Search results for: foggy image restoration

Authors: Manaranjan Pradhan, Shailaja Grover, U. Dinesh Kumar

Abstract:

Farming community in India, as well as other parts of the world, is one of the highly stressed communities due to reasons such as increasing input costs (cost of seeds, fertilizers, pesticide), droughts, reduced revenue leading to farmer suicides. Lack of integrated farm advisory system in India adds to the farmers problems. Farmers need right information during the early stages of crop’s lifecycle to prevent damage and loss in revenue. In this paper, we use deep learning techniques to develop an early warning system for detection of crop diseases using images taken by farmers using their smart phone. The research work leads to building a smart assistant using analytics and big data which could help the farmers with early diagnosis of the crop diseases and corrective actions. The classical approach for crop disease management has been to identify diseases at crop level. Recently, ImageNet Classification using the convolutional neural network (CNN) has been successfully used to identify diseases at individual plant level. Our model uses convolution filters, max pooling, dense layers and dropouts (to avoid overfitting). The models are built for binary classification (healthy or not healthy) and multi class classification (identifying which disease). Transfer learning is used to modify the weights of parameters learnt through ImageNet dataset and apply them on crop diseases, which reduces number of epochs to learn. One shot learning is used to learn from very few images, while data augmentation techniques are used to improve accuracy with images taken from farms by using techniques such as rotation, zoom, shift and blurred images. Models built using combination of these techniques are more robust for deploying in the real world. Our model is validated using tomato crop. In India, tomato is affected by 10 different diseases. Our model achieves an accuracy of more than 95% in correctly classifying the diseases. The main contribution of our research is to create a personal assistant for farmers for managing plant disease, although the model was validated using tomato crop, it can be easily extended to other crops. The advancement of technology in computing and availability of large data has made possible the success of deep learning applications in computer vision, natural language processing, image recognition, etc. With these robust models and huge smartphone penetration, feasibility of implementation of these models is high resulting in timely advise to the farmers and thus increasing the farmers' income and reducing the input costs.

Keywords: analytics in agriculture, CNN, crop disease detection, data augmentation, image recognition, one shot learning, transfer learning

Procedia PDF Downloads 117

Authors: Lumindu Dilumka, Ranaweera I. D., Sudusinghe S. P., Sanduni Kanchana A. M. K.

Abstract:

This study was motivated by the challenges experienced by parents and guardians in ensuring the safety of children in cyberspace. In the last two or three years, online education has become very popular all over the world due to the Covid 19 pandemic. Therefore, parents, guardians and teachers must ensure the safety of children in cyberspace. Children are more likely to go astray and there are plenty of online programs are waiting to get them on the wrong track and also, children who are engaging in the online education can be distracted at any moment. Therefore, parents should keep a close check on their children's online activity. Apart from that, due to the unawareness of children, they tempt to share their sensitive information, causing a chance of being a victim of phishing attacks from outsiders. These problems can be overcome through the proposed web-based system. We use feature extraction, web tracking and analysis mechanisms, image processing and name entity recognition to implement this web-based system.

Keywords: online education, cyber bullying, social media, face recognition, web tracker, privacy data

Procedia PDF Downloads 86

Authors: Diana Islam, Juan Fang, Vito Fanelli, Bing Han, Julie Khang, Jianfeng Wu, Arthur S. Slutsky, Haibo Zhang

Abstract:

Introduction: MSC delivery in preclinical models of ARDS has demonstrated significant improvements in lung function and recovery from acute injury. However, the role of MSC delivery in ARDS associated pulmonary fibrosis is not well understood. Some animal studies using bleomycin, asbestos, and silica-induced pulmonary fibrosis show that MSC delivery can suppress fibrosis. While other animal studies using radiation induced pulmonary fibrosis, liver, and kidney fibrosis models show that MSC delivery can contribute to fibrosis. Hypothesis: The beneficial and deleterious effects of MSC in ARDS are modulated by the lung microenvironment at the time of MSC delivery. Methods: To induce ARDS a two-hit mouse model of Hydrochloric acid (HCl) aspiration (day 0) and mechanical ventilation (MV) (day 2) was used. HCl and injurious MV generated fibrosis within 14-28 days. 0.5x106 mouse MSCs were delivered (via both intratracheal and intravenous routes) either in the active inflammatory phase (day 2) or during the remodeling phase (day 14) of ARDS (mouse fibroblasts or PBS used as a control). Lung injury accessed using inflammation score and elastance measurement. Pulmonary fibrosis was accessed using histological score, tissue collagen level, and collagen expression. In addition alveolar epithelial (E) and mesenchymal (M) marker expression profile was also measured. All measurements were taken at day 2, 14, and 28. Results: MSC delivery 2 days after HCl exacerbated lung injury and fibrosis compared to HCl alone, while the day 14 delivery showed protective effects. However in the absence of HCl, MSC significantly reduced the injurious MV-induced fibrosis. HCl injury suppressed E markers and up-regulated M markers. MSC delivery 2 days after HCl further amplified M marker expression, indicating their role in myofibroblast proliferation/activation. While with 14-day delivery E marker up-regulation was observed indicating their role in epithelial restoration. Conclusions: Early MSC delivery can be protective of injurious MV. Late MSC delivery during repair phase may also aid in recovery. However, early MSC delivery during the exudative inflammatory phase of HCl-induced ARDS can result in pro-fibrotic profiles. It is critical to understand the interaction between MSC and the lung microenvironment before MSC-based therapies are utilized for ARDS.

