Pulse Coupled Neural Network

Many factors contribute to the aberrations induced in an optical system. Atmospheric turbulence between the object and the imaging system, physical or thermal perturbations in optical elements degrade the system's point spread function, and misaligned optics are the primary sources of aberrations that affect image quality. The design of a nonconventional real-time adaptive optic system using a micro-mirror device for wavefront correction is presented. The unconventional compensated imaging system presented offers advantages in speed, cost, power consumption, and weight. A pulsed-coupled neural network is used to as a preprocessor to enhance the performance of the wavefront sensor for low-light applications. Modeling results that characterize the system performance are presented.

The invention of ever faster and more complex computers makes more and more applications of di-gital signal processing possible. And in turn, new application areas require increasingly more compu-tational power. One of the most computationally in-tensive areas is that of ( ...

The use of general descriptive names, registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use.

This paper introduces a novel method for human face recognition based on a simplified approach for the Pulse Coupled Neural Network (PCNN) Algorithm. The face image is introduced to the Intersecting Cortical Model (ICM) to be iterated 200 times, and then the time signals for the faces are compared to make a decision. Experimental results for human face recognition confirm that the proposed method lends itself to higher classification accuracy relative to existing techniques.

Recent advances in the understanding of the mammalian visual cortex have led to new approaches for image processing techniques. As a result of this, computer simulations using the proposed visual cortex model have become very useful in the field of image processing. Models of this kind have the ability to efficiently extract image segments, edges and texture. They Ž. operate by generating a set of pulse images images with binary pixels for a static input. These pulse images display synchronized activity of neighboring neurons, and it is these images in which the information about segments, edges and texture are displayed. Pulse image generation is dependent on autowaves that travel throughout the image. In order to extend pulse Ž. image processing to multidimensional data i.e., data cubes , the autowaves are designed to expand in all of the cube's dimensions. In this fashion, pulse cubes can be created and the same analysis techniques that have been applied to two-dimensional pulse images can be applied to pulse image cubes. This paper examines and discusses multidimensional pulse image Ž. analysis applied to three-dimensional 3D chemical structural data of 17b-estradiol.

The compilation of Jiangxi General Annals was one of the most important cultural projects for Liu Kunyi, then Jiangxi's governor and a key member of the Hunan army. During the compilation process, Li Wentian, the provincial administrator of cultural and educational affairs, tried to reward the local gentry to raise a personal reputation. Liu objected to Li's approach invited Zhao Zhiqian to join this work and fully supported him in "the struggle between scholars from Jiangxi and those from other provinces". It took seven years to complete the compilation. By controlling the personnel of compilation, Liu not only proved his ability to lead great cultural undertakings, but also demonstrated his control over the situation. The compilation work has never deviated from his basic ideas, which ensured that the legitimacy of the Hunan Army Group and Liu's personal power were constructed in the book which was endorsed by the highest state power.

Pulse Coupled Neural Networks (PCNN) is a visual cortex-inspired neural networks and characterized by the global coupling and pulse synchronization of neurons. It has been proven suitable for image processing and successfully employed in image fusion. However, in most PCNN-based fusion algorithms, only single pixel value is input to motivate PCNN neuron. This is not effective enough because humans are often sensitive to features, not only pixel value. In this paper, novel orientation information is considered as features to motivate PCNN. Visual observation and objective performance evaluation criteria demonstrate that the proposed algorithm outperforms typical wavelet-based, lapacian pyramid transform-based and PCNN-based fusion algorithms.

In a multi-spectral environment information about the presence of a target is manifest across the spectra. Detection of these targets requires the fusion of these different kinds of data. However , image fusion is difficult due to the large volume of data. Typically each detector channel does not provide enough information to detect the target with a significant level of confidence. Thus, each channel provides clues only and hints as to the presence of the target. The Pulse-Coupled Neural Network (PCNN) has been shown to be a very powerful image processing tool [Kinser, 96a, 96b, 96c] [Ranganath, 95] for single channel images. Its usefulness is its inherent ability to extract edges, texture, and segments from an image. The input image is converted to a sequence of pulse images , which clearly present the objects and edges in the image at different times. While the peNN does not recognise any target, it does process the data in a manner that makes recognition very easy. A common tool for analysing images is a Fourier filter. However, on 'real-world' images Fourier filters have a difficult time providing the desired results. Fortunately, Fourier filters can analyse PCNN pulse images with relative ease. The Fractional Power Filter (FPF) [Brasher, 94] is a composite Fourier filter that also allows the user to manipulate the trade-off between generalisation and discrimination that is inherent in first order filters. The PCNN produces image autowaves that travel through the image. Autowaves are expanding waves that do not reflect or refract and annihilate each other as they collide. These autowaves are the key to the extraction of pertinent information from the image as will be demonstrated. The fusion process requires analysis of each channel of the image and the combination of these analyses. Thus, the image fusion process proposed here will allow multiple PCNNs to create intra-channel T. Lindblad et al.

