modes of the turbulent boundary layer flow can be discussed.

Keywords: rotating facility, PIV, boundary layer flow, spatial and temporal resolution

Procedia PDF Downloads 180

Authors: Samane Sharifi Monfared, Lavdie Rada

Abstract:

Nowadays, autonomous vehicles are developing rapidly toward facilitating human car driving. One of the main issues is road lane detection for a suitable guidance direction and car accident prevention. This paper aims to improve and optimize road line detection based on a combination of camera calibration, the Hough transform, and Canny edge detection. The video processing is implemented using the Open CV library with the novelty of having a scale able region masking. The aim of the study is to introduce automatic road lane detection techniques with the user’s minimum manual intervention.

Keywords: hough transform, canny edge detection, optimisation, scaleable masking, camera calibration, improving the quality of image, image processing, video processing

Procedia PDF Downloads 94

Authors: Y. C. Khoo, W. T. Lai

Abstract:

The single orifice hydraulic flat spray nozzle was evaluated with two global imaging techniques to characterize various aspects of the resulting spray. The two techniques were high resolution flow visualization and Particle Image Velocimetry (PIV). A CCD camera with 29 million pixels was used to capture shadowgraph images to realize ligament formation and collapse as well as droplet interaction. Quantitative analysis was performed to give the sizing information of the droplets and ligaments. This camera was then applied with a PIV system to evaluate the overall velocity field of the spray, from nozzle exit to droplet discharge. PIV images were further post-processed to determine the inclusion angle of the spray. The results from those investigations provided significant quantitative understanding of the spray structure. Based on the quantitative results, detailed understanding of the spray behavior was achieved.

Keywords: spray, flow visualization, PIV, shadowgraph, quantitative sizing, velocity field

Procedia PDF Downloads 381

Authors: Sonia Sassi, Mostapha Tarfaoui, Hamza Ben Yahia

Abstract:

In this investigation, an experimental technique in which the dynamic response, damage kinetic and heat dissipation are measured simultaneously during high strain rates on adhesively bonded joints materials. The material used in this study is widely used in the design of structures for military applications. It was composed of a 45° Bi-axial fiber-glass mat of 0.286 mm thickness in a Polyester resin matrix. In adhesive bonding, a NORPOL Polyvinylester of 1 mm thickness was used to assemble the composite substrate. The experimental setup consists of a compression Split Hopkinson Pressure Bar (SHPB), a high-speed infrared camera and a high-speed Fastcam rapid camera. For the dynamic compression tests, 13 mm x 13 mm x 9 mm samples for out-of-plane tests were considered from 372 to 1030 s-1. Specimen surface is controlled and monitored in situ and in real time using the high-speed camera which acquires the damage progressive in specimens and with the infrared camera which provides thermal images in time sequence. Preliminary compressive stress-strain vs. strain rates data obtained show that the dynamic material strength increases with increasing strain rates. Damage investigations have revealed that the failure mainly occurred in the adhesive/adherent interface because of the brittle nature of the polymeric adhesive. Results have shown the dependency of the dynamic parameters on strain rates. Significant temperature rise was observed in dynamic compression tests. Experimental results show that the temperature change depending on the strain rate and the damage mode and their maximum exceed 100 °C. The dependence of these results on strain rate indicates that there exists a strong correlation between damage rate sensitivity and heat dissipation, which might be useful when developing damage models under dynamic loading tacking into account the effect of the energy balance of adhesively bonded joints.

Keywords: adhesive bonded joints, Hopkinson bars, out-of-plane tests, dynamic compression properties, damage mechanisms, heat dissipation

Procedia PDF Downloads 212

Authors: Shehnila Kanwal

Abstract:

Murree-Kotli Sattian-Kahta National Park (MKKNP) was declared in 2009. However, not much is known about the diversity and relative abundance of the mammalian fauna of this park. In the current study, we used field sign survey and infrared camera trapping techniques to get an insight into the diversity of mammalian species and their relative abundance. We conducted field surveys in different areas of the park at various elevations from April 2023 up to March 2024 to record the field signs (scats, pug marks etc.) of the mammals’ species; in addition, we deployed a total of 22 infrared trail camera traps in different areas of the park, for 116 nights. We obtained a total of 5201 photographs using camera trapping. Results of camera trapping coupled with field sign surveys confirmed the presence of a total of twenty-one different mammalian species (large, meso and small mammals) recorded in the study area. The common leopard was recorded at four different sites in the park, with an altitudinal range between 648m-1533m. Distribution of Asiatic jackal and a red fox was recorded positive at all the sites surveyed in the park with an altitudinal range between 498m-1287m and 433m-2049m, respectively. Leopard cats were recorded at two different sites within the altitudinal range between 498m-894m. Jungle cat was recorded at three sites within an altitudinal range between 498m-846. Asian palm civets and small Indian civets were both recorded at three sites. Grey mongoose and small Indian mongoose were recorded at four and three sites. We also collected a total of 75 scats of different mammal species in the park to further confirm their occurrence. For the Indian pangolin, we recorded three field burrows at two different sites. Diversity index (H’=2.369960) and species evenness (E=0.81995) were calculated. Analysis of data revealed that wild boar (Sus sucrofa) was the most abundant species in the park; most of the mammal species were found nocturnal; these remain active from dusk throughout the night, and some of them remain active at dawn time. Leopard and Asian palm civets were highly overlapping species in the study area. Their temporal activity pattern overlapped 61%. Barking deer and Indian crested porcupine were also found to be nocturnal species they remained active throughout the night.

Keywords: MKKNP, diversity, abundance, evenness, distribution, mammals, overlapped

Procedia PDF Downloads 18

Authors: Juan Sebastian Huertas Piedrahita, Jaime Arturo Lopez Duque, Eduardo Luis Perez Londoño, Julián S. Rodríguez

Abstract:

The process of generating computer vision is called artificial vision. The artificial vision is a branch of artificial intelligence that allows the obtaining, processing, and analysis of any type of information especially the ones obtained through digital images. Actually the artificial vision is used in manufacturing areas for quality control and production, as these processes can be realized through counting algorithms, positioning, and recognition of objects that can be measured by a single camera (or more). On the other hand, the companies use assembly lines formed by conveyor systems with actuators on them for moving pieces from one location to another in their production. These devices must be previously programmed for their good performance and must have a programmed logic routine. Nowadays the production is the main target of every industry, quality, and the fast elaboration of the different stages and processes in the chain of production of any product or service being offered. The principal base of this project is to program a computer that recognizes geometric figures (circle, square, and triangle) through a camera, each one with a different color and link it with a group of conveyor systems to organize the mentioned figures in cubicles, which differ from one another also by having different colors. This project bases on artificial vision, therefore the methodology needed to develop this project must be strict, this one is detailed below: 1. Methodology: 1.1 The software used in this project is QT Creator which is linked with Open CV libraries. Together, these tools perform to realize the respective program to identify colors and forms directly from the camera to the computer. 1.2 Imagery acquisition: To start using the libraries of Open CV is necessary to acquire images, which can be captured by a computer’s web camera or a different specialized camera. 1.3 The recognition of RGB colors is realized by code, crossing the matrices of the captured images and comparing pixels, identifying the primary colors which are red, green, and blue. 1.4 To detect forms it is necessary to realize the segmentation of the images, so the first step is converting the image from RGB to grayscale, to work with the dark tones of the image, then the image is binarized which means having the figure of the image in a white tone with a black background. Finally, we find the contours of the figure in the image to detect the quantity of edges to identify which figure it is. 1.5 After the color and figure have been identified, the program links with the conveyor systems, which through the actuators will classify the figures in their respective cubicles. Conclusions: The Open CV library is a useful tool for projects in which an interface between a computer and the environment is required since the camera obtains external characteristics and realizes any process. With the program for this project any type of assembly line can be optimized because images from the environment can be obtained and the process would be more accurate.

Keywords: artificial intelligence, artificial vision, binarized, grayscale, images, RGB

Procedia PDF Downloads 378

Authors: Ege Can Onal, Enver Ersen, Meltem Elitas

Abstract:

Laparoscopy is a surgical operation, well known as keyhole surgery. The operation is performed through small holes, hence, scars of a patient become much smaller, patients can recover in a short time and the hospital stay becomes shorter in comparison to an open surgery. Several tools are used at laparoscopic operations; among them, the laparoscope has a crucial role. It provides the vision during the operation, which will be the main focus in here. Since the operation area is very small, motion of the surgical tools might be limited in laparoscopic operations compared to traditional surgeries. To overcome this limitation, most of the laparoscopic tools have become more precise, dexterous, multi-functional or automated. Here, we present a robotic-assisted laparoscope that is controlled with pedals directly by a surgeon. Thus, the movement of the laparoscope might be controlled better, so there will not be a need to calibrate the camera during the operation. The need for an assistant that controls the movement of the laparoscope will be eliminated. The duration of the laparoscopic operation might be shorter since the surgeon will directly operate the camera.