Keywords: acute respiratory distress syndrome (ARDS), mesenchymal stem cells (MSC), hydrochloric acid (HCl), mechanical ventilation (MV)

Procedia PDF Downloads 667

Authors: Jaeyoung Lee

Abstract:

Autonomous driving systems require high reliability to provide people with a safe and comfortable driving experience. However, despite the development of a number of vehicle sensors, it is difficult to always provide high perceived performance in driving environments that vary from time to season. The image segmentation method using deep learning, which has recently evolved rapidly, provides high recognition performance in various road environments stably. However, since the system controls a vehicle in real time, a highly complex deep learning network cannot be used due to time and memory constraints. Moreover, efficient networks are optimized for GPU environments, which degrade performance in embedded processor environments equipped simple hardware accelerators. In this paper, a semantic segmentation network, matrix multiplication accelerator network (MMANet), optimized for matrix multiplication accelerator (MMA) on Texas instrument digital signal processors (TI DSP) is proposed to improve the recognition performance of autonomous driving system. The proposed method is designed to maximize the number of layers that can be performed in a limited time to provide reliable driving environment information in real time. First, the number of channels in the activation map is fixed to fit the structure of MMA. By increasing the number of parallel branches, the lack of information caused by fixing the number of channels is resolved. Second, an efficient convolution is selected depending on the size of the activation. Since MMA is a fixed, it may be more efficient for normal convolution than depthwise separable convolution depending on memory access overhead. Thus, a convolution type is decided according to output stride to increase network depth. In addition, memory access time is minimized by processing operations only in L3 cache. Lastly, reliable contexts are extracted using the extended atrous spatial pyramid pooling (ASPP). The suggested method gets stable features from an extended path by increasing the kernel size and accessing consecutive data. In addition, it consists of two ASPPs to obtain high quality contexts using the restored shape without global average pooling paths since the layer uses MMA as a simple adder. To verify the proposed method, an experiment is conducted using perfsim, a timing simulator, and the Cityscapes validation sets. The proposed network can process an image with 640 x 480 resolution for 6.67 ms, so six cameras can be used to identify the surroundings of the vehicle as 20 frame per second (FPS). In addition, it achieves 73.1% mean intersection over union (mIoU) which is the highest recognition rate among embedded networks on the Cityscapes validation set.

Keywords: edge network, embedded network, MMA, matrix multiplication accelerator, semantic segmentation network

Procedia PDF Downloads 128

Authors: Ihsane Modhaffar

Abstract:

Fiber orientation is essential for the physical properties of composite materials. The theoretical parameters of a given reinforcement are usually known and widely used to predict the behavior of the material. In this work, we propose an image processing approach to estimate true principal directions and fiber orientation during injection molding processes of short fiber reinforced thermoplastics. Generally, a group of fibers are described in terms of probability distribution function or orientation tensor. Numerical techniques for the prediction of fiber orientation are also considered for concentrated situations. The flow was considered to be incompressible, and behave as Newtonian fluid containing suspensions of short-fibers. The governing equations, of this problem are: the continuity, the momentum and the energy. The obtained results were compared to available experimental findings. A good agreement between the numerical results and the experimental data was achieved.

Keywords: injection, composites, short-fiber reinforced thermoplastics, fiber orientation, incompressible fluid, numerical simulation

Procedia PDF Downloads 530

Authors: I. Gkolfinopoulos, N. Chijiwa

Abstract:

To evaluate the remaining capacity in concrete tensioned members, it is important to accurately estimate damage in precast concrete tendons. In this research Johnson-Cook model and damage parameters of high-strength steel material were calculated by static and dynamic uniaxial tensile tests. Replication of experimental results was achieved through finite element analysis for both single 8-noded three-dimensional element as well as the full-scale dob-bone shaped model and relevant model parameters are proposed. Finally, simulation results in terms of strain and deformation were verified using digital image correlation analysis.

Keywords: DIC analysis, Johnson-Cook, quasi-static, dynamic, rupture, tendon

Procedia PDF Downloads 145

Authors: Rakkrit Duangsoithong, Jermphiphut Jaruenpunyasak, Alba Garcia

Abstract:

The use of foot recognition can be applied in many medical fields such as the gait pattern analysis and the knee exercises of patients in rehabilitation. Generally, a camera-based foot recognition system is intended to capture a patient image in a controlled room and background to recognize the foot in the limited views. However, this system can be inconvenient to monitor the knee exercises at home. In order to overcome these problems, this paper proposes to use the deep learning method using Convolutional Neural Networks (CNNs) for foot recognition. The results are compared with the traditional classification method using LBP and HOG features with kNN and SVM classifiers. According to the results, deep learning method provides better accuracy but with higher complexity to recognize the foot images from online databases than the traditional classification method.

