Search results for: LEON Processor

Authors: Qian Liu, Steve Furber

Abstract:

To explore how the brain may recognise objects in its general,accurate and energy-efficient manner, this paper proposes the use of a neuromorphic hardware system formed from a Dynamic Video Sensor (DVS) silicon retina in concert with the SpiNNaker real-time Spiking Neural Network (SNN) simulator. As a first step in the exploration on this platform a recognition system for dynamic hand postures is developed, enabling the study of the methods used in the visual pathways of the brain. Inspired by the behaviours of the primary visual cortex, Convolutional Neural Networks (CNNs) are modelled using both linear perceptrons and spiking Leaky Integrate-and-Fire (LIF) neurons. In this study’s largest configuration using these approaches, a network of 74,210 neurons and 15,216,512 synapses is created and operated in real-time using 290 SpiNNaker processor cores in parallel and with 93.0% accuracy. A smaller network using only 1/10th of the resources is also created, again operating in real-time, and it is able to recognise the postures with an accuracy of around 86.4% - only 6.6% lower than the much larger system. The recognition rate of the smaller network developed on this neuromorphic system is sufficient for a successful hand posture recognition system, and demonstrates a much improved cost to performance trade-off in its approach.

Keywords: Spiking neural network (SNN), convolutional neural network (CNN), posture recognition, neuromorphic system.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2001

Authors: Jia-Rong Yeh, Ai-Hsien Li, Jiann-Shing Shieh, Yen-An Su, Chi-Yu Yang

Abstract:

In this paper, an automatic detecting algorithm for QRS complex detecting was applied for analyzing ECG recordings and five criteria for dangerous arrhythmia diagnosing are applied for a protocol type of automatic arrhythmia diagnosing system. The automatic detecting algorithm applied in this paper detected the distribution of QRS complexes in ECG recordings and related information, such as heart rate and RR interval. In this investigation, twenty sampled ECG recordings of patients with different pathologic conditions were collected for off-line analysis. A combinative application of four digital filters for bettering ECG signals and promoting detecting rate for QRS complex was proposed as pre-processing. Both of hardware filters and digital filters were applied to eliminate different types of noises mixed with ECG recordings. Then, an automatic detecting algorithm of QRS complex was applied for verifying the distribution of QRS complex. Finally, the quantitative clinic criteria for diagnosing arrhythmia were programmed in a practical application for automatic arrhythmia diagnosing as a post-processor. The results of diagnoses by automatic dangerous arrhythmia diagnosing were compared with the results of off-line diagnoses by experienced clinic physicians. The results of comparison showed the application of automatic dangerous arrhythmia diagnosis performed a matching rate of 95% compared with an experienced physician-s diagnoses.

Keywords: Signal processing, electrocardiography (ECG), QRS complex, arrhythmia.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1476

Authors: María Araya Léon, Ainoa Abella, Aura Murillo, Ricardo Guasch, Laura Clèries

Abstract:

This article aims to examine scientific theory framed within the term hospitals of the future from a multidisciplinary and cross-sectional perspective. To understand the connection that the various cross-sectional areas, we studied have with architectural spaces and to determine the future outlook of the works examined and how they can be classified into the categories of need/solution, evolution/revolution, collective/individual, and preventive/corrective. The changes currently taking place within the context of healthcare demonstrate how important these projects are and the need for companies to face future changes. A systematic review has been carried out focused on what will the hospitals of the future be like in relation to the elements that form part of their use, design, and architectural space experience, using the WOS database from 2016 to 2019. The large number of works about sensoring & big data and the scarce amount related to the area of materials is worth highlighting. Furthermore, no growth concerning future issues is envisaged over time. Regarding classifications, the articles we reviewed address evolutionary and collective solutions more, and in terms of preventive and corrective solutions, they were found at a similar level. Although our research focused on the future of hospitals, there is little evidence representing this approach. We also detected that, given the relevance of the research on how the built environment influences human health and well-being, these studies should be promoted within the context of healthcare. This article allows to find evidence on the future perspective from within the domain of hospital architecture, in order to create bridges between the productive sector of architecture and scientific theory. This will make it possible to detect R&D opportunities in each analyzed cross-section.

Keywords: Hospitals, trends, architectural space, disruptive approach.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 213

Authors: Citlali Aguilera Lira, David Lynch Steinicke, Andrea León Garcia

Abstract:

One of the challenges that higher education faces is to find how to approach the sustainability in an inclusive way to the student within all the different academic areas, how to move the sustainable development from the abstract field to the operational field. This research comes from the ecoliteracy and the pedagogical praxis as tools for rebuilding the teaching processes inside of universities. The purpose is to determine and describe which are the factors involved in the process of learning particularly in the Greenhouse-School Siembra UV. In the Greenhouse-School Siembra UV, of the University of Veracruz, are cultivated vegetables, medicinal plants and small cornfields under the usage of eco-technologies such as hydroponics, Wickingbed and Hugelkultur, which main purpose is the saving of space, labor and natural resources, as well as function as agricultural production alternatives in the urban and periurban zones. The sample was formed with students from different academic areas and who are actively involved in the greenhouse, as well as institutes from the University of Veracruz and governmental and nongovernmental departments. This project comes from a pedagogic praxis approach, from filling the needs that the different professional profiles of the university students have. All this with the purpose of generate a pragmatic dialogue with the sustainability. It also comes from the necessity to understand the factors that intervene in the students’ praxis. In this manner is how the students are the fundamental unit in the sphere of sustainability. As a result, it is observed that those University of Veracruz students who are involved in the Greenhouse-school, Siembra UV, have enriched in different levels the sense of urban and periurban agriculture because of the diverse academic approaches they have and the interaction between them. It is concluded that the ecotechnologies act as fundamental tools for ecoliteracy in society, where it is strengthen the nutritional and food security from a sustainable development approach.

Keywords: Farming eco-technologies, food security, multidisciplinary, pedagogical praxis.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1564

Authors: Laura Elena De León Prado, Dora Alicia Cortés Hernández, Javier Sánchez

Abstract:

Among the different cancer treatments that are currently used, hyperthermia has a promising potential due to the multiple benefits that are obtained by this technique. In general terms, hyperthermia is a method that takes advantage of the sensitivity