Keywords: laparoscope, laparoscopy, low-cost, minimally invasive surgery, robotic-assisted surgery

Procedia PDF Downloads 342

Authors: David Tena-Gago, Jose M. Alcaraz Calero, Qi Wang

Abstract:

The evolution and integration of automated vehicles have become more and more tangible in recent years. State-of-the-art technological advances in the field of camera-based Artificial Intelligence (AI) and computer vision greatly favor the performance and reliability of the Advanced Driver Assistance System (ADAS), leading to a greater knowledge of vehicular operation and resembling human behavior. However, the exclusive use of this technology still seems insufficient to control vehicular operation at 100%. To reveal the degree of accuracy of the current camera-based automated driving AI modules, this paper studies the structure and behavior of one of the main solutions in a controlled testing environment. The results obtained clearly outline the lack of reliability when using exclusively the AI model in the perception stage, thereby entailing using additional complementary sensors to improve its safety and performance.

Keywords: accuracy assessment, AI-driven mobility, artificial intelligence, automated vehicles

Procedia PDF Downloads 113

Authors: (Ted) Edward Holmberg, Mahdi Abdelguerfi, Elias Ioup

Abstract:

This research details a coverage measure for evaluating the effectiveness of observer node placements in a spatial bipartite network. This coverage measure can be used to optimize the configuration of stationary or mobile spatially oriented observer nodes, or a hybrid of the two, over time in order to fully utilize their capabilities. To demonstrate the practical application of this approach, we construct a SpatioTemporal Bipartite Network (STBN) using real-time crime center (RTCC) camera nodes and NOPD calls for service (CFS) event nodes from New Orleans, La (NOLA). We use the coverage measure to identify optimal placements for moving mobile RTCC camera vans to improve coverage of vulnerable areas based on temporal patterns.

Keywords: coverage measure, mobile node dynamics, Monte Carlo simulation, observer nodes, observable nodes, spatiotemporal bipartite knowledge graph, temporal spatial analysis

Procedia PDF Downloads 113

Authors: Madhavi T., Sireesha Sakhamuri, Hema Sri A., Harika K.

Abstract:

Today as technological advancements are taking place, these advancements are being used by the armed forces to reduce the risk of their losses and to defeat their enemies. The development of sophisticated technology relies mostly on the use of high- tech weapons or machinery. Robotics is one of the hot spheres of the modern age in which nations concentrate on the state of war and peace for military purposes. They have been in use for demining and rescue operations for some time now but are being propelled by using them for combat and spy missions. This project focuses on creating a LoRa-based spying robot with a wireless IP camera attached to it that can rising the human target. This robot transmits the signal via an IP camera to the base station. One of this project’s major applications can be analyzed using a PC that can be used to control the robot’s movement. The robot sends the signal through the LoRa transceiver at the base station to the LoRa transceiver mounted on the robot. With this function, the, robot can relay videos in real- time along with anti-collision capabilities and the enemies in the war zone cannot recognize them. More importantly, this project focuses on increasing communication using LoRa.

Keywords: lora, IP cam, metal detector, laser shoot

Procedia PDF Downloads 111

Authors: N. Chetty, K. Singh

Abstract:

In this work, the relationship between the melanin content in a tissue and subsequent absorption of light through that tissue was determined using a digital camera. This technique proved to be simple, cost effective, efficient and reliable. Tissue phantom samples were created using milk and soy sauce to simulate the optical properties of melanin content in human tissue. Increasing the concentration of soy sauce in the milk correlated to an increase in melanin content of an individual. Two methods were employed to measure the light transmitted through the sample. The first was direct measurement of the transmitted intensity using a conventional lux meter. The second method involved correctly calibrating an ordinary digital camera and using image analysis software to calculate the transmitted intensity through the phantom. The results from these methods were then graphically compared to the theoretical relationship between the intensity of transmitted light and the concentration of absorbers in the sample. Conclusions were then drawn about the effectiveness and efficiency of these low cost methods.

Keywords: tissue phantoms, scattering coefficient, albedo, low-cost method

Procedia PDF Downloads 271

Authors: James Purnama, Riri Fitri Sari

Abstract:

We investigate the machine learning algorithm selection problem in the term of a depth image based eye gaze estimation, with respect to its essential difficulty in reducing the number of required training samples and duration time of training. Statistics based prediction accuracy are increasingly used to assess and evaluate prediction or estimation in gaze estimation. This article evaluates Root Mean Squared Error (RMSE) and R-Squared statistical analysis to assess machine learning methods on depth camera data for gaze estimation. There are 4 machines learning methods have been evaluated: Random Forest Regression, Regression Tree, Support Vector Machine (SVM), and Linear Regression. The experiment results show that the Random Forest Regression has the lowest RMSE and the highest R-Squared, which means that it is the best among other methods.