Keywords: foot recognition, deep learning, knee rehabilitation, convolutional neural network

Procedia PDF Downloads 160

Authors: Ayalew Yimam Ali

Abstract:

The Y-shaped microchannel system is used to mix up low or high viscosities of different fluids, and the laminar flow with high-viscous water-glycerol fluids makes the mixing at the entrance Y-junction region a challenging issue. Acoustic streaming (AS) is time-average, a steady second-order flow phenomenon that could produce rolling motion in the microchannel by oscillating low-frequency range acoustic transducer by inducing acoustic wave in the flow field is the promising strategy to enhance diffusion mass transfer and mixing performance in laminar flow phenomena. In this study, the 3D trapezoidal Structure has been manufactured with advanced CNC machine cutting tools to produce the molds of trapezoidal structure with the 3D sharp edge tip angles of 30° and 0.3mm spine sharp-edge tip depth from PMMA glass (Polymethylmethacrylate) and the microchannel has been fabricated using PDMS (Polydimethylsiloxane) which could be grown-up longitudinally in Y-junction microchannel mixing region top surface to visualized 3D rolling steady acoustic streaming and mixing performance evaluation using high-viscous miscible fluids. The 3D acoustic streaming flow patterns and mixing enhancement were investigated using the micro-particle image velocimetry (μPIV) technique with different spine depth lengths, channel widths, high volume flow rates, oscillation frequencies, and amplitude. The velocity and vorticity flow fields show that a pair of 3D counter-rotating streaming vortices were created around the trapezoidal spine structure and observing high vorticity maps up to 8 times more than the case without acoustic streaming in Y-junction with the high-viscosity water-glycerol mixture fluids. The mixing experiments were performed by using fluorescent green dye solution with de-ionized water on one inlet side, de-ionized water-glycerol with different mass-weight percentage ratios on the other inlet side of the Y-channel and evaluated its performance with the degree of mixing at different amplitudes, flow rates, frequencies, and spine sharp-tip edge angles using the grayscale value of pixel intensity with MATLAB Software. The degree of mixing (M) characterized was found to significantly improved to 0.96.8% with acoustic streaming from 67.42% without acoustic streaming, in the case of 0.0986 μl/min flow rate, 12kHz frequency and 40V oscillation amplitude at y = 2.26 mm. The results suggested the creation of a new 3D steady streaming rolling motion with a high volume flow rate around the entrance junction mixing region, which promotes the mixing of two similar high-viscosity fluids inside the microchannel, which is unable to mix by the laminar flow with low viscous conditions.

Keywords: nano fabrication, 3D acoustic streaming flow visualization, micro-particle image velocimetry, mixing enhancement

Procedia PDF Downloads 31

Authors: Francisco Bulnes

Abstract:

Through the Fukaya conjecture and the wrapped Floer cohomology, the correspondences between paths in a loop space and states of a wrapping space of states in a Hamiltonian space (the ramification of field in this case is the connection to the operator that goes from TM to T*M) are demonstrated where these last states are corresponding to bosonic extensions of a spectrum of the space-time or direct image of the functor Spec, on space-time. This establishes a distinguished diffeomorphism defined by the mapping from the corresponding loops space to wrapping category of the Floer cohomology complex which furthermore relates in certain proportion D-branes (certain D-modules) with strings. This also gives to place to certain conjecture that establishes equivalences between moduli spaces that can be consigned in a moduli identity taking as space-time the Hitchin moduli space on G, whose dual can be expressed by a factor of a bosonic moduli spaces.

Keywords: Floer cohomology, Fukaya conjecture, Lagrangian submanifolds, quantum topological diffeomorphism

Procedia PDF Downloads 307

Authors: Jeyeon Kim

Abstract:

This paper describes the development of the automatic battery charging device for the rotor wings type unmanned aerial vehicle (UAV) and the positioning method that can be accurately landed on the charging device when landing. The developed automatic battery charging device is considered by simple maintenance, durability, cost and error of the positioning when landing. In order to for the UAV accurately land on the charging device, two kinds of markers (a color marker and a light marker) installed on the charging device is detected by the camera mounted on the UAV. And then, the UAV is controlled so that the detected marker becomes the center of the image and is landed on the device. We conduct the performance evaluation of the proposal positioning method by the outdoor experiments at day and night, and show the effectiveness of the system.

Keywords: unmanned aerial vehicle, automatic battery charging, positioning

Procedia PDF Downloads 361

Authors: Poopak Dehshahri

Abstract:

The applicability of intellectual property rights in the sports industry from the present paper’s perspective includes athletic skills, which are comprised of two parts: athletic movements and athletic methods. Also, the applicability pertaining to the athletes᾽ personality, such as the Name, the Image, the Voice, the Signature and their Shirt Number, are deemed as related to the sports natural persons. Radio and TV broadcasting rights of the sports events, the signs and symbols of the athletic institutions including the sign and symbol, trademark (brand name), the name and the place of residence of the sports clubs, the Sports events and the special sports, special slogan of the sports clubs or sports competitions and the sports clothing design are Included under the athletic institutions᾽ applicability of intellectual property rights.