Safety concerns in the operation of autonomous aerial systems require safe-landing protocols be followed during situations where the mission should be aborted due to mechanical or other failure. This article presents a pulse-coupled neural network (pcnn) to assist in the vegetation classification in a vision-based landing site detection system for an unmanned aircraft. We propose a heterogeneous computing architecture and an Opencl implementation of a pcnn feature generator. Its performance is compared across Opencl kernels designed for cpu, gpu, and fpga platforms. This comparison examines the compute

For over a decade, the Pulse Coupled Neural Network (PCNN) based algorithms have been successfully used in image interpretation applications including image segmentation. There are several versions of the PCNN based image segmentation methods, and the segmentation accuracy of all of them is very sensitive to the values of the network parameters. Most methods treat PCNN parameters like linking coefficient and primary firing threshold as global parameters, and determine them by trial-and-error. The automatic determination of appropriate values for linking coefficient, and primary firing threshold is a challenging problem and deserves further research. This paper presents a method for obtaining global as well as local values for the linking coefficient and the primary firing threshold for neurons directly from the image statistics. Extensive simulation results show that the proposed approach achieves excellent segmentation accuracy comparable to the best accuracy obtainable by trial-and-error for a variety of images.

For over a decade, the Pulse Coupled Neural Network (PCNN) based algorithms have been successfully used in image interpretation applications including image segmentation. There are several versions of the PCNN based image segmentation methods, and the segmentation accuracy of all of them is very sensitive to the values of the network parameters. Most methods treat PCNN parameters like linking coefficient and primary firing threshold as global parameters, and determine them by trial-and-error. The automatic determination of appropriate values for linking coefficient, and primary firing threshold is a challenging problem and deserves further research. This paper presents a method for obtaining global as well as local values for the linking coefficient and the primary firing threshold for neurons directly from the image statistics. Extensive simulation results show that the proposed approach achieves excellent segmentation accuracy comparable to the best accuracy obtainable by trial-an...

Pulse coupled neural network: PCNN is an algorithm pattern imitating the visual mechanism of living beings, has been widely used in the fields of image processing. As image fusion is the main part of PCNN image processing, the study on how to improve the algorithm performance of image fusion becomes hot. We present adaptive PCNN image fusion method in accordance with PCNN parameter automatic adjustment of local mean and variance. The experimental results show that, in terms of visual quality and objective evaluation, the adaptive PCNN fusion images do better than those in pyramid algorithms of image fusion, principal component analysis fusion method and wavelet transform method, etc. It has also some value on broadening the theoretical study and practical application of PCNN.

The paper puts forward an image de-noising method based on 2D wavelet transform with the application of the method in agricultural data collection system. As the there are influences of various factors in the collection process through wireless image sensor network, the detail signals of each scale are obtained from multi-scale analysis to replace the original signals with smooth low-frequency signals by applying 2D wavelet transform in de-nosing the images collected. The experiment result shows that the application of 2D wavelet transform image de-noising algorithm can achieve good subjective and objective image quality and help to collect high quality data and analyze the images for the data center with optimum effects.

This work involved human subjects or animals in its research. Approval of all ethical and experimental procedures and protocols was granted by the Center for Regional Development, The Scientific Ethics Committee, under Reference No. 21063693.

We describe the concept of a vision system based on an optoelectronic hardware neural processor. The proposed system is composed of a pulse coupled neural network (PCNN) preprocessor stage that converts an input image into a temporal pulsed pattern. These pulses are inputs to the optical broadcast neural network (OBNN) processor, which classifies the input pattern between a set of reference patterns based on a pattern matching strategy. The PCNN is to provide immunity to the scale, rotation, and translation of objects in the image. The OBNN provides high parallelism and a high speed hardware neural processor.