Keywords: gaze estimation, gaze tracking, eye tracking, kinect, regression model, orange python

Procedia PDF Downloads 538

Authors: Lukáš Klein, Karel Žídek

Abstract:

Hyperspectral imaging counts among the most frequently used multidimensional sensing methods. While there are many approaches to capturing a hyperspectral data cube, optical compression is emerging as a valuable tool to reduce the setup complexity and the amount of data storage needed. Hyperspectral compressive imagers have been created in the past; however, they have primarily focused on relatively narrow sections of the electromagnetic spectrum. A broader spectral study of samples can provide helpful information, especially for applications involving the harmonic generation and advanced material characterizations. We demonstrate a broadband hyperspectral microscope based on the single-pixel camera principle. Captured spatially encoded data are processed to reconstruct a hyperspectral cube in a combined visible and near-infrared spectrum (from 400 to 2500 nm). Hyperspectral cubes can be reconstructed with a spectral resolution of up to 3 nm and spatial resolution of up to 7 µm (subject to diffraction) with a high compressive ratio.

Keywords: compressive imaging, hyperspectral imaging, near-infrared spectrum, single-pixel camera, visible spectrum

Procedia PDF Downloads 89

Authors: Sandro Bartolini, Alessandro Mecocci, Alessio Medaglini

Abstract:

The growing popularity of systems based on unmanned aerial vehicles (UAVs) is highlighting their vulnerability, particularly in relation to the positioning system used. Typically, UAV architectures use the civilian GPS, which is exposed to a number of different attacks, such as jamming or spoofing. This is why it is important to develop alternative methodologies to accurately estimate the actual UAV position without relying on GPS measurements only. In this paper, we propose a position estimate method for UAVs based on monocular visual odometry. We have developed a flight control system capable of keeping track of the entire trajectory travelled, with a reduced dependency on the availability of GPS signals. Moreover, the simplicity of the developed solution makes it applicable to a wide range of commercial drones. The final goal is to allow for safer flights in all conditions, even under cyber-attacks trying to deceive the drone.

Keywords: visual odometry, autonomous uav, position measurement, autonomous outdoor flight

Procedia PDF Downloads 217

Authors: Azam Fazelpour, Saeed R. Dehghani, Vlastimil Masek, Yuri S. Muzychka

Abstract:

The study uses an image analysis technique employing thermal imaging to measure the percentage of areas with various temperatures on a freezing surface. An image segmentation method using threshold values is applied to a sequence of image recording the freezing process. The phenomenon is transient and temperatures vary fast to reach the freezing point and complete the freezing process. Freezing salt water is subjected to the salt rejection that makes the freezing point dynamic and dependent on the salinity at the phase interface. For a specific area of freezing, nucleation starts from one side and end to another side, which causes a dynamic and transient temperature in that area. Thermal cameras are able to reveal a difference in temperature due to their sensitivity to infrared radiance. Using Experimental setup, a video is recorded by a thermal camera to monitor radiance and temperatures during the freezing process. Image processing techniques are applied to all frames to detect and classify temperatures on the surface. Image processing segmentation method is used to find contours with same temperatures on the icing surface. Each segment is obtained using the temperature range appeared in the image and correspond pixel values in the image. Using the contours extracted from image and camera parameters, stratified areas with different temperatures are calculated. To observe temperature contours on the icing surface using the thermal camera, the salt water sample is dropped on a cold surface with the temperature of -20°C. A thermal video is recorded for 2 minutes to observe the temperature field. Examining the results obtained by the method and the experimental observations verifies the accuracy and applicability of the method.

Keywords: ice contour boundary, image processing, image segmentation, salt ice, thermal image

Procedia PDF Downloads 320

Authors: Chang Liu, John Nash, Stephen D. Prior

Abstract:

This paper presents the design and implementation details of a complete unmanned aerial system (UAS) based on commercial-off-the-shelf (COTS) components, focusing on safety, security, search and rescue scenarios in GPS-denied environments. In particular, the aerial platform is capable of semi-autonomously navigating through extremely low-light, GPS-denied indoor environments based on onboard sensors only, including a downward-facing optical flow camera. Besides, an additional low-cost payload camera system is developed to stream both infrared video and visible light video to a ground station in real-time, for the purpose of detecting sign of life and hidden humans. The total cost of the complete system is estimated to be $1150, and the effectiveness of the system has been tested and validated in practical scenarios.