Keywords: sport industry, intellectual property, sport skills, right to fame, radio and television broadcasting right, sport sign

Procedia PDF Downloads 65

Authors: Peyman Sheikholharam Mashhadi

Abstract:

Difficulties in analyzing real world images in classical image processing and machine vision framework have motivated researchers towards considering the biology-based vision. It is a common belief that mammalian visual cortex has been adapted to the statistics of the real world images through the evolution process. There are two well-known successful models of mammalian visual cortical cells: Independent Component Analysis (ICA) and Independent Subspace Analysis (ISA). In this paper, we statistically analyze the dependencies which remain in the components after applying these models to the natural images. Also, we investigate the response of feature detectors to gratings with various parameters in order to find optimal parameters of the feature detectors. Finally, the selectiveness of feature detectors to phase, in both models is considered.

Keywords: statistics, independent component analysis, independent subspace analysis, phase, natural images

Procedia PDF Downloads 338

Authors: Diego Aguirre, Olac Fuentes

Abstract:

In this paper, we present a taxonomy of weight initialization schemes used in deep learning. We survey the most representative techniques in each class and compare them in terms of overhead cost, convergence rate, and applicability. We also introduce a new weight initialization scheme. In this technique, we perform an initial feedforward pass through the network using an initialization mini-batch. Using statistics obtained from this pass, we initialize the weights of the network, so the following properties are met: 1) weight matrices are orthogonal; 2) ReLU layers produce a predetermined number of non-zero activations; 3) the output produced by each internal layer has a unit variance; 4) weights in the last layer are chosen to minimize the error in the initial mini-batch. We evaluate our method on three popular architectures, and a faster converge rates are achieved on the MNIST, CIFAR-10/100, and ImageNet datasets when compared to state-of-the-art initialization techniques.

Keywords: deep learning, image classification, supervised learning, weight initialization

Procedia PDF Downloads 134

Authors: Marlon M. Hernández Cely, Victor E. C. Baptistella, Oscar M. H. Rodríguez

Abstract:

Particle Image Velocimetry (PIV) is a well-established technique in the field of fluid flow measurement and provides instantaneous velocity fields over global domains. It has been applied to external and internal flows and in single and two-phase flows. Regarding internal flow, works about the application of PIV in annular ducts are scanty. An experimental work is presented, where flow of water is studied in an annular duct of inner diameter of 60 mm and outer diameter of 155 mm and 10.5-m length, with the goal of obtaining detailed velocity measurements. Depending on the flow rates of water, it can be laminar, transitional or turbulent. In this study, the water flow rate was kept at three different values for the annular duct, allowing the analysis of one laminar and two turbulent flows. Velocity fields and statistic quantities of the turbulent flow were calculated.

Keywords: PIV, annular duct, laminar, turbulence, velocity profile

Procedia PDF Downloads 349

Authors: Paulo Fernando Silva Filho, Elcio Hideiti Shiguemori

Abstract:

The selection of specific landmarks for an Unmanned Aerial Vehicles’ Visual Navigation systems based on Automatic Landmark Recognition has significant influence on the precision of the system’s estimated position. At the same time, manual selection of the landmarks does not guarantee a high recognition rate, which would also result on a poor precision. This work aims to develop an automatic landmark selection that will take the image of the flight area and identify the best landmarks to be recognized by the Visual Navigation Landmark Recognition System. The criterion to select a landmark is based on features detected by ORB or AKAZE and edges information on each possible landmark. Results have shown that disposition of possible landmarks is quite different from the human perception.

Keywords: clustering, edges, feature points, landmark selection, X-means

Procedia PDF Downloads 278

Authors: Sara Fayez Fawzy Mikhael

Abstract:

Inclusive education services for students with autism are still developing in Thailand. Although many more children with intellectual disabilities have been attending school since the Thai government enacted the Education for Persons with Disabilities Act in 2008, facilities for students with disabilities and their families are generally inadequate. This comprehensive study used the Attitudes and Preparedness for Teaching Students with Autism Scale (APTSAS) to examine the attitudes and preparedness of 110, elementary teachers in teaching students with autism in the general education setting. Descriptive statistical analyzes showed that the most important factor in the formation of a negative image of teachers with autism is student attitudes. Most teachers also stated that their pre-service training did not prepare them to meet the needs of children with special needs who cannot speak. The study is important and provides directions for improving non-formal teacher education in Thailand.

Keywords: attitude, autism, teachers, thailandsports activates, movement skills, motor skills

Procedia PDF Downloads 60

Authors: C. Hamamura, V. Gialluca

Abstract:

Much of the Eclectricity Distributors has about 70% of its electricity interruptions arising from cause "trees", alone or associated with wind and rain and with or without falling branch and / or trees. This contributes inexorably and significantly to outages, resulting in high costs as compensation in addition to the operation and maintenance costs. On the other hand, there is little data structure and solutions to better organize the trees pruning plan effectively, minimizing costs and environmentally friendly. This work describes the development of an algorithm to provide data of trees associated to power grid. The method is accomplished on several steps using satellite imagery and geographically vectorized grid. A sliding window like approach is performed to seek the area around the grid. The proposed method counted 764 trees on a patch of the grid, which was very close to the 738 trees counted manually. The trees data was used as a part of a larger project that implements a system to optimize tree pruning plan.

