Search results for: JPOLE algorithm

Authors: Abdullah Bal, Sevdenur Bal

Abstract:

This paper purposes a new type of hardware application for training of cellular neural networks (CNN) using optical joint transform correlation (JTC) architecture for image feature extraction. CNNs require much more computation during the training stage compare to test process. Since optoelectronic hardware applications offer possibility of parallel high speed processing capability for 2D data processing applications, CNN training algorithm can be realized using Fourier optics technique. JTC employs lens and CCD cameras with laser beam that realize 2D matrix multiplication and summation in the light speed. Therefore, in the each iteration of training, JTC carries more computation burden inherently and the rest of mathematical computation realized digitally. The bipolar data is encoded by phase and summation of correlation operations is realized using multi-object input joint images. Overlapping properties of JTC are then utilized for summation of two cross-correlations which provide less computation possibility for training stage. Phase-only JTC does not require data rearrangement, electronic pre-calculation and strict system alignment. The proposed system can be incorporated simultaneously with various optical image processing or optical pattern recognition techniques just in the same optical system.

Keywords: CNN training, image processing, joint transform correlation, optoelectronic hardware

Procedia PDF Downloads 506

Authors: Allan F. F. Alves, Fernando A. Bacchim Neto, Guilherme Giacomini, Marcela de Oliveira, Ana L. M. Pavan, Maria E. D. Rosa, Diana R. Pina

Abstract:

Stroke is a worldwide concern, only in Brazil it accounts for 10% of all registered deaths. There are 2 stroke types, ischemic (87%) and hemorrhagic (13%). Early diagnosis is essential to avoid irreversible cerebral damage. Non-enhanced computed tomography (NECT) is one of the main diagnostic techniques used due to its wide availability and rapid diagnosis. Detection depends on the size and severity of lesions and the time spent between the first symptoms and examination. The Alberta Stroke Program Early CT Score (ASPECTS) is a subjective method that increases the detection rate. The aim of this work was to implement an image segmentation system to enhance ischemic stroke and to quantify the area of ischemic and hemorrhagic stroke lesions in CT scans. We evaluated 10 patients with NECT examinations diagnosed with ischemic stroke. Analyzes were performed in two axial slices, one at the level of the thalamus and basal ganglion and one adjacent to the top edge of the ganglionic structures with window width between 80 and 100 Hounsfield Units. We used different image processing techniques such as morphological filters, discrete wavelet transform and Fuzzy C-means clustering. Subjective analyzes were performed by a neuroradiologist according to the ASPECTS scale to quantify ischemic areas in the middle cerebral artery region. These subjective analysis results were compared with objective analyzes performed by the computational algorithm. Preliminary results indicate that the morphological filters actually improve the ischemic areas for subjective evaluations. The comparison in area of the ischemic region contoured by the neuroradiologist and the defined area by computational algorithm showed no deviations greater than 12% in any of the 10 examination tests. Although there is a tendency that the areas contoured by the neuroradiologist are smaller than those obtained by the algorithm. These results show the importance of a computer aided diagnosis software to assist neuroradiology decisions, especially in critical situations as the choice of treatment for ischemic stroke.

Keywords: ischemic stroke, image processing, CT scans, Fuzzy C-means

Procedia PDF Downloads 366

Authors: Palash Dey, Sudip Talukdar

Abstract:

In this work, a cross-breed approach is presented for obtaining both the amount of the damage intensity and location of damage existing in thin-walled members. This cross-breed approach is developed based on response surface methodology (RSM) and genetic algorithm (GA). Theoretical finite element (FE) model of cracked angle section thin walled curved beam has been linked to the developed approach to carry out trial experiments to generate response surface functions (RSFs) of free, forced and heterogeneous dynamic response data. Subsequently, the error between the computed response surface functions and measured dynamic response data has been minimized using GA to find out the optimum damage parameters (amount of the damage intensity and location). A single crack of varying location and depth has been considered in this study. The presented approach has been found to reveal good accuracy in prediction of crack parameters and possess great potential in crack detection as it requires only the current response of a cracked beam.

Keywords: damage parameters, finite element, genetic algorithm, response surface methodology, thin walled curved beam

Procedia PDF Downloads 248

Authors: Tayyaba Hussain, Sobia Jangsher, Saqib Ali, Saqib Ejaz

Abstract:

Unmanned Aerial vehicles (UAV’s) have gained great popularity due to their remoteness, ease of deployment and high maneuverability in different applications like real-time surveillance, image capturing, weather atmospheric studies, disaster site monitoring and mapping. These applications can involve a real-time communication with the ground station. However, altitude and mobility possess a few challenges for the communication. UAV’s at high altitude usually require more transmit power. One possible solution can be with the use of multi hops (UAV’s acting as relays) and exploiting the mobility pattern of the UAV’s. In this paper, we studied a relay (UAV’s acting as relays) selection for a reliable transmission to a destination UAV. We exploit the mobility information of the UAV’s to propose a Mobility-Aware Relay Selection (MARS) algorithm with the objective of giving improved data rates. The results are compared with Non Mobility-Aware relay selection scheme and optimal values. Numerical results show that our proposed MARS algorithm gives 6% better achievable data rates for the mobile UAV’s as compared with Non MobilityAware relay selection scheme. On average a decrease of 20.2% in data rate is achieved with MARS as compared with SDP solver in Yalmip.