Sign language is the language used by deaf people to communicate with one another and with common people. An interpreter is typically required when an ordinary person wishes to communicate with a deaf person. Visual sign language recognition is a dynamic area of research in computer vision. A functional sign language recognition (SLR) system can provide an opportunity for a deaf person to communicate with non-signed persons without the need for an interpreter. It can be used to produce speech or text that makes the mute more autonomous. Arabic Sign Language Recognition (ArSLR) is an important area of study. It helps to eliminate barriers between the deaf and the community. This paper includes a review of the Arabic Sign Language Recognition System (ArSLRS). Present challenges and future research opportunities are also illustrated.

Multiresolution signal analysis via orthogonal discrete wavelet transform (ODWT) separates signals into lower and higher frequency contents. For images as 2-D signals, each decomposition level consists of a low pass filtered partrepresenting the approximation of the image and high pass filtered parts regarding the details such as noise, edges, sharp transitions between intensity values of pixels etc. of the image. In this study, we propose a new method for extraction of the necessary information from high and mid frequency bands. For this reason, relative energy and entropy concept is applied to the subimages of the original image in order to understand which frequency bands should be further decomposed via multiresolution signal analysis.

Some objects in specific poses cannot be distinguished using a single view. A model is proposed and developed for 3D object recognition based on multipleviews; it was applied on hand postures recognition. A pulse-coupled neural network is used to generate features vector for single view. Two views with different view angles are used; each view generates its features' vector. The two 2D-vectors are then linearly combined into one 3D vector. The hand postures are then combined to construct a dynamic gesture (word). The reconstruction is performed using best-match search algorithm. The experiment was conducted on 50 words and the result was 96% recognition accuracy confirming objects dataset offline extendibility.

Accurate removal of magnetic resonance imaging (MRI) signal outside the brain, a.k.a., skull stripping, is a key step in the brain image pre-processing pipelines. In rodents, this is mostly achieved by manually editing a brain mask, which is time-consuming and operator dependent. Automating this step is particularly challenging in rodents as compared to humans, because of differences in brain/scalp tissue geometry, image resolution with respect to brain-scalp distance, and tissue contrast around the skull. In this study, we proposed a deep-learning-based framework, U-Net, to automatically identify the rodent brain boundaries in MR images. The U-Net method is robust against inter-subject variability and eliminates operator dependence. To benchmark the efficiency of this method, we trained and validated our model using both in-house collected and publicly available datasets. In comparison to current state-of-the-art methods, our approach achieved superior averaged Dice similarity coefficient to ground truth T2-weighted rapid acquisition with relaxation enhancement and T2 *-weighted echo planar imaging data in both rats and mice (all p < 0.05), demonstrating robust performance of our approach across various MRI protocols.

As a large-scale water conservancy project with a long history on the Chengdu Plain, Dujiangyan can maintain operation for a long time and bring benefits steadily without the corresponding water management system and operation mechanism. However, the traditional management method is only designed to make Dujiang generally function, and it is difficult to take into account and resolve the contradictions and disputes between the various branches and weirs. During the period of the Republic of China, the water management organization in Dujiangyan Irrigation District changed, and counties in the irrigation district have successively set up water conservancy committees under the instructions of the Sichuan Provincial Government to take charge of the management of the Minyan, performing government affairs. Although the county water conservancy committees in the Dujiangyan Irrigation District have clear and unified organizational settings and management methods, when the state power penetrated into water management, they were initially resisted by local gentlemen. With the efforts of the Sichuan Provincial Government and the county water conservancy committees, the water conservancy cause in Dujiangyan Irrigation District has been better developed, and the water conservancy society has been more accepted by local gentry. We can see from this change that modern state power is the continuous expansion of government management and the transformation of traditional water management from the traditional to modern.

Oscillatory neural networks are a recent approach for applications in image segmentation. In this context, the LEGION (Locally Excitatory Globally Inhibitory Oscillator Network) is the most consistent proposal. As positive aspects, the network has got parallel architecture and capacity to separate the segments in time. On the other hand, the structure based on differential equations presents high computational complexity and limited capacity of segmentation, which restricts practical applications. In this paper, a proposal of parallel architecture for implementation of an oscillatory neural network suitable for image segmentation is presented. The proposed network keeps the positive features of the LEGION network, offering lower complexity for implementation in digital hardware and capacity of segmentation not limited, as well as few parameters, with intuitive setting. Preliminary results confirm the successful operation of the proposed network in applications of image segmentation.