Keywords: unmanned aerial system, commercial-off-the-shelf, extremely low-light, GPS-denied, optical flow, infrared video

Procedia PDF Downloads 327

Authors: S. Maleczek, K. Drabczyk, L. Bogdan, A. Iwan

Abstract:

It is known that for industrial applications using solar panel constructed from silicon solar cells require high-efficiency performance. One of the main problems in solar panels is different mechanical and structural defects, causing the decrease of generated power. To analyse defects in solar cells, various techniques are used. However, the thermal imaging is fast and simple method for locating defects. The main goal of this work was to analyze defects in constructed flexible silicon photovoltaic modules via thermal imaging and electroluminescence method. This work is realized for the GEKON project (No. GEKON2/O4/268473/23/2016) sponsored by The National Centre for Research and Development and The National Fund for Environmental Protection and Water Management. Thermal behavior was observed using thermographic camera (VIGOcam v50, VIGO System S.A, Poland) using a DC conventional source. Electroluminescence was observed by Steinbeis Center Photovoltaics (Stuttgart, Germany) equipped with a camera, in which there is a Si-CCD, 16 Mpix detector Kodak KAF-16803type. The camera has a typical spectral response in the range 350 - 1100 nm with a maximum QE of 60 % at 550 nm. In our work commercial silicon solar cells with the size 156 × 156 mm were cut for nine parts (called single solar cells) and used to create photovoltaic modules with the size of 160 × 70 cm (containing about 80 single solar cells). Flexible silicon photovoltaic modules on polyamides or polyester fabric were constructed and investigated taking into consideration anomalies on the surface of modules. Thermal imaging provided evidence of visible voltage-activated conduction. In electro-luminescence images, two regions are noticeable: darker, where solar cell is inactive and brighter corresponding with correctly working photovoltaic cells. The electroluminescence method is non-destructive and gives greater resolution of images thereby allowing a more precise evaluation of microcracks of solar cell after lamination process. Our study showed good correlations between defects observed by thermal imaging and electroluminescence. Finally, we can conclude that the thermographic examination of large scale photovoltaic modules allows us the fast, simple and inexpensive localization of defects at the single solar cells and modules. Moreover, thermographic camera was also useful to detection electrical interconnection between single solar cells.

Keywords: electro-luminescence, flexible devices, silicon solar cells, thermal imaging

Procedia PDF Downloads 315

Authors: Ahmad Bilal

Abstract:

Digital and convergent media can be helpful in terms of acquiring film production skills and knowledge, and it has also reduced the cost of production. Thus, allowing filmmakers greater opportunities and access to the medium of film. Both these dimensions of new and convergent media have been challenging the established cinema of Pakistan, as traditionally, it has been controlled by the authorities through censorship policies. The use of the smartphone as a movie camera, editing machine, and a transmitter can further challenge the control in a postcolonial society. To explore the impact of new and convergent media on the art of filmmaking, a film 'Sohni Dharti: An untrue story' is produced. It is shot both on a smartphone and a Digital Single Lens Reflex Camera (DSLR), with almost zero budgets. It is distributed through Vimeo from Pakistan. This process reveals how the technologies that are available today, and the increased knowledge of film production that they bring, allow a more inclusive experience of the film production and distribution. At the same time, however, it also discloses the limitations that accompany new technologies within the context of a postcolonial society. This paper will investigate the role of technology to bring filmmaking at a level of pencil and paper.

Keywords: convergent media, filmmaking, smartphone, Pakistan

Procedia PDF Downloads 280

Authors: Chad Goldsworthy, B. Rajeswari Matam

Abstract:

The use of deep learning for species identification in camera trap images has revolutionised our ability to study, conserve and monitor species in a highly efficient and unobtrusive manner, with state-of-the-art models achieving accuracies surpassing the accuracy of manual human classification. The high imbalance of camera trap datasets, however, results in poor accuracies for minority (rare or endangered) species due to their relative insignificance to the overall model accuracy. This paper investigates the use of Focal Loss, in comparison to the traditional Cross Entropy Loss function, to improve the identification of minority species in the “255 Bird Species” dataset from Kaggle. The results show that, although Focal Loss slightly decreased the accuracy of the majority species, it was able to increase the F1-score by 0.06 and improve the identification of the bottom two, five and ten (minority) species by 37.5%, 15.7% and 10.8%, respectively, as well as resulting in an improved overall accuracy of 2.96%.