Keywords: mobility aware, relay selection, time division multiple acess, unmanned aerial vehicle

Procedia PDF Downloads 238

Authors: Mohammad Tarek Al Muallim, Ozhan Duzenli, Ceyhun Ilguy

Abstract:

Harbor protection systems are so important. The need for automatic protection systems has increased over the last years. Diver detection active sonar has great significance. It used to detect underwater threats such as divers and autonomous underwater vehicle. To automatically detect such threats the sonar image is processed by algorithms. These algorithms used to detect, track and classify of underwater objects. In this work, divers tracking and classification algorithm is improved be proposing a robust wake detection method. To detect objects the sonar images is normalized then segmented based on fixed threshold. Next, the centroids of the segments are found and clustered based on distance metric. Then to track the objects linear Kalman filter is applied. To reduce effect of noise and creation of false tracks, the Kalman tracker is fine tuned. The tuning is done based on our active sonar specifications. After the tracks are initialed and updated they are subjected to a filtering stage to eliminate the noisy and unstable tracks. Also to eliminate object with a speed out of the diver speed range such as buoys and fast boats. Afterwards the result tracks are subjected to a classification stage to deiced the type of the object been tracked. Here the classification stage is to deice wither if the tracked object is an open circuit diver or a close circuit diver. At the classification stage, a small area around the object is extracted and a novel wake detection method is applied. The morphological features of the object with his wake is extracted. We used support vector machine to find the best classifier. The sonar training images and the test images are collected by ARMELSAN Defense Technologies Company using the portable diver detection sonar ARAS-2023. After applying the algorithm to the test sonar data, we get fine and stable tracks of the divers. The total classification accuracy achieved with the diver type is 97%.

Keywords: harbor protection, diver detection, active sonar, wake detection, diver classification

Procedia PDF Downloads 238

Authors: Adewale O. Ogunde, Emmanuel O. Ajibade

Abstract:

The task of selecting a few elective courses from a variety of available elective courses has been a difficult one for many students over the years. In many higher institutions, guidance and counselors or level advisers are usually employed to assist the students in picking the right choice of courses. In reality, these counselors and advisers are most times overloaded with too many students to attend to, and sometimes they do not have enough time for the students. Most times, the academic strength of the student based on past results are not considered in the new choice of electives. Recommender systems implement advanced data analysis techniques to help users find the items of their interest by producing a predicted likeliness score or a list of top recommended items for a given active user. Therefore, in this work, a collaborative filtering-based recommender system that will dynamically recommend elective courses to undergraduate students based on their past grades in related courses was developed. This approach employed the use of the k-nearest neighbor algorithm to discover hidden relationships between the related courses passed by students in the past and the currently available elective courses. Real students’ results dataset was used to build and test the recommendation model. The developed system will not only improve the academic performance of students, but it will also help reduce the workload on the level advisers and school counselors.

Keywords: collaborative filtering, elective courses, k-nearest neighbor algorithm, recommender systems

Procedia PDF Downloads 164

Authors: Latika Choudhary

Abstract:

“The potential benefits of Artificial Intelligence are huge, So are the dangers.” - Dave Water. Artificial intelligence is one of the facet of Information technology domain which despite several attempts does not have a clear definition or ambit. However it can be understood as technology to solve problems via automated decisions and predictions. Artificial intelligence is essentially an algorithm based technology which analyses the large amounts of data and then solves problems by detecting useful patterns. Owing to its automated feature it will not be wrong to say that humans & AI have more utility than humans alone or computers alone.1 For many decades AI experienced enthusiasm as well as setbacks, yet it has today become part and parcel of our everyday life, making it convenient or at times problematic. AI and related technology encompass Intellectual Property in multiple ways, the most important being AI technology for management of Intellectual Property, IP for protecting AI and IP as a hindrance to the transparency of AI systems. Thus the relationship between the two is of reciprocity as IP influences AI and vice versa. While AI is a recent concept, the IP laws for protection or even dealing with its challenges are relatively older, raising the need for revision to keep up with the pace of technological advancements. This paper will analyze the relationship between AI and IP to determine how beneficial or conflictual the same is, address how the old concepts of IP are being stretched to its maximum limits so as to accommodate the unwanted consequences of the Artificial Intelligence and propose ways to mitigate the situation so that AI becomes the friend it is and not turn into a potential foe it appears to be.

Keywords: intellectual property rights, information technology, algorithm, artificial intelligence

Procedia PDF Downloads 87

Authors: Hanqing Yang, Xiang Meng, Qi Li, Weirong Chen

Abstract:

The SOC (state of charge) based droop control has limitations on the load power sharing among different energy storage units, due to the line impedance. In this paper, a distributed control strategy for dispersed energy storage units in the DC microgrid based on discrete consensus is proposed. Firstly, a sparse information communication network is built. Thus, local controllers can communicate with its neighbors using voltage, current and SOC information. An average voltage of grid can be evaluated to compensate voltage offset by droop control, and an objective virtual resistance fulfilling above requirement can be dynamically calculated to distribute load power according to the SOC of the energy storage units. Then, the stability of the whole system and influence of communication delay are analyzed. It can be concluded that this control strategy can improve the robustness and flexibility, because of having no center controller. Finally, a model of DC microgrid with dispersed energy storage units and loads is built, the discrete distributed algorithm is established and communication protocol is developed. The co-simulation between Matlab/Simulink and JADE (Java agent development framework) has verified the effectiveness of proposed control strategy.