Spiking cortical model (SCM) is applied to image segmentation. A natural image is processed to produce a series of spike images by SCM, and the segmented result is obtained by the integration of the series of spike images. An appropriate maximum iterative times is selected to achieve an optimal thresh old of SCM. In each iteration, neurons that produced spikes correspond to pixels with an intensity of the input natural image approximately. SCM synchronizes the output spikes via the fast linking synaptic modulation, which makes objects in the image as homogeneous as possible. Experimental results show that the output image not only separates objects and background well, but also pixels in each object are homogeneous. The proposed method performs well over other methods and the quantitative metrics are consistent with the visual performance.

The use of general descriptive names, registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use.

The FE method is the main tool used in industry to simulate large scale forming processes and it gives good results. Plastometric tests at various deformation conditions (temperatures, strain rates, etc.) combined with an inverse analysis to eliminate various heterogeneities are usually performed to obtain accurate flow stress data necessary for the FE analysis. Since large scale problems containing bilions of grains are usually considered, the major assumtion of this approach is that behavior and interaction of particular grains is homogenized in the form of a single flow stress model [1]. This procedure is well established and widely used to improve and develop new manufacturing cycles composed of processes of rolling, forging, stamping, etc.

A new method that exploits shape to localize the auroral oval in satellite imagery is introduced. The core of the method is driven by the linear least-squares (LLS) randomized Hough transform (RHT). The LLS-RHT is a new fast variant of the RHT suitable when not all necessary conditions of the RHT can be satisfied. The method is also compared with the three existing methods for aurora localization, namely the histogram-based k-means [C.C. Hung, G. Germany, K-means and iterative selection algorithms in image segmentation,

Se presenta un esquema para el reconocimiento del lenguaje de señas usando técnicas de extracción de parámetros basados en firmas de correlación rotacional que toman como información de entrada imágenes en dos dimensiones y generan señales en una dimensión. Uno de los principales problemas del reconocimiento de lenguaje de señas es la extracción adecuada de la Región de Interés (ROI por sus siglas en inglés), este trabajo contempla el uso de un sistema de visión por computadora que evita la implementación de sensores electrónicos y marcadores visuales o físicos que limitan la practicidad de generar los gestos del lenguaje de señas, permitiendo así una expresión fluida y natural del individuo que se esté analizando. Se eligieron los signos de los dígitos del 1 al 9 del lenguaje de señas empleando un perceptrón multicapa con validación cruzada alcanzado un 100%.  Se presenta un esquema para el reconocimiento del lenguaje de señas usando técnicas de extracción de parámetros basados en firmas de correlación rotacional que toman como información de entrada imágenes en dos dimensiones y generan señales en una dimensión. Uno de los principales problemas del reconocimiento de lenguaje de señas es la extracción adecuada de la Región de Interés (ROI por sus siglas en inglés), este trabajo contempla el uso de un sistema de visión por computadora que evita la implementación de sensores electrónicos y marcadores visuales o físicos que limitan la practicidad de generar los gestos del lenguaje de señas, permitiendo así una expresión fluida y natural del individuo que se esté analizando. Se eligieron los signos de los dígitos del 1 al 9 del lenguaje de señas empleando un perceptrón multicapa con validación cruzada alcanzado un 100%.

Pulse coupled neural network: PCNN is an algorithm pattern imitating the visual mechanism of living beings, has been widely used in the fields of image processing. As image fusion is the main part of PCNN image processing, the study on how to improve the algorithm performance of image fusion becomes hot. We present adaptive PCNN image fusion method in accordance with PCNN parameter automatic adjustment of local mean and variance. The experimental results show that, in terms of visual quality and objective evaluation, the adaptive PCNN fusion images do better than those in pyramid algorithms of image fusion, principal component analysis fusion method and wavelet transform method, etc. It has also some value on broadening the theoretical study and practical application of PCNN.