Keywords: image pattern recognition, trees pruning, trees recognition, neural network

Procedia PDF Downloads 498

Authors: Yonggu Jang, Jisong Ryu, Woosik Lee

Abstract:

The study aims to address the limitations of existing underground facility exploration equipment in terms of exploration depth range, relative depth measurement, data processing time, and human-centered ground penetrating radar image interpretation. The study proposed the use of 3D absolute positioning technology to develop a precision underground facility exploration system. The aim of this study is to establish a precise exploration system for underground facilities based on 3D absolute positioning technology, which can accurately survey up to a depth of 5m and measure the 3D absolute location of precise underground facilities. The study developed software and hardware technologies to build the precision exploration system. The software technologies developed include absolute positioning technology, ground surface location synchronization technology of GPR exploration equipment, GPR exploration image AI interpretation technology, and integrated underground space map-based composite data processing technology. The hardware systems developed include a vehicle-type exploration system and a cart-type exploration system. The data was collected using the developed exploration system, which employs 3D absolute positioning technology. The GPR exploration images were analyzed using AI technology, and the three-dimensional location information of the explored precise underground facilities was compared to the integrated underground space map. The study successfully developed a precision underground facility exploration system based on 3D absolute positioning technology. The developed exploration system can accurately survey up to a depth of 5m and measure the 3D absolute location of precise underground facilities. The system comprises software technologies that build a 3D precise DEM, synchronize the GPR sensor's ground surface 3D location coordinates, automatically analyze and detect underground facility information in GPR exploration images and improve accuracy through comparative analysis of the three-dimensional location information, and hardware systems, including a vehicle-type exploration system and a cart-type exploration system. The study's findings and technological advancements are essential for underground safety management in Korea. The proposed precision exploration system significantly contributes to establishing precise location information of underground facility information, which is crucial for underground safety management and improves the accuracy and efficiency of exploration. The study addressed the limitations of existing equipment in exploring underground facilities, proposed 3D absolute positioning technology-based precision exploration system, developed software and hardware systems for the exploration system, and contributed to underground safety management by providing precise location information. The developed precision underground facility exploration system based on 3D absolute positioning technology has the potential to provide accurate and efficient exploration of underground facilities up to a depth of 5m. The system's technological advancements contribute to the establishment of precise location information of underground facility information, which is essential for underground safety management in Korea.

Keywords: 3D absolute positioning, AI interpretation of GPR exploration images, complex data processing, integrated underground space maps, precision exploration system for underground facilities

Procedia PDF Downloads 60

Authors: J. T. Pitale, S. Ghassab, H. Ay, N. Berme

Abstract:

Biomechanics researchers commonly use marker-based optical motion capture (MoCap) systems to extract human body kinematic data. These systems use cameras to detect passive or active markers placed on the subject. The cameras use triangulation methods to form images of the markers, which typically require each marker to be visible by at least two cameras simultaneously. Cameras in a conventional optical MoCap system are mounted at a distance from the subject, typically on walls, ceiling as well as fixed or adjustable frame structures. To accommodate for space constraints and as portable force measurement systems are getting popular, there is a need for smaller and smaller capture volumes. When the efficacy of a MoCap system is investigated, it is important to consider the tradeoff amongst the camera distance from subject, pixel density, and the field of view (FOV). If cameras are mounted relatively close to a subject, the area corresponding to each pixel reduces, thus increasing the image resolution. However, the cross section of the capture volume also decreases, causing reduction of the visible area. Due to this reduction, additional cameras may be required in such applications. On the other hand, mounting cameras relatively far from the subject increases the visible area but reduces the image quality. The goal of this study was to develop a quantitative methodology to investigate marker occlusions and optimize camera placement for a given capture volume and subject postures using three-dimension computer-aided design (CAD) tools. We modeled a 4.9m x 3.7m x 2.4m (LxWxH) MoCap volume and designed a mounting structure for cameras using SOLIDWORKS (Dassault Systems, MA, USA). The FOV was used to generate the capture volume for each camera placed on the structure. A human body model with configurable posture was placed at the center of the capture volume on CAD environment. We studied three postures; initial contact, mid-stance, and early swing. The human body CAD model was adjusted for each posture based on the range of joint angles. Markers were attached to the model to enable a full body capture. The cameras were placed around the capture volume at a maximum distance of 2.7m from the subject. We used the Camera View feature in SOLIDWORKS to generate images of the subject as seen by each camera and the number of markers visible to each camera was tabulated. The approach presented in this study provides a quantitative method to investigate the efficacy and efficiency of a MoCap camera setup. This approach enables optimization of a camera setup through adjusting the position and orientation of cameras on the CAD environment and quantifying marker visibility. It is also possible to compare different camera setup options on the same quantitative basis. The flexibility of the CAD environment enables accurate representation of the capture volume, including any objects that may cause obstructions between the subject and the cameras. With this approach, it is possible to compare different camera placement options to each other, as well as optimize a given camera setup based on quantitative results.