Keywords: dispersed energy storage units, discrete consensus algorithm, state of charge, communication delay

Procedia PDF Downloads 280

Authors: Delphine Ribes Lemay, Valentine Bernasconi, André Andrade, Lara DéFayes, Mathieu Salzmann, FréDéRic Kaplan, Nicolas Henchoz

Abstract:

Digitization and preservation of large heritage induce high maintenance costs to keep up with the technical standards and ensure sustainable access. Creating impactful usage is instrumental to justify the resources for long-term preservation. The Museum für Gestaltung of Zurich holds one of the biggest poster collections of the world from which 52’000 were digitised. In the process of building a digital installation to valorize the collection, one objective was to develop an algorithm capable of predicting the next poster to show according to the ones already displayed. The work presented here describes the steps to build an algorithm able to automatically create sequences of posters reflecting associations performed by curator and professional designers. The exposed challenge finds similarities with the domain of song playlist algorithms. Recently, artificial intelligence techniques and more specifically, deep-learning algorithms have been used to facilitate their generations. Promising results were found thanks to Recurrent Neural Networks (RNN) trained on manually generated playlist and paired with clusters of extracted features from songs. We used the same principles to create the proposed algorithm but applied to a challenging medium, posters. First, a convolutional autoencoder was trained to extract features of the posters. The 52’000 digital posters were used as a training set. Poster features were then clustered. Next, an RNN learned to predict the next cluster according to the previous ones. RNN training set was composed of poster sequences extracted from a collection of books from the Gestaltung Museum of Zurich dedicated to displaying posters. Finally, within the predicted cluster, the poster with the best proximity compared to the previous poster is selected. The mean square distance between features of posters was used to compute the proximity. To validate the predictive model, we compared sequences of 15 posters produced by our model to randomly and manually generated sequences. Manual sequences were created by a professional graphic designer. We asked 21 participants working as professional graphic designers to sort the sequences from the one with the strongest graphic line to the one with the weakest and to motivate their answer with a short description. The sequences produced by the designer were ranked first 60%, second 25% and third 15% of the time. The sequences produced by our predictive model were ranked first 25%, second 45% and third 30% of the time. The sequences produced randomly were ranked first 15%, second 29%, and third 55% of the time. Compared to designer sequences, and as reported by participants, model and random sequences lacked thematic continuity. According to the results, the proposed model is able to generate better poster sequencing compared to random sampling. Eventually, our algorithm is sometimes able to outperform a professional designer. As a next step, the proposed algorithm should include a possibility to create sequences according to a selected theme. To conclude, this work shows the potentiality of artificial intelligence techniques to learn from existing content and provide a tool to curate large sets of data, with a permanent renewal of the presented content.

Keywords: Artificial Intelligence, Digital Humanities, serendipity, design research

Procedia PDF Downloads 184

Authors: Gautam Kumar Saharia, Sagardeep Talukdar, Riki Dutta, Sudipta Nandy

Abstract:

The physics-informed neural network (PINN) method opens up an approach for numerically solving nonlinear partial differential equations leveraging fast calculating speed and high precession of modern computing systems. We construct the PINN based on a strong universal approximation theorem and apply the initial-boundary value data and residual collocation points to weekly impose initial and boundary conditions to the neural network and choose the optimization algorithms adaptive moment estimation (ADAM) and Limited-memory Broyden-Fletcher-Golfard-Shanno (L-BFGS) algorithm to optimize learnable parameter of the neural network. Next, we improve the PINN with a weighted loss function to obtain both the bright and dark soliton solutions of the Fokas-Lenells equation (FLE). We find the proposed scheme of adjustable weight coefficients into PINN has a better convergence rate and generalizability than the basic PINN algorithm. We believe that the PINN approach to solve the partial differential equation appearing in nonlinear optics would be useful in studying various optical phenomena.

Keywords: deep learning, optical soliton, physics informed neural network, partial differential equation

Procedia PDF Downloads 70

Authors: Dawei Li, Cheng Liu, Yiteng Liu

Abstract:

This paper introduces an embedded digital image system for Chinese space environment vertical exploration sounding rocket. In order to record the flight status of the sounding rocket as well as the payloads, an onboard embedded image processing system based on ADV212, a JPEG2000 compression chip, is designed in this paper. Since the sounding rocket is not designed to be recovered, all image data should be transmitted to the ground station before the re-entry while the downlink band used for the image transmission is only about 600 kbps. Under the same condition of compression ratio compared with other algorithm, JPEG2000 standard algorithm can achieve better image quality. So JPEG2000 image compression is applied under this condition with a limited downlink data band. This embedded image system supports lossless to 200:1 real time compression, with two cameras to monitor nose ejection and motor separation, and two cameras to monitor boom deployment. The encoder, ADV7182, receives PAL signal from the camera, then output the ITU-R BT.656 signal to ADV212. ADV7182 switches between four input video channels as the program sequence. Two SRAMs are used for Ping-pong operation and one 512 Mb SDRAM for buffering high frame-rate images. The whole image system has the characteristics of low power dissipation, low cost, small size and high reliability, which is rather suitable for this sounding rocket application.

Keywords: ADV212, image system, JPEG2000, sounding rocket

Procedia PDF Downloads 421

Authors: Evan Lowhorn, Rocio Alba-Flores

Abstract:

The purpose of this work is to develop a method for classifying hand signals and using the output in a drone control algorithm. To achieve this, methods based on Convolutional Neural Networks (CNN) were applied. CNN's are a subset of deep learning, which allows grid-like inputs to be processed and passed through a neural network to be trained for classification. This type of neural network allows for classification via imaging, which is less intrusive than previous methods using biosensors, such as EMG sensors. Classification CNN's operate purely from the pixel values in an image; therefore they can be used without additional exteroceptive sensors. A development bench was constructed using a desktop computer connected to a high-definition webcam mounted on a scissor arm. This allowed the camera to be pointed downwards at the desk to provide a constant solid background for the dataset and a clear detection area for the user. A MATLAB script was created to automate dataset image capture at the development bench and save the images to the desktop. This allowed the user to create their own dataset of 12,000 images within three hours. These images were evenly distributed among seven classes. The defined classes include forward, backward, left, right, idle, and land. The drone has a popular flip function which was also included as an additional class. To simplify control, the corresponding hand signals chosen were the numerical hand signs for one through five for movements, a fist for land, and the universal “ok” sign for the flip command. Transfer learning with PyTorch (Python) was performed using a pre-trained 18-layer residual learning network (ResNet-18) to retrain the network for custom classification. An algorithm was created to interpret the classification and send encoded messages to a Ryze Tello drone over its 2.4 GHz Wi-Fi connection. The drone’s movements were performed in half-meter distance increments at a constant speed. When combined with the drone control algorithm, the classification performed as desired with negligible latency when compared to the delay in the drone’s movement commands.