Bioinformatics-Trends and Methodologies 632 2. Background 2.1 Pattern recognition, machine learning and data mining Pattern recognition can be defined as the categorization of the input data into identifiable classes via the extraction of significant features or attributes of the data from a background of irrelevant detail (Duda et al, 2000). The task of pattern recognition is also viewed as the transformation from the measurement space to the feature space and finally to a decision space. Machine learning techniques aim at producing a system that can learn and adapt from the environment and hence exhibits a kind of intelligence essential for applications that lack known solutions (Alpydin, 2004). Machine learning models very often attempt to optimize a criterion function through exploiting information from training examples. Data mining, on the other hand, can be thought of as a collection of statistical, machine learning, pattern recognition and artificial intelligence tools that help uncover and extract 'hidden' knowledge from data. Particularly in the medical domain data mining refers often to techniques and methods that analyze large amounts of data. These techniques include among many others classification, clustering, association rule mining and regression or prediction. Cluster analysis usually addresses segmentation problems. The objective of this analysis is to separate data with similar characteristics from the dissimilar ones. Cluster analysis is frequently the first required task of the mining process. Cluster analysis can also be used for outlier detection to identify samples with peculiar behavior. Among the most simple and efficient clustering techniques are K-means, fuzzy K-means, Self Organizing maps; in addition to more advanced clustering methods like evolving clustering techniques and distributed clustering. The purpose of association rule mining, on the other hand, is to search for the most significant relationship across large number of variables or attributes. Sometimes, association is viewed as one type of dependencies where affinities of data items are described (e.g., describing data items or events that frequently occur together or in sequence). Some techniques for association analysis are nonlinear regression, rule induction, Apriori algorithm and Bayesian networks. Time Series prediction is also an important aspect in data mining whereby the temporal structure and ordering of the data is utilized to estimate some future value based on current and past data samples. Time-series prediction encompasses a wide variety of applications. As mentioned earlier, the purpose of this chapter is to provide a broad introduction to the fundamentals of machine learning suitable for bioinformatics. The rest of the chapter will mainly focus on the classification problem. 2.2 Machine learning models for classification Classification is usually referred to as the process of devising models that can predict categorical (discrete, unordered) class labels. Often machine learning models are used for these purposes that learn the class functions using a set of given training examples. Popular machine learning classification models are decision tree classifiers, Bayesian classifiers, Bayesian belief networks, rule based classifiers and Backpropagation-Multi layer neural network (Hand et. al, 2001). More recent approaches to classification include support vector machines and ensemble methods. In addition, other approaches are frequently encountered in the literature like k-nearest-neighbor classifiers, case-based reasoning, genetic algorithms, rough sets and fuzzy logic techniques. According to a recent ranking (KDnuggets : Polls, 2006) common classification models used in the data mining community are decision trees, decision rules, logistic regression, artificial www.intechopen.com Novel Machine Learning Techniques for Micro-Array Data Classification

Bu çalışma, basıncın sıvıların kaynama sıcaklığı üzerine etkisini öğretmede öğretmene rehberlik edecek bir çalışma yaprağı geliştirmek amacıyla yapılmıştır. Çalışmanın başlangıç aşamasında lise seviyesinde toplam 36 öğrenciyle bireysel ve grup mülakatları yürütülerek, basıncın kaynama sıcaklığına etkisiyle ilgili yanılgılar tespit edilmiştir. Yanılgıları gidermek ve etkili kavram öğretimini sağlamak amacıyla konuyla ilgili bir çalışma yaprağı geliştirilmiştir. Materyalin geliştirilmesinde bütünleştirici (constructivist) öğrenme teorisi göz önünde bulundurulmuş ve hazırlama safhasında öğrencilerin basınç-kaynama ilişkisi ile ilgili yanılgıları dikkate alınmıştır. Uygulama, 24 kişilik Lise 2 öğrenci grubu ile birlikte gerçekleştirilmiştir. Uygulama sonunda materyalin değerlendirme bölümündeki sorulara öğrencilerin verdikleri cevaplardan elde edilen bulgular, öğrencilerin sahip olduğu kavram yanılgılarını gidermede ve basınç-kaynama ilişkisini anlamada etkili olduğunu kanıtlamıştır. Elde edilen sonuçlar ışığında, materyalin öğretim sürecinde etkili bir şekilde uygulanabilmesine yönelik bazı önerilerde bulunulmuştur.

A Pulse Coupled Neural Network (PCNN) is proposed as a numerical model of cat visual cortex, and it has been applied to the engineering fields especially in an image processing, e.g., segmentation, edge enhancement, and so on. The PCNN model consists of neurons with two kind of inputs, namely feeding input and linking input and they each have a lot of parameters. The Parameters are used to be defined empirically and the optimization of parameters has been known as one of the remaining problem of PCNN. According to the recent studies, parameters in PCNN will be able to be given using parameter learning rule or evolutionary programming. However these methods require teaching images for the learning. In this study, we propose a parameter adjustment method of PCNN for the image segmentation. The proposed method changes the parameters through the iterations of trial of segmentation and the method doesn't require any teaching signal or teaching pattern. The successful results are obtained in the simulations, and we conclude that the proposed method shows good performance for the parameter adjustment of PCNNs.