Keywords: motion capture, cameras, biomechanics, gait analysis

Procedia PDF Downloads 309

Authors: Saeedeh Barzegar Khalilsaraei, Manoocheher Kelarestaghi, Farshad Eshghi

Abstract:

The vehicle color, as a prominent and stable feature, helps to identify a vehicle more accurately. As a result, vehicle color recognition is of great importance in intelligent transportation systems. Unlike conventional methods which use only a single Convolutional Neural Network (CNN) for feature extraction or classification, in this paper, four CNNs, with different architectures well-performing in different classes, are trained to extract various features from the input image. To take advantage of the distinct capability of each network, the multiple outputs are combined using a stack generalization algorithm as an ensemble technique. As a result, the final model performs better than each CNN individually in vehicle color identification. The evaluation results in terms of overall average accuracy and accuracy variance show the proposed method’s outperformance compared to the state-of-the-art rivals.

Keywords: Vehicle Color Recognition, Ensemble Algorithm, Stack Generalization, Convolutional Neural Network

Procedia PDF Downloads 81

Authors: Jiahe Liu, HongyangYu, Feng Qian, Miao Luo

Abstract:

This study aimed to address the problem of improving the confidence of key points by fusing multi-view information, thereby estimating human posture more accurately. We first obtained multi-view image information and then used the MvP algorithm to fuse this multi-view information together to obtain a set of high-confidence human key points. We used these as the input for the Spatio-Temporal Graph Convolution (ST-GCN). ST-GCN is a deep learning model used for processing spatio-temporal data, which can effectively capture spatio-temporal relationships in video sequences. By using the MvP algorithm to fuse multi-view information and inputting it into the spatio-temporal graph convolution model, this study provides an effective method to improve the accuracy of human posture estimation and provides strong support for further research and application in related fields.

Keywords: multi-view, pose estimation, ST-GCN, joint fusion

Procedia PDF Downloads 68

Authors: M. De Iuliis, O. Kammouh, G. P. Cimellaro, S. Tesfamariam

Abstract:

Community Resilience has gained a significant attention due to the recent unexpected natural and man-made disasters. Resilience is the process of maintaining livable conditions in the event of interruptions in normally available services. Estimating the resilience of systems, ranging from individuals to communities, is a formidable task due to the complexity involved in the process. The most challenging parameter involved in the resilience assessment is the 'downtime'. Downtime is the time needed for a system to recover its services following a disaster event. Estimating the exact downtime of a system requires a lot of inputs and resources that are not always obtainable. The uncertainties in the downtime estimation are usually handled using probabilistic methods, which necessitates acquiring large historical data. The estimation process also involves ignorance, imprecision, vagueness, and subjective judgment. In this paper, a fuzzy-based approach to estimate the downtime of building structures following earthquake events is proposed. Fuzzy logic can integrate descriptive (linguistic) knowledge and numerical data into the fuzzy system. This ability allows the use of walk down surveys, which collect data in a linguistic or a numerical form. The use of fuzzy logic permits a fast and economical estimation of parameters that involve uncertainties. The first step of the method is to determine the building’s vulnerability. A rapid visual screening is designed to acquire information about the analyzed building (e.g. year of construction, structural system, site seismicity, etc.). Then, a fuzzy logic is implemented using a hierarchical scheme to determine the building damageability, which is the main ingredient to estimate the downtime. Generally, the downtime can be divided into three main components: downtime due to the actual damage (DT1); downtime caused by rational and irrational delays (DT2); and downtime due to utilities disruption (DT3). In this work, DT1 is computed by relating the building damageability results obtained from the visual screening to some already-defined components repair times available in the literature. DT2 and DT3 are estimated using the REDITM Guidelines. The Downtime of the building is finally obtained by combining the three components. The proposed method also allows identifying the downtime corresponding to each of the three recovery states: re-occupancy; functional recovery; and full recovery. Future work is aimed at improving the current methodology to pass from the downtime to the resilience of buildings. This will provide a simple tool that can be used by the authorities for decision making.

Keywords: resilience, restoration, downtime, community resilience, fuzzy logic, recovery, damage, built environment

Procedia PDF Downloads 158

Authors: Manal Anis, Bin Bakhti Sayeed

Abstract:

With the encroachment of automobile upon our communities for decades and the concomitant urban sprawl resulting in a loss of public place, it was only a matter of time before people, realizing the role of streets in stimulating urban prosperity, would start reclaiming them to rebuild their communities. In order for this restoration of communities to take effect it is imperative that streets be freed from the dominance of motor vehicles. A holistic approach to pedestrian-friendly street environment can help build communities that embody the cities in which they are found. While the developed countries are finding more and more innovative ways to integrate walkable streets to foster communal living, the developing countries still have a long way to go. Since Dhaka is still struggling to balance the growing needs of accommodating automobiles for increased population with the loss of urban community life that comes with it, it is high time that alternate approaches are looked into. This study aims to understand streets as a living corridor through which one discovers and identifies with the city. The research area is chosen to be Manik Mia Avenue, overlooking the South Plaza of the National Parliament Building in Dhaka city. Being the site of supreme power, it is precisely this symbolic importance that the National Parliament Building has in the psyche of Bangladeshis, which has given Manik Mia Avenue a significant place in the country’s history. Above all, being an avenue it is essentially a neutral territory, universally accessible, inclusive and pluralist. The needs of the Avenue’s frequent users are analyzed with the help of a multi-method approach to survey consisting of an empirical study, a questionnaire survey and interview with relevant users. The research then tries to understand the concept of walkability by exploring the different ways in which the built environment influences walking. For this analysis, the six dimensions of Matthew Carmona are taken as a guideline for a holistic approach toward the different interacting facets of an urban public space. Based on the studies, a set of criteria is proposed to evaluate, plan and design streets that are more contextual in nature. The study concludes with how the existing street patterns of Dhaka city can be rethought and redesigned to cater to peoples’ need for a public place. The proposal is meant to be an inspiration for further studies in this respect in the context of Bangladesh.

Keywords: public space, six dimensions, street, urban, walkability

Procedia PDF Downloads 221

Authors: Ayyaz Hussain, Tariq Sadad

Abstract:

Breast cancer (BC) is one of the widespread ailments among females globally. The early prognosis of BC can decrease the mortality rate. Exact findings of benign tumors can avoid unnecessary biopsies and further treatments of patients under investigation. However, due to variations in images, it is a tough job to isolate cancerous cases from normal and benign ones. The machine learning technique is widely employed in the classification of BC pattern and prognosis. In this research, a deep convolution neural network (DCNN) called AlexNet architecture is employed to get more discriminative features from breast tissues. To achieve higher accuracy, K-nearest neighbor (KNN) classifiers are employed as a substitute for the softmax layer in deep learning. The proposed model is tested on a widely used breast image database called MIAS dataset for experimental purposes and achieved 99% accuracy.

Keywords: breast cancer, DCNN, KNN, mammography

Procedia PDF Downloads 135

Authors: Thi Thi Zin, Pyke Tin, Ikuo Kobayashi, Yoichiro Horii

Abstract:

Today dairy farm experts and farmers have well recognized the importance of dairy cow Body Condition Score (BCS) since these scores can be used to optimize milk production, managing feeding system and as an indicator for abnormality in health even can be utilized to manage for having healthy calving times and process. In tradition, BCS measures are done by animal experts or trained technicians based on visual observations focusing on pin bones, pin, thurl and hook area, tail heads shapes, hook angles and short and long ribs. Since the traditional technique is very manual and subjective, the results can lead to different scores as well as not cost effective. Thus this paper proposes an algebraic geometric imaging approach for an automatic dairy cow BCS system. The proposed system consists of three functional modules. In the first module, significant landmarks or anatomical points from the cow image region are automatically extracted by using image processing techniques. To be specific, there are 23 anatomical points in the regions of ribs, hook bones, pin bone, thurl and tail head. These points are extracted by using block region based vertical and horizontal histogram methods. According to animal experts, the body condition scores depend mainly on the shape structure these regions. Therefore the second module will investigate some algebraic and geometric properties of the extracted anatomical points. Specifically, the second order polynomial regression is employed to a subset of anatomical points to produce the regression coefficients which are to be utilized as a part of feature vector in scoring process. In addition, the angles at thurl, pin, tail head and hook bone area are computed to extend the feature vector. Finally, in the third module, the extracted feature vectors are trained by using Markov Classification process to assign BCS for individual cows. Then the assigned BCS are revised by using multiple regression method to produce the final BCS score for dairy cows. In order to confirm the validity of proposed method, a monitoring video camera is set up at the milk rotary parlor to take top view images of cows. The proposed method extracts the key anatomical points and the corresponding feature vectors for each individual cows. Then the multiple regression calculator and Markov Chain Classification process are utilized to produce the estimated body condition score for each cow. The experimental results tested on 100 dairy cows from self-collected dataset and public bench mark dataset show very promising with accuracy of 98%.

Keywords: algebraic geometric imaging approach, body condition score, Markov classification, polynomial regression

Procedia PDF Downloads 156

Authors: Kunal Shah, Swati Patel

Abstract:

Many real world problems involve data which can be considered as multi-label data streams. Efficient methods exist for multi-label classification in non streaming scenarios. However, learning in evolving streaming scenarios is more challenging, as the learners must be able to adapt to change using limited time and memory. Classification is used to predict class of unseen instance as accurate as possible. Multi label classification is a variant of single label classification where set of labels associated with single instance. Multi label classification is used by modern applications, such as text classification, functional genomics, image classification, music categorization etc. This paper introduces the task of multi-label classification, methods for multi-label classification and evolution measure for multi-label classification. Also, comparative analysis of multi label classification methods on the basis of theoretical study, and then on the basis of simulation was done on various data sets.