Keywords: classification, computer vision, convolutional neural networks, drone control

Procedia PDF Downloads 210

Authors: Hassan Qandil

Abstract:

Using a statistical algorithm incorporated in MATLAB, four types of non-imaging Fresnel lenses are designed; spot-flat, linear-flat, dome-shaped and semi-cylindrical-shaped. The optimization employs a statistical ray-tracing methodology of the incident light, mainly considering effects of chromatic aberration, varying focal lengths, solar inclination and azimuth angles, lens and receiver apertures, and the optimum number of prism grooves. While adopting an equal-groove-width assumption of the Poly-methyl-methacrylate (PMMA) prisms, the main target is to maximize the ray intensity on the receiver’s aperture and therefore achieving higher values of heat flux. The algorithm outputs prism angles and 2D sketches. 3D drawings are then generated via AutoCAD and linked to COMSOL Multiphysics software to simulate the lenses under solar ray conditions, which provides optical and thermal analysis at both the lens’ and the receiver’s apertures while setting conditions as per the Dallas-TX weather data. Once the lenses’ characterization is finalized, receivers are designed based on its optimized aperture size. Several cavity shapes; including triangular, arc-shaped and trapezoidal, are tested while coupled with a variety of receiver materials, working fluids, heat transfer mechanisms, and enclosure designs. A vacuum-reflective enclosure is also simulated for an enhanced thermal absorption efficiency. Each receiver type is simulated via COMSOL while coupled with the optimized lens. A lab-scale prototype for the optimum lens-receiver configuration is then fabricated for experimental evaluation. Application-based testing is also performed for the selected configuration, including that of a photovoltaic-thermal cogeneration system and solar furnace system. Finally, some future research work is pointed out, including the coupling of the collector-receiver system with an end-user power generator, and the use of a multi-layered genetic algorithm for comparative studies.

Keywords: COMSOL, concentrator, energy, fresnel, optics, renewable, solar

Procedia PDF Downloads 155

Authors: Sai Hung Cheung, Zhe Shao

Abstract:

Recent earthquake-induced tsunamis in Padang, 2004 and Tohoku, 2011 brought huge losses of lives and properties. Maintaining vertical evacuation systems is the most crucial strategy to effectively reduce casualty during the tsunami event. Thus, it is of our great interest to quantify the risk to structural dynamic systems due to earthquake-induced tsunamis. Despite continuous advancement in computational simulation of the tsunami and wave-structure interaction modeling, it still remains computationally challenging to evaluate the reliability (or its complement failure probability) of a structural dynamic system when uncertainties related to the system and its modeling are taken into account. The failure of the structure in a tsunami-wave-structural system is defined as any response quantities of the system exceeding specified thresholds during the time when the structure is subjected to dynamic wave impact due to earthquake-induced tsunamis. In this paper, an approach based on a novel integration of the Subset Simulation algorithm and a recently proposed moving least squares response surface approach for stochastic sampling is proposed. The effectiveness of the proposed approach is discussed by comparing its results with those obtained from the Subset Simulation algorithm without using the response surface approach.

Keywords: response surface model, subset simulation, structural reliability, Tsunami risk

Procedia PDF Downloads 383

Authors: Indri Mahadiraka Rumamby, R. R. Dwinanti Rika Marthanty, Jessica Sjah

Abstract:

Smoothed particle hydrodynamics (SPH) was originally created to simulate nonaxisymmetric phenomena in astrophysics. However, this method still has several shortcomings, namely the high computational cost required to model values with high resolution and problems with boundary conditions. The difficulty of modeling boundary conditions occurs because the SPH method is influenced by particle deficiency due to the integral of the kernel function being truncated by boundary conditions. This research aims to answer if SPH modeling with a focus on boundary layer interactions and continuous flow can produce quantifiably accurate values with low computational cost. This research will combine algorithms and coding in the main program of meandering river, continuous flow algorithm, and solid-fluid algorithm with the aim of obtaining quantitatively accurate results on solid-fluid interactions with the continuous flow on a meandering channel using the SPH method. This study uses the Fortran programming language for modeling the SPH (Smoothed Particle Hydrodynamics) numerical method; the model is conducted in the form of a U-shaped meandering open channel in 3D, where the channel walls are soil particles and uses a continuous flow with a limited number of particles.

Keywords: smoothed particle hydrodynamics, computational fluid dynamics, numerical simulation, fluid mechanics

Procedia PDF Downloads 131

Authors: Ganesh B. Shinde, Vijaya B. Musande

Abstract:

Cotton crop identification based on the timely information has significant advantage to the different implications of food, economic and environment. Due to the significant advantages, the accurate detection of cotton crop regions using supervised learning procedure is challenging problem in remote sensing. Here, classifiers on the direct image are played a major role but the results are not much satisfactorily. In order to further improve the effectiveness, variety of vegetation indices are proposed in the literature. But, recently, the major challenge is to find the better vegetation indices for the cotton crop identification through the proposed methodology. Accordingly, fuzzy c-means clustering is combined with neural network algorithm, trained by Levenberg-Marquardt for cotton crop classification. To experiment the proposed method, five LISS-III satellite images was taken and the experimentation was done with six vegetation indices such as Simple Ratio, Normalized Difference Vegetation Index, Enhanced Vegetation Index, Green Atmospherically Resistant Vegetation Index, Wide-Dynamic Range Vegetation Index, Green Chlorophyll Index. Along with these indices, Green Leaf Area Index is also considered for investigation. From the research outcome, Green Atmospherically Resistant Vegetation Index outperformed with all other indices by reaching the average accuracy value of 95.21%.