Keywords: convolutional neural networks, data imbalance, deep learning, focal loss, species classification, wildlife conservation

Procedia PDF Downloads 190

Authors: Murad Almadani, Khalil Abu-Hantash, Xinyu Wang, Herbert Jelinek, Kinda Khalaf

Abstract:

Gait analysis is used by healthcare professionals as a tool to gain a better understanding of the movement impairment and track progress. In most circumstances, monitoring patients in their real-life environments with low-cost equipment such as cameras and wearable sensors is more important. Inertial sensors, on the other hand, cannot provide enough information on angular dynamics. This research offers a method for tracking 2D joint coordinates using cutting-edge vision algorithms and a single RGB camera. We provide an end-to-end comprehensive deep learning pipeline for marker-less gait parameter estimation, which, to our knowledge, has never been done before. To make our pipeline function in real-time for real-world applications, we leverage the AlphaPose human posture prediction model and a deep learning transformer. We tested our approach on the well-known GPJATK dataset, which produces promising results.

Keywords: gait analysis, human pose estimation, deep learning, real time gait estimation, AlphaPose, transformer

Procedia PDF Downloads 118

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: Amit Kumar Happy

Abstract:

This paper is motivated by the importance of multi-sensor image fusion with a specific focus on infrared (IR) and visual image (VI) fusion for various applications, including military reconnaissance. Image fusion can be defined as the process of combining two or more source images into a single composite image with extended information content that improves visual perception or feature extraction. These images can be from different modalities like visible camera & IR thermal imager. While visible images are captured by reflected radiations in the visible spectrum, the thermal images are formed from thermal radiation (infrared) that may be reflected or self-emitted. A digital color camera captures the visible source image, and a thermal infrared camera acquires the thermal source image. In this paper, some image fusion algorithms based upon multi-scale transform (MST) and region-based selection rule with consistency verification have been proposed and presented. This research includes the implementation of the proposed image fusion algorithm in MATLAB along with a comparative analysis to decide the optimum number of levels for MST and the coefficient fusion rule. The results are presented, and several commonly used evaluation metrics are used to assess the suggested method's validity. Experiments show that the proposed approach is capable of producing good fusion results. While deploying our image fusion algorithm approaches, we observe several challenges from the popular image fusion methods. While high computational cost and complex processing steps of image fusion algorithms provide accurate fused results, they also make it hard to become deployed in systems and applications that require a real-time operation, high flexibility, and low computation ability. So, the methods presented in this paper offer good results with minimum time complexity.

Keywords: image fusion, IR thermal imager, multi-sensor, multi-scale transform

Procedia PDF Downloads 115

Authors: Yumee Kim, Seung-Guk Hyeon, Kukjin Chun

Abstract:

An adjustable aperture using a liquid crystal is proposed for real-time range detection and obtaining images simultaneously. The adjustable aperture operates as two types of aperture stops which can create two different Depth of Field images. By analyzing these two images, the distance can be extracted from camera to object. Initially, the aperture stop has large size with zero voltage. When the input voltage is applied, the aperture stop transfer to smaller size by orientational transition of liquid crystal molecules in the device. The diameter of aperture stop is 1.94mm and 1.06mm. The proposed device has low driving voltage of 7.0V and fast response time of 6.22m. Compact size aperture of 6×6×1.1 mm3 is assembled in conventional camera which contain 1/3” HD image sensor and focal length of 3.3mm that can be used in autonomous. The measured range was up to 5m. The adjustable aperture has high stability due to no mechanically moving parts. This range sensor can be applied to the various field of 3D depth map application which is the Advanced Driving Assistance System (ADAS), drones and manufacturing machine.

Keywords: adjustable aperture, dual aperture, liquid crystal, ranging and imaging, ADAS, range sensor

Procedia PDF Downloads 381

Authors: Mohkammad Nur Cahyadi, Hepi Hapsari Handayani, Agus Budi Raharjo, Ronny Mardianto, Daud Wahyu Imani, Arizal Bawazir, Luki Adi Triawan

Abstract:

PDAM (local water company) determines customer charges by considering the customer's building or house. Charges determination significantly affects PDAM income and customer costs because the PDAM applies a subsidy policy for customers classified as small households. Periodic updates are needed so that pricing is in line with the target. A thorough customer survey in Surabaya is needed to update customer building data. However, the survey that has been carried out so far has been by deploying officers to conduct one-by-one surveys for each PDAM customer. Surveys with this method require a lot of effort and cost. For this reason, this research offers a technology called moblie mapping, a mapping method that is more efficient in terms of time and cost. The use of this tool is also quite simple, where the device will be installed in the car so that it can record the surrounding buildings while the car is running. Mobile mapping technology generally uses lidar sensors equipped with GNSS, but this technology requires high costs. In overcoming this problem, this research develops low-cost mobile mapping technology using a webcam camera sensor added to the GNSS and IMU sensors. The camera used has specifications of 3MP with a resolution of 720 and a diagonal field of view of 78⁰. The principle of this invention is to integrate four camera sensors, a GNSS webcam, and GPS to acquire photo data, which is equipped with location data (latitude, longitude) and IMU (roll, pitch, yaw). This device is also equipped with a tripod and a vacuum cleaner to attach to the car's roof so it doesn't fall off while running. The output data from this technology will be analyzed with artificial intelligence to reduce similar data (Cosine Similarity) and then classify building types. Data reduction is used to eliminate similar data and maintain the image that displays the complete house so that it can be processed for later classification of buildings. The AI method used is transfer learning by utilizing a trained model named VGG-16. From the analysis of similarity data, it was found that the data reduction reached 50%. Then georeferencing is done using the Google Maps API to get address information according to the coordinates in the data. After that, geographic join is done to link survey data with customer data already owned by PDAM Surya Sembada Surabaya.

Keywords: mobile mapping, GNSS, IMU, similarity, classification

Procedia PDF Downloads 84

Authors: C. Gebauer, C. Henke, R. Vossen

Abstract:

Within the Mohamed Bin Zayed International Robotics Challenge (MBZIRC) 2020, a team of unmanned aerial vehicles (UAV) is used to capture intruder drones by physical interaction. The challenge is motivated by UAV safety. The purpose of this work is to investigate the agility of a quadcopter being controlled visually. The aim is to track and follow a highly dynamic target, e.g., an intruder quadcopter. The following is realized in close range and the opponent has a velocity of up to 10 m/s. Additional limitations are given by the hardware itself, where only monocular vision is present, and no additional knowledge about the targets state is available. An image based visual servoing (IBVS) approach is applied in combination with a Linear Quadratic Regulator (LQR). The IBVS is integrated into the LQR and an optimal trajectory is computed within the projected three-dimensional image-space. The approach has been evaluated on real quadcopter systems in different flight scenarios to demonstrate the system's stability.

Keywords: image based visual servoing, quadcopter, dynamic object tracking, linear-quadratic regulator

Procedia PDF Downloads 149

Authors: Shih-Hao Chen, Chi-Wai Chow

Abstract:

Multiple-input and multiple-output (MIMO) scheme can extend the transmission capacity for the light-emitting-diode (LED) visible light communication (VLC) system. The MIMO VLC system using the popular mobile-phone camera as the optical receiver (Rx) to receive MIMO signal from n x n Red-Green-Blue (RGB) LED array is desirable. The key step of decoding the received RGB LED array signals is detecting the direction of received array signals. If the LED transmitter (Tx) is rotated, the signal may not be received correctly and cause an error in the received signal. In this work, we propose and demonstrate a novel hierarchical transmission scheme which can reduce the computation complexity of rotation detection in LED array VLC system. We use the n x n RGB LED array as the MIMO Tx. A novel two dimension Hadamard coding scheme is proposed and demonstrated. The detection correction rate is above 95% in the indoor usage distance. Experimental results confirm the feasibility of the proposed scheme.

Keywords: Visible Light Communication (VLC), Multiple-input and multiple-output (MIMO), Red-Green-Blue (RGB), Hadamard coding scheme

Procedia PDF Downloads 419

Authors: Giuseppe Spampinato, Salvatore Curti, Ivana Guarneri, Arcangelo Bruna

Abstract:

This document describes an advanced system and methodology for Cross Traffic Alert (CTA), able to detect vehicles that move into the vehicle driving path from the left or right side. The camera is supposed to be not only on a vehicle still, e.g. at a traffic light or at an intersection, but also moving slowly, e.g. in a car park. In all of the aforementioned conditions, a driver’s short loss of concentration or distraction can easily lead to a serious accident. A valid support to avoid these kinds of car crashes is represented by the proposed system. It is an extension of our previous work, related to a clustering system, which only works on fixed cameras. Just a vanish point calculation and simple optical flow filtering, to eliminate motion vectors due to the car relative movement, is performed to let the system achieve high performances with different scenarios, cameras and resolutions. The proposed system just uses as input the optical flow, which is hardware implemented in the proposed platform and since the elaboration of the whole system is really speed and power consumption, it is inserted directly in the camera framework, allowing to execute all the processing in real-time.