As the size and complexity of a software system increase, the principal design decisions and the global structure of a system become more important than the selection of specific algorithms and data structures in order to determine its quality.

Bug-Tracking mechanism is employed only is some of the large software development houses. Most of the others never bothered with bug tracking at all, and instead simply relied on shared lists and email to monitor the status of defects. This procedure is errorprone and tends to cause those bugs judged least significant by developers to be dropped or ignored. Bug-Tracking System is an ideal solution to track the bugs of a product, solution or an application. Bug Tacking System allows individual or groups of developers to keep track of outstanding bugs in their product effectively. This can also be called as Defect Tracking System. The Bug Tracking System can dramatically increase the productivity and accountability of individual employees by providing a documented work flow and positive feedback for good performance.

Education in the 21st century equips students with knowledge and information and the success of achieving academic achievements during the learning process. Students' academic achievement can be seen from various aspects: the Grade Point Average. So far, efforts to predict GPA have not been made. In fact, if the student's Grade Point Average can be predicted from an early age, the study program can implement a policy to improve graduates' quality and make planning, study escort, and guidance more intensive. Based on this urgency, this study aims to produce a predictive model for the GPA of STMIK Amik Riau students in the odd semester of 2019, using the Backpropagation Neural Network algorithm and Multiple Linear Regression. Backpropagation's architectural model is 8 architectures, and 4-5-1 is the best architectural model with MSE at the time of training = 0.00099965532 and MSE during network validation = 0.0038793 with an epoch of 102 iterations and the resulting accuracy value of 95.24%. Meanwhile, the GPA prediction results, after testing using the Multiple Linear Regression algorithm, obtained an MSE value of 0. 0.27966667%, with a Multiple Correlation coefficient (R) of R = 0.9774925 and a coefficient of determination (R2) = 0.95549159. Thus the prediction of student GPA using MLR is accurate because the value of the coefficient of determination (R2) is close to 1.

Resumen. Realizar pronósticos de múltiples pasos en el tiempo no es una tarea sencilla; por lo tanto, es importante desarrollar y usar técnicas que simplifiquen esta tarea. Este trabajo prueba modelos de Inteligencia Artificial para tres diferentes estrategias de pronósticos a múltiples pasos: Iterativa, Directa, y Multiple Input-Multiple Output (MIMO). El primer paso consiste en preprocesar las series de tiempo para limpiar (eliminar datos atípicos y ruido) y transformar los datos en secuencias o imágenes; después, se usan tres tipos de Artificial Neural Networks (ANN) para construir modelos de pronóstico: Multilayer Perceptron (MLP), Long Short Term Memory (LSTM) y Convolutional Neural Networks (CNN). Se muestran experimentos con series de tiempo obtenidas en estaciones climáticas en Michoacán, México, donde se puede observar el desempeño que se obtuvo con cada modelo y estrategia.

Ureteroscopy has evolved into a routine technique for treatment of kidney stones. Laser lithotripsy is commonly used to fragment the kidney stones until they are small enough to be removed. Poor image quality, presence of floating debris and severe occlusions in the endoscopy video make it difficult to target stones during the ureteroscopy procedure. A potential solution is automated localization and segmentation of the stone fragments. However, the heterogeneity of stones in terms of shape, texture, as well as colour and the presence of moving debris make the task of stone segmentation challenging. Further, dynamic background, motion blur, local deformations, occlusions and varying illumination conditions need to be taken into account during segmentation. To address these issues, we compliment state-of-the-art U-Net based segmentation strategy with the learned motion information. This technique leverages difference in motion between the large stones and surrounding debris and additionally tackles problems due to illumination variability, occlusions and other factors that are present in the frame-of-interest. The proposed motion induced U-Net (MI-UNet) architecture consists of two main components: 1) U-Net and 2) DVFNet. The quantitative results show consistent performance and improvement over most evaluation metrics. The qualitative validation also illustrate that our complimentary DVF computation is able to effectively reduce the effect of surrounding debris in contrast to U-Net.

Este proyecto plantea el desarrollo de técnicas de modelado, diseño, análisis, optimización y testeo de sistemas embebidos con restricciones de tiempo y requerimientos heterogéneos, para diferentes plataformas de software y hardware, en la búsqueda de un apropiado balance entre prestaciones, rendimiento y eficiencia.