Keywords: Fuzzy C-Means clustering (FCM), neural network, Levenberg-Marquardt (LM) algorithm, vegetation indices

Procedia PDF Downloads 318

Authors: Alexandru-Ion Marinescu

Abstract:

There exist a plethora of methods in the scientific literature which tackle the well-established task of credit score evaluation. In its most abstract form, a credit scoring algorithm takes as input several credit applicant properties, such as age, marital status, employment status, loan duration, etc. and must output a binary response variable (i.e. “GOOD” or “BAD”) stating whether the client is susceptible to payment return delays. Data imbalance is a common occurrence among financial institution databases, with the majority being classified as “GOOD” clients (clients that respect the loan return calendar) alongside a small percentage of “BAD” clients. But it is the “BAD” clients we are interested in since accurately predicting their behavior is crucial in preventing unwanted loss for loan providers. We add to this whole context the constraint that the algorithm must yield an actual, tractable mathematical formula, which is friendlier towards financial analysts. To this end, we have turned to genetic algorithms and genetic programming, aiming to evolve actual mathematical expressions using specially tailored mutation and crossover operators. As far as data representation is concerned, we employ a very flexible mechanism – LINQ expression trees, readily available in the C# programming language, enabling us to construct executable pieces of code at runtime. As the title implies, they model trees, with intermediate nodes being operators (addition, subtraction, multiplication, division) or mathematical functions (sin, cos, abs, round, etc.) and leaf nodes storing either constants or variables. There is a one-to-one correspondence between the client properties and the formula variables. The mutation and crossover operators work on a flattened version of the tree, obtained via a pre-order traversal. A consequence of our chosen technique is that we can identify and discard client properties which do not take part in the final score evaluation, effectively acting as a dimensionality reduction scheme. We compare ourselves with state of the art approaches, such as support vector machines, Bayesian networks, and extreme learning machines, to name a few. The data sets we benchmark against amount to a total of 8, of which we mention the well-known Australian credit and German credit data sets, and the performance indicators are the following: percentage correctly classified, area under curve, partial Gini index, H-measure, Brier score and Kolmogorov-Smirnov statistic, respectively. Finally, we obtain encouraging results, which, although placing us in the lower half of the hierarchy, drive us to further refine the algorithm.

Keywords: expression trees, financial credit scoring, genetic algorithm, genetic programming, symbolic evolution

Procedia PDF Downloads 117

Authors: Thein Thein, Kalyar Myo San

Abstract:

Lip reading system is one of the different supportive technologies for hearing impaired, or elderly people or non-native speakers. For normal hearing persons in noisy environments or in conditions where the audio signal is not available, lip reading techniques can be used to increase their understanding of spoken language. Hearing impaired persons have used lip reading techniques as important tools to find out what was said by other people without hearing voice. Thus, visual speech information is important and become active research area. Using visual information from lip movements can improve the accuracy and robustness of a speech recognition system and the need for lip reading system is ever increasing for every language. However, the recognition of lip movement is a difficult task because of the region of interest (ROI) is nonlinear and noisy. Therefore, this paper proposes method to detect the accurate lips shape and to localize lip movement towards automatic lip tracking by using the combination of Otsu global thresholding technique and Moore Neighborhood Tracing Algorithm. Proposed method shows how accurate lip localization and tracking which is useful for speech recognition. In this work of study and experiments will be carried out the automatic lip localizing the lip shape for Myanmar consonants using the only visual information from lip movements which is useful for visual speech of Myanmar languages.

Keywords: lip reading, lip localization, lip tracking, Moore neighborhood tracing algorithm

Procedia PDF Downloads 352

Authors: Andres Manuel Garcia Lamparte, Rocio Losada Iglesias, Marcos BoullóN Magan, David Miranda Barros

Abstract:

The biodiversity strategy of the European Union for 2030, mentions climate change as one of the key factors for biodiversity loss and considers green infrastructure as one of the solutions to this problem. In this line, the European Commission has developed a green infrastructure strategy which commits members states to consider green infrastructure in their territorial planning. This green infrastructure is aimed at granting the provision of a wide number of ecosystem services to support biodiversity and human well-being by countering the effects of climate change. Yet, there are not too many tools available to delimit green infrastructure. The available ones consider the potential of the territory to provide ecosystem services. However, these methods usually aggregate several maps of ecosystem services potential without considering possible trade-offs. This can lead to excluding areas with a high potential for providing ecosystem services which have many trade-offs with other ecosystem services. In order to tackle this problem, a methodology is proposed to consider ecosystem services trade-offs in the objective function of a simulated annealing algorithm aimed at delimiting green infrastructure multifunctional buffer areas. To this end, the provision potential maps of the regulating ecosystem services considered to delimit the multifunctional buffer areas are clustered in groups, so that ecosystem services that create trade-offs are excluded in each group. The normalized provision potential maps of the ecosystem services in each group are added to obtain a potential map per group which is normalized again. Then the potential maps for each group are combined in a raster map that shows the highest provision potential value in each cell. The combined map is then used in the objective function of the simulated annealing algorithm. The algorithm is run both using the proposed methodology and considering the ecosystem services individually. The results are analyzed with spatial statistics and landscape metrics to check the number of ecosystem services that the delimited areas produce, as well as their regularity and compactness. It has been observed that the proposed methodology increases the number of ecosystem services produced by delimited areas, improving their multifunctionality and increasing their effectiveness in preventing climate change impacts.