Keywords: clustering, cross traffic alert, optical flow, real time, vanishing point

Procedia PDF Downloads 203

Authors: Thilini M. Yatanwala

Abstract:

CCTV (Closed-Circuit-Television) Surveillance System is being used in public places over decades and a large variety of data is being produced every moment. However, most of the CCTV data is stored in isolation without having integrity. As a result, identification of the behavior of suspicious people along with their location has become strenuous. This research was conducted to acquire more accurate and reliable timely information from the CCTV video records. The implemented system can identify human objects in public places based on outfit colors. Inter-process communication technologies were used to implement the CCTV camera network to track people in the premises. The research was conducted in three stages and in the first stage human objects were filtered from other movable objects available in public places. In the second stage people were uniquely identified based on their outfit colors and in the third stage an individual was continuously tracked in the CCTV network. A face detection algorithm was implemented using cascade classifier based on the training model to detect human objects. HAAR feature based two-dimensional convolution operator was introduced to identify features of the human face such as region of eyes, region of nose and bridge of the nose based on darkness and lightness of facial area. In the second stage outfit colors of human objects were analyzed by dividing the area into upper left, upper right, lower left, lower right of the body. Mean color, mod color and standard deviation of each area were extracted as crucial factors to uniquely identify human object using histogram based approach. Color based measurements were written in to XML files and separate directories were maintained to store XML files related to each camera according to time stamp. As the third stage of the approach, inter-process communication techniques were used to implement an acknowledgement based CCTV camera network to continuously track individuals in a network of cameras. Real time analysis of XML files generated in each camera can determine the path of individual to monitor full activity sequence. Higher efficiency was achieved by sending and receiving acknowledgments only among adjacent cameras. Suspicious incidents such as a person staying in a sensitive area for a longer period or a person disappeared from the camera coverage can be detected in this approach. The system was tested for 150 people with the accuracy level of 82%. However, this approach was unable to produce expected results in the presence of group of people wearing similar type of outfits. This approach can be applied to any existing camera network without changing the physical arrangement of CCTV cameras. The study of human identification and suspicious incident detection using outfit color analysis can achieve higher level of accuracy and the project will be continued by integrating motion and gait feature analysis techniques to derive more information from CCTV videos.

Keywords: CCTV surveillance, human detection and identification, image processing, inter-process communication, security, suspicious detection

Procedia PDF Downloads 181

Authors: Mingxin Li, Liya Ni

Abstract:

Autonomous parking is a valuable feature applicable to many robotics applications such as tour guide robots, UV sanitizing robots, food delivery robots, and warehouse robots. With auto-parking, the robot will be able to park at the charging zone and charge itself without human intervention. As compared to self-driving vehicles, auto-parking is more challenging for a small-scale mobile robot only equipped with a front camera due to the camera view limited by the robot’s height and the narrow Field of View (FOV) of the inexpensive camera. In this research, auto-parking of a small-scale mobile robot with a front camera only was achieved in a four-step process: Firstly, transfer learning was performed on the AlexNet, a popular pre-trained convolutional neural network (CNN). It was trained with 150 pictures of empty parking slots and 150 pictures of occupied parking slots from the view angle of a small-scale robot. The dataset of images was divided into a group of 70% images for training and the remaining 30% images for validation. An average success rate of 95% was achieved. Secondly, the image of detected empty parking space was processed with edge detection followed by the computation of parametric representations of the boundary lines using the Hough Transform algorithm. Thirdly, the positions of the entrance point and center of available parking space were predicted based on the robot kinematic model as the robot was driving closer to the parking space because the boundary lines disappeared partially or completely from its camera view due to the height and FOV limitations. The robot used its wheel speeds to compute the positions of the parking space with respect to its changing local frame as it moved along, based on its kinematic model. Lastly, the predicted entrance point of the parking space was used as the reference for the motion control of the robot until it was replaced by the actual center when it became visible again by the robot. The linear and angular velocities of the robot chassis center were computed based on the error between the current chassis center and the reference point. Then the left and right wheel speeds were obtained using inverse kinematics and sent to the motor driver. The above-mentioned four subtasks were all successfully accomplished, with the transformed learning, image processing, and target prediction performed in MATLAB, while the motion control and image capture conducted on a self-built small scale differential drive mobile robot. The small-scale robot employs a Raspberry Pi board, a Pi camera, an L298N dual H-bridge motor driver, a USB power module, a power bank, four wheels, and a chassis. Future research includes three areas: the integration of all four subsystems into one hardware/software platform with the upgrade to an Nvidia Jetson Nano board that provides superior performance for deep learning and image processing; more testing and validation on the identification of available parking space and its boundary lines; improvement of performance after the hardware/software integration is completed.

Keywords: autonomous parking, convolutional neural network, image processing, kinematics-based prediction, transfer learning

Procedia PDF Downloads 132