Submitted for the DFD14 Meeting of The American Physical Society On the wake dynamics of flapping inverted flags 1 ANVAR GILMANOV, FOTIS SOTIROPOULOS, Univ of Minn-Minneapolis, JULIA COSSE, MORY GHARIB, California Institute of Technology-As recently shown experimentally by Kim et al. (JFM, 2013), when a flexible flag with a fixed trailing edge (an inverted flag) is exposed to a uniform inflow it can exhibit complex structural response and rich fluid-structure interaction (FSI) dynamics. We employ a new FSI numerical method to carry out large-eddy simulation (LES) of inverted flags in the range where large-amplitude flapping instabilities have been found experimentally. The numerical method integrates the curvilinear immersed boundary (CURVIB) FSI method of Borazjani et al. (JCP, 2008) with the thin-shell, rotationfree, finite-element (FE) formulation of Stolarski et al. (Int. JNME, 2013) and is able to simulate FSI of flexible thin bodies undergoing oscillations of arbitrarily large amplitude. The dynamic Smagorinsky model is employed for subgrid scale closure and a wall model is employed for reconstructing velocity boundary conditions. Comparisons with the experimental data show that the simulations are able to capture the structural response of the flag with very good accuracy. The computed results are analyzed to elucidate the structure and dynamics of the massively separated, unsteady flow shed off the flag edges.

A swarm model that induces emergent collective behaviors from the individuals in the swarm is proposed for the planar formations of multiple autonomous unmanned aerial vehicles (UAVs). The stability of the formations, or cohesiveness of the swarm, is guaranteed via the the Direct Method of Lyapunov which is used to construct the instantaneous velocity of each individual. A Lyapunov-like function, from which the velocity controllers are constructed, expresses the three well-known Reynolds' flocking rules as artificial potential fields for interindividual attraction and avoidance. Basic patterns of planar formation which are similar to emergent behaviors distinctive of swarming in nature are demonstrated via computer simulations. Different emergent patterns are obtained with variations in the system parameters.

, except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden. The use in this publication of trade names, trademarks, service marks and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights.

The aim of this paper is to develop a finite element procedure for crack prediction in vibrating beams. Based on this procedure, full frictional contact conditions are introduced between the crack surfaces in order to consider the breathing of crack. The region surrounding the crack is simulated by two-dimensional finite elements. An incremental-iterative procedure is employed to solve the nonlinear dynamic equations governing this problem. The obtained time response is processed with Fast Fourier Transform to extract its frequency components. The first three natural frequencies are input to a trained Artificial Neural Network for depth and position prediction of the crack. This study is validated for a dynamic loading cantilever beam. It is found that the proposed procedure is capable of predicting the crack depth and position with high accuracy.

This paper may be helpful to handicapped people can live with ease life. The wheelchair can be operated with the help of any moving body part having red colour pointer. It makes the handicapped people as a independent. System comprises of integration of Red colour pointer operation using digital image processing and embedded technology. Disability affects hundreds of millions of families in developing countries. According to the World Health Organization (WHO), Currently around 10 per cent of the total world&#39;s population, or roughly 650 million people, live with a disability. In indiadisabled population has increased by 22.4% between 2001 and 2011. The number of disabled, which was 2.19 crore in 2001, rose in 2011 to 2.68 crore-1.5 crore males and 1.18 crore females. This indicates that the most common disability is motor, followed by blindness, deafness, intellectual and language. Wheelchairs is necessary for disable persons to move from one place to other i.e indoor or outdoor...

Besides the commercial and surveillance applications of drones increasing enormously, there is no doubt in the efficacy. This article is on surveillance check and implementation of security system for ATM robbery. The recent incidences, challenging the security and safety measures currently used at ATM machines. In this paper, whenever robbery occurs, vibration sensor synchronised with GSM module and PIC controller alerts the management system, as well as closes the door and sprays Anaesthesia to make the thief unconscious using stepper motor synchronised with PIC controller. The application of drone is to track the thief, even he/she escapes from the encounter. Camera continuously process and sends the video to the PC and it will be saved in computer. In this proposal, the main idea is to save the money from robber(s) besides capturing them using a sensor system and drone.