Keywords: ecosystem services trade-offs, green infrastructure delineation, multifunctional buffer areas, climate change

Procedia PDF Downloads 174

Authors: Regina Chua

Abstract:

To reduce the costs of re-fielding a large survey, a Random Forest classifier was applied to measure the accuracy of classifying individuals into their assigned segments with the fewest possible questions. Given a long survey, one needed to determine the most predictive ten or fewer questions that would accurately assign new individuals to custom segments. Furthermore, the solution needed to be quick in its classification and usable in non-Python environments. In this paper, a supervised Random Forest classifier was modeled on a dataset with 7,000 individuals, 60 questions, and 254 features. The Random Forest consisted of an iterative collection of individual decision trees that result in a predicted segment with robust precision and recall scores compared to a single tree. A random 70-30 stratified sampling for training the algorithm was used, and accuracy trade-offs at different depths for each segment were identified. Ultimately, the Random Forest classifier performed at 87% accuracy at a depth of 10 with 20 instead of 254 features and 10 instead of 60 questions. With an acceptable accuracy in prioritizing feature selection, new tools were developed for non-Python environments: a worksheet with a formulaic version of the algorithm and an embedded function to predict the segment of an individual in real-time. Random Forest was determined to be an optimal classification model by its feature selection, performance, processing speed, and flexible application in other environments.

Keywords: machine learning, supervised learning, data science, random forest, classification, prediction, predictive modeling

Procedia PDF Downloads 94

Authors: Shaik Ayesha Fathima, Shaik Noor Jahan, Duvvada Rajeswara Rao

Abstract:

Earth's environment and its evolution can be seen through satellite images in near real-time. Through satellite imagery, remote sensing data provide crucial information that can be used for a variety of applications, including image fusion, change detection, land cover classification, agriculture, mining, disaster mitigation, and monitoring climate change. The objective of this project is to propose a method for classifying satellite images according to multiple predefined land cover classes. The proposed approach involves collecting data in image format. The data is then pre-processed using data pre-processing techniques. The processed data is fed into the proposed algorithm and the obtained result is analyzed. Some of the algorithms used in satellite imagery classification are U-Net, Random Forest, Deep Labv3, CNN, ANN, Resnet etc. In this project, we are using the DeepLabv3 (Atrous convolution) algorithm for land cover classification. The dataset used is the deep globe land cover classification dataset. DeepLabv3 is a semantic segmentation system that uses atrous convolution to capture multi-scale context by adopting multiple atrous rates in cascade or in parallel to determine the scale of segments.

Keywords: area calculation, atrous convolution, deep globe land cover classification, deepLabv3, land cover classification, resnet 50

Procedia PDF Downloads 140

Authors: Karchung, S. Ruangsinchaiwanich

Abstract:

This paper presents open-loop vector control method of induction motor with space vector pulse width modulation (SVPWM) technique. Normally, the closed loop speed control is preferred and is believed to be more accurate. However, it requires a position sensor to track the rotor position which is not desirable to use it for certain workspace applications. This paper exhibits the performance of three-phase induction motor with the simplest control algorithm without the use of a position sensor nor an estimation block to estimate rotor position for sensorless control. The motor stator currents are measured and are transformed to synchronously rotating (d-q-axis) frame by use of Clarke and Park transformation. The actual control happens in this frame where the measured currents are compared with the reference currents. The error signal is fed to a conventional PI controller, and the corrected d-q voltage is generated. The controller outputs are transformed back to three phase voltages and are fed to SVPWM block which generates PWM signal for the voltage source inverter. The open loop vector control model along with SVPWM algorithm is modeled in MATLAB/Simulink software and is experimented and validated in TMS320F28335 DSP board.

Keywords: electric drive, induction motor, open-loop vector control, space vector pulse width modulation technique

Procedia PDF Downloads 147

Authors: Youngsun Moon, Yeong-Ju Go, Jong-Soo Choi

Abstract:

Most drones that currently have surveillance/reconnaissance missions are basically equipped with optical equipment, but we also need to use a microphone array to estimate the location of the acoustic source. This can provide additional information in the absence of optical equipment. The purpose of this study is to estimate Direction of Arrival (DOA) based on Time Difference of Arrival (TDOA) estimation of the acoustic source in the drone. The problem is that it is impossible to measure the clear target acoustic source because of the drone noise. To overcome this problem is to separate the drone noise and the target acoustic source using Blind Source Separation(BSS) based on Independent Component Analysis(ICA). ICA can be performed assuming that the drone noise and target acoustic source are independent and each signal has non-gaussianity. For maximized non-gaussianity each signal, we use Negentropy and Kurtosis based on probability theory. As a result, we can improve TDOA estimation and DOA estimation of the target source in the noisy environment. We simulated the performance of the DOA algorithm applying BSS algorithm, and demonstrated the simulation through experiment at the anechoic wind tunnel.

Keywords: aeroacoustics, acoustic source detection, time difference of arrival, direction of arrival, blind source separation, independent component analysis, drone

Procedia PDF Downloads 162

Authors: U. Yerlikaya, R. T. Balkan

Abstract:

In this research, a method is developed to obtain high-dimensional configuration space for path planning problems. In typical cases, the path planning problems are solved directly in the 3-dimensional (D) workspace. However, this method is inefficient in handling the robots with various geometrical and mechanical restrictions. To overcome these difficulties, path planning may be formalized and solved in a new space which is called configuration space. The number of dimensions of the configuration space comes from the degree of freedoms of the system of interest. The method can be applied in two ways. In the first way, the point clouds of all the bodies of the system and interaction of them are used. The second way is performed via using the clearance function of simulation software where the minimum distances between surfaces of bodies are simultaneously measured. A double-turret system is held in the scope of this study. The 4-D configuration space of a double-turret system is obtained in these two ways. As a result, the difference between these two methods is around 1%, depending on the density of the point cloud. The disparity between the two forms steadily decreases as the point cloud density increases. At the end of the study, in order to verify 4-D configuration space obtained, 4-D path planning problem was realized as 2-D + 2-D and a sample path planning is carried out with using A* algorithm. Then, the accuracy of the configuration space is proved using the obtained paths on the simulation model of the double-turret system.

Keywords: A* algorithm, autonomous turrets, high-dimensional C-space, manifold C-space, point clouds

Procedia PDF Downloads 139

Authors: Masoud Shahmanzari

Abstract:

The aim of this paper is to present a Variable Neighborhood Search (VNS) with Tabu Search (TS) conditions for the Roaming Salesman Problem (RSP). The RSP is a special case of the well-known traveling salesman problem (TSP) where a set of cities with time-dependent rewards and a set of campaign days are given. Each city can be visited on any day and a subset of cities can be visited multiple times. The goal is to determine an optimal campaign schedule consist of daily open/closed tours that visit some cities and maximizes the total net benefit while respecting daily maximum tour duration constraints and the necessity to return campaign base frequently. This problem arises in several real-life applications and particularly in election logistics where depots are not fixed. We formulate the problem as a mixed integer linear programming (MILP), in which we capture as many real-world aspects of the RSP as possible. We also present a hybrid metaheuristic algorithm based on a VNS with TS conditions. The initial feasible solution is constructed via a new matheuristc approach based on the decomposition of the original problem. Next, this solution is improved in terms of the collected rewards using the proposed local search procedure. We consider a set of 81 cities in Turkey and a campaign of 30 days as our largest instance. Computational results on real-world instances show that the developed algorithm could find near-optimal solutions effectively.

Keywords: optimization, routing, election logistics, heuristics

Procedia PDF Downloads 92

Authors: Chin-Chia Jane

Abstract:

In a transportation network, travel time refers to the transmission time from source node to destination node, whereas reliability refers to the probability of a successful connection from source node to destination node. With an increasing emphasis on quality of service (QoS), both performance indexes are significant in the design and analysis of transportation systems. In this work, we extend the well-known flow network model for transportation networks so that travel time and reliability are integrated into the QoS measurement simultaneously. In the extended model, in addition to the general arc capacities, each intermediate node has a time weight which is the travel time for per unit of commodity going through the node. Meanwhile, arcs and nodes are treated as binary random variables that switch between operation and failure with associated probabilities. For pre-specified travel time limitation and demand requirement, the QoS of a transportation network is the probability that source can successfully transport the demand requirement to destination while the total transmission time is under the travel time limitation. This work is pioneering, since existing literatures that evaluate travel time reliability via a single optimization path, the proposed QoS focuses the performance of the whole network system. To compute the QoS of transportation networks, we first transfer the extended network model into an equivalent min-cost max-flow network model. In the transferred network, each arc has a new travel time weight which takes value 0. Each intermediate node is replaced by two nodes u and v, and an arc directed from u to v. The newly generated nodes u and v are perfect nodes. The new direct arc has three weights: travel time, capacity, and operation probability. Then the universal set of state vectors is recursively decomposed into disjoint subsets of reliable, unreliable, and stochastic vectors until no stochastic vector is left. The decomposition is made possible by applying existing efficient min-cost max-flow algorithm. Because the reliable subsets are disjoint, QoS can be obtained directly by summing the probabilities of these reliable subsets. Computational experiments are conducted on a benchmark network which has 11 nodes and 21 arcs. Five travel time limitations and five demand requirements are set to compute the QoS value. To make a comparison, we test the exhaustive complete enumeration method. Computational results reveal the proposed algorithm is much more efficient than the complete enumeration method. In this work, a transportation network is analyzed by an extended flow network model where each arc has a fixed capacity, each intermediate node has a time weight, and both arcs and nodes are independent binary random variables. The quality of service of the transportation network is an integration of customer demands, travel time, and the probability of connection. We present a decomposition algorithm to compute the QoS efficiently. Computational experiments conducted on a prototype network show that the proposed algorithm is superior to existing complete enumeration methods.

Keywords: quality of service, reliability, transportation network, travel time

Procedia PDF Downloads 221

Authors: Vitalii Yu. Kapitan, Aleksandr V. Perzhu, Konstantin V. Nefedev

Abstract:

The necessity of studying magnetic multilayer structures is explained by the prospects of their practical application as a technological base for creating new storages medium. Magnetic multilayer films have many unique features that contribute to increasing the density of information recording and the speed of storage devices. Multilayer structures are structures of alternating magnetic and nonmagnetic layers. In frame of the classical Heisenberg model, lattice spin systems with direct short- and long-range exchange interactions were investigated by Monte Carlo methods. The thermodynamic characteristics of multilayer structures, such as the temperature behavior of magnetization, energy, and heat capacity, were investigated. The processes of magnetization reversal of multilayer structures in external magnetic fields were investigated. The developed software is based on the new, promising programming language Rust. Rust is a new experimental programming language developed by Mozilla. The language is positioned as an alternative to C and C++. For the Monte Carlo simulation, the Metropolis algorithm and its parallel implementation using MPI and the Wang-Landau algorithm were used. We are planning to study of magnetic multilayer films with asymmetric Dzyaloshinskii–Moriya (DM) interaction, interfacing effects and skyrmions textures. This work was supported by the state task of the Ministry of Education and Science of the Russia # 3.7383.2017/8.9

Keywords: The Monte Carlo methods, Heisenberg model, multilayer structures, magnetic skyrmion

Procedia PDF Downloads 166

Authors: Zahzouh Zoubir, Bouzaouit Azzeddine, Gahgah Mounir

Abstract:

The use of more and more electronic power switches with a nonlinear behavior generates non-sinusoidal currents in distribution networks, which causes damage to domestic and industrial equipment. The multi-level shunt power active filter is subsequently shown to be an adequate solution to the problem raised. Nevertheless, the difficulty of adjusting the active filter DC supply voltage requires another technology to ensure it. In this article, a photovoltaic generator is associated with the DC bus power terminals of the active filter. The proposed system consists of a field of solar panels, three multi-level voltage inverters connected to the power grid and a non-linear load consisting of a six-diode rectifier bridge supplying a resistive-inductive load. Current control techniques of active and reactive power are used to compensate for both harmonic currents and reactive power as well as to inject active solar power into the distribution network. An algorithm of the search method of the maximum power point of type Perturb and observe is applied. Simulation results of the system proposed under the MATLAB/Simulink environment shows that the performance of control commands that reassure the solar power injection in the network, harmonic current compensation and power factor correction.

Keywords: Actif power filter, MPPT, pertub&observe algorithm, PV array, PWM-control

Procedia PDF Downloads 339

Authors: Yingkun Huang, Li Guo, Zekang Lan, Kai Tian

Abstract:

Hard disk drives (HDD) failure of the exascale supercomputing system may lead to service interruption and invalidate previous calculations, and it will cause permanent data loss. Therefore, initiating corrective actions before hard drive failures materialize is critical to the continued operation of jobs. In this paper, a highly accurate analysis model based on CNN-LSTM and oversampling technique was proposed, which can correctly predict the necessity of a disk replacement even ten days in advance. Generally, the learning-based method performs poorly on a training dataset with long-tail distribution, especially fault prediction is a very classic situation as the scarcity of failure data. To overcome the puzzle, a new oversampling was employed to augment the data, and then, an improved CNN-LSTM with the shortcut was built to learn more effective features. The shortcut transmits the results of the previous layer of CNN and is used as the input of the LSTM model after weighted fusion with the output of the next layer. Finally, a detailed, empirical comparison of 6 prediction methods is presented and discussed on a public dataset for evaluation. The experiments indicate that the proposed method predicts disk failure with 0.91 Precision, 0.91 Recall, 0.91 F-measure, and 0.90 MCC for 10 days prediction horizon. Thus, the proposed algorithm is an efficient algorithm for predicting HDD failure in supercomputing.

Keywords: HDD replacement, failure, CNN-LSTM, oversampling, prediction

Procedia PDF Downloads 79

Authors: Christine Böckmann, Julia Rosemann

Abstract:

We present an inversion algorithm that is used in the European Aerosol Lidar Network for the inversion of data collected with multi-wavelength Raman lidar. These instruments measure backscatter coefficients at 355, 532, and 1064 nm, and extinction coefficients at 355 and 532 nm. The algorithm is based on manually controlled inversion of optical data which allows for detailed sensitivity studies and thus provides us with comparably high quality of the derived data products. The algorithm allows us to derive particle effective radius, volume, surface-area concentration with comparably high confidence. The retrieval of the real and imaginary parts of the complex refractive index still is a challenge in view of the accuracy required for these parameters in climate change studies in which light-absorption needs to be known with high accuracy. Single-scattering albedo (SSA) can be computed from the retrieve microphysical parameters and allows us to categorize aerosols into high and low absorbing aerosols. From mathematical point of view the algorithm is based on the concept of using truncated singular value decomposition as regularization method. This method was adapted to work for the retrieval of the particle size distribution function (PSD) and is called hybrid regularization technique since it is using a triple of regularization parameters. The inversion of an ill-posed problem, such as the retrieval of the PSD, is always a challenging task because very small measurement errors will be amplified most often hugely during the solution process unless an appropriate regularization method is used. Even using a regularization method is difficult since appropriate regularization parameters have to be determined. Therefore, in a next stage of our work we decided to use two regularization techniques in parallel for comparison purpose. The second method is an iterative regularization method based on Pade iteration. Here, the number of iteration steps serves as the regularization parameter. We successfully developed a semi-automated software for spherical particles which is able to run even on a parallel processor machine. From a mathematical point of view, it is also very important (as selection criteria for an appropriate regularization method) to investigate the degree of ill-posedness of the problem which we found is a moderate ill-posedness. We computed the optical data from mono-modal logarithmic PSD and investigated particles of spherical shape in our simulations. We considered particle radii as large as 6 nm which does not only cover the size range of particles in the fine-mode fraction of naturally occurring PSD but also covers a part of the coarse-mode fraction of PSD. We considered errors of 15% in the simulation studies. For the SSA, 100% of all cases achieve relative errors below 12%. In more detail, 87% of all cases for 355 nm and 88% of all cases for 532 nm are well below 6%. With respect to the absolute error for non- and weak-absorbing particles with real parts 1.5 and 1.6 in all modes the accuracy limit +/- 0.03 is achieved. In sum, 70% of all cases stay below +/-0.03 which is sufficient for climate change studies.

Keywords: aerosol particles, inverse problem, microphysical particle properties, regularization

Procedia PDF Downloads 343