Keywords: binary relevance, concept drift, data stream mining, MLSC, multiple window with buffer

Procedia PDF Downloads 583

Authors: Cherng-Jye Jeng, Ming-Feng Hou, Hsing-Yuan Liu, Chuan-Feng Chang, Lih-Rong Wang, Yen-Chin Lin

Abstract:

Introduction: The incidence of breast cancer has gradually increased in Taiwan, and the characteristic of younger ages impact these women in their middle age, and may also cause challenges in terms of family, work, and illness. Breasts are symbols of femininity, as well as of sex. For women, breasts are important organs for the female identity and sexual expression. Losing breasts not only affects the female role, but would also affect sexual attraction and sexual desire. Thus, women with breast cancer who have need for mastectomies experience physical incompletion, which affects women’s self-confidence, physical image, and self-orientation. Purposes: 1. To understand the physical experience of women with breast cancer. 2. To explore the issue of sexual issues on the health effects of women with breast cancer. 3. To construct a domestic sex life issue group model for domestic women with breast cancer. 4. To explore the accompaniment experiences and sexual relationship adjustments of spouses when women have breast cancer. Method: After the research plan passes IRB review, participants will be recruited at breast surgery clinic in the affiliated hospital, to screen suitable subjects for entry into the group. Between March and May 2015, two sexual health and sex life consultation groups were conducted, which were (1) 10 in postoperative groups for women with cancer; (2) 4 married couples group for postoperative women with cancer. After sharing experiences and dialogue, women can achieve mutual support and growth. Data organization and analysis underwent descriptive analysis in qualitative research, and the group process was transcribed into transcripts for overall-content and category-content analysis. Results: Ten women with breast cancer believed that participating in group can help them exchange experiences, and elevate sexual health. The main issues include: (1) after breast cancer surgery, patients generally received chemotherapy or estrogen suppressants, causing early menopause; in particular, vaginal dryness can cause pain or bleeding in intercourse, reducing their desire for sexual activity; (2) breast cancer accentuates original spousal or family and friend relationships; some people have support and care from their family, and spouses emphasize health over the appearance of breasts; however, some people do not have acceptance and support from their family, and some even hear spousal sarcasm about loss of breasts; (3) women with breast cancer have polarized expressions of optimism and pessimism in regards to their emotions, beliefs, and body image regarding cancer; this is related to the women’s original personalities, attribution of causes of cancer, and extent of worry about relapse. Conclusion: The research results can be provided as a reference to medical institutions or breast cancer volunteer teams, to pay attention to maintaining the health of women with breast cancer.

Keywords: women with breast cancer, experiences of objectifying the body, quality of sex life, sexual health

Procedia PDF Downloads 318

Authors: Sajidullah Khan, Najeeb Ullah, Wang Yin Chai, Chai Soo See

Abstract:

In this paper, we propose an approach to remove impulse and Poisson noise from X-ray images. Many filters have been used for impulse noise removal from color and gray scale images with their own strengths and weaknesses but X-ray images contain Poisson noise and unfortunately there is no intelligent filter which can detect impulse and Poisson noise from X-ray images. Our proposed filter uses the upgraded layer discrimination approach to detect both Impulse and Poisson noise corrupted pixels in X-ray images and then restores only those detected pixels with a simple efficient and reliable one line equation. Our Proposed algorithms are very effective and much more efficient than all existing filters used only for Impulse noise removal. The proposed method uses a new powerful and efficient noise detection method to determine whether the pixel under observation is corrupted or noise free. Results from computer simulations are used to demonstrate pleasing performance of our proposed method.

Keywords: X-ray image de-noising, impulse noise, poisson noise, PRWF

Procedia PDF Downloads 381

Authors: Ujjal Chattaraj, P. C. Dasbebartta, S. Bhuyan

Abstract:

This paper aims to propose a method to detect the action of the driver’s eyes, using the concept of face detection. There are three major key contributing methods which can rapidly process the framework of the facial image and hence produce results which further can program the reactions of the vehicles as pre-programmed for the traffic safety. This paper compares and analyses the methods on the basis of their reaction time and their ability to deal with fluctuating images of the driver. The program used in this study is simple and efficient, built using the AdaBoost learning algorithm. Through this program, the system would be able to discard background regions and focus on the face-like regions. The results are analyzed on a common computer which makes it feasible for the end users. The application domain of this experiment is quite wide, such as detection of drowsiness or influence of alcohols in drivers or detection for the case of identification.

Keywords: AdaBoost learning algorithm, face detection, framework, traffic safety

Procedia PDF Downloads 156

Authors: Eva Štěpánková

Abstract:

Environmental management implementation is presently one of the ways of organization success and value improvement. Increasing an organization motivation to environmental measures introduction is caused primarily by the rising pressure of the society that generates various incentives to endeavor for the environmental performance improvement. The aim of the paper is to identify and characterize the key incentives and expectations leading organizations to the environmental management implementation. The author focuses on five businesses of different size and field, operating in the Czech Republic. The qualitative approach and grounded theory procedure are used in research. The results point out that the significant incentives for environmental management implementation represent primarily demands of customers, the opportunity to declare the environmental commitment and image improvement. The researched enterprises less commonly expect the economical contribution, competitive advantage increase or export rate improvement. The results show that marketing contributions are primarily expected from the environmental management implementation.

Keywords: environmental management, environmental management system, ISO 14001, Czech Republic

Procedia PDF Downloads 384