Perkembangan teknologi semakin canggih dengan adanya internet. Internet yang dapat dengan mudah diakses untuk mencari informasi dan dokumen dapat memicu adanya tindak plagiarisme.  Setiap dokumen yang melakukan tindakan plagiarisme akan susah dikenali tanpa adanya sistem yang dapat mengenali kesamaan dokumen. Sistem yang dapat mendeteksi plagiarisme dengan mencari kemiripan pada dokumen dibutuhkan. Dalam penelitian ini digunakan algoritma  Nazief & Andriani dalam proses penentuan kemiripan pada dokumen teks dan algoritma  Smith-Waterman untuk mengidentifikasi kesamaan yang paling signifikan ( local alignment ) dari dua buah rangkaian sekuens string. Hasil akhir yang didapatkan perbandingan dua  sequence  dengan bantuan preprocessing memiliki tingkat perhitungan similarity  yang lebih besar dalam mendeteksi kemiripan dokumen.  Kata Kunci:  Dokumen, Plagiarisme, Algoritma  Smith-Waterman , Algoritma  Nazief & Andriani

Fire and explosion protection in industrial conditions requires multidimensional approach. Usually the risk of hazardous zone creation is unavoidable, if the combustible material is processed. Therefore controlling of potential ignition sources is introduced. One of most popular sources of ignition is electrostatic discharge. Depending on the type of the discharge, as well as on exact discharge conditions, energy released might reach hundreds or even thousands of mJ, being able to ignite most of gaseous or dust-air hazardous mixtures. A dedicated methodology was created to record the discharge with fast camera with maximum speed of 1M fps and with the oscilloscope up to 25 GS/s. Dedicated test stand allows to obtain high voltage to create the conditions for electrostatic discharge. The aim of presented research was to analyze the course of electrostatic spark discharge in laboratory conditions and to place the outcomes in the context of explosion safety in the industrial conditions....

Water, the primary source of human existence, vastly occurs in the earth, our home. The major share of water resource comes from ocean and sea which is saline in nature. The remaining constitutes the freshwater or surface water resources on which we depend for our day to day life for domestic uses and consumption. In this context it is very necessary that better quality water ensure for human population and the environmental protection. Among various quality indices, Total Dissolved Solvents are considered as predominant that accounts for hardness and salinity, for the salt organic and inorganic constituents present. The lack of knowledge about the contents prevents common man to use such water or ill treatment of water bodies results in misuse of water resources. The work focuses on an alternative by identifying the solvents present in the water body ensuring to desalinate the particular ionic compound or metal.

In teaching and learning languages, it is common to use code-switching. Code-switching help speaker and hearer share the ability to speak more than one language. This research was focused on the code-switching that was done by the lecturer, the most common type that appeared in the classroom interaction and the function of the code-switching found in the analysis. Descriptive qualitative research was applied in this research which the data got from the interaction of the lecturer and students in the classroom in teaching and learning English for Arabic students. Based on the analysis, in the classroom interaction, the three types of code-switching was found; tag-switching, inter-sentential switching and intra-sentential (Romaine, 1995). The most common type found in the interaction was intra-sentential switching. The used of code-switching in the classroom interaction was to improve language skills, give clarification, emphasis the topic, and make good relation with interlocutor tha...

In this paper, it was tried to provide a comprehensive analysis of the term &quot;artificial intelligence&quot; and its essence definitions existing in the scientific literature. It is noted that nowadays, they developed several approaches to the interpretation of &quot;artificial intelligence&quot; definitions. At the same time, despite a wide range of studies on the interpretation, various areas of application, as well as the legal aspects of &quot;artificial intelligence&quot; concept, the unexplored horizons of this definition are quite wide. For this reason, the level of interest in the multidimensional consideration and analysis of the term &quot;artificial intelligence&quot; is quite high, which allows us to talk about fairly broad prospects for the development of legal approaches to its study. It is shown that the most convincing and consistent position of those scientists who are inclined to describe the concept under consideration by designating its essential properties an...

English. Presence of regional and minority languages over digital media is an indicator of their vitality. In this paper, we want to investigate quantitative aspects of the use on Facebook of the Sardinian language. In particular, we want to focus on the coexistence of diatopic varieties. We extracted linguistic data from public pages and, through the translation of the most frequent words, we find out similarities and differences between varieties. Italiano. La presenza e l' uso delle lingue regionali e minoritarie sui mezzi digitaliè un indicatore della loro vitalità. In questo lavoro vogliamo concentrarci sugli aspetti quantitativi del sardo usato su Facebook. In particolare, vogliamo analizzare le varietà diatopiche estraendo i dati linguistici dalle pagine pubbliche. Mediante la traduzione delle parole più frequenti abbiamo trovato similarità e differenze tra le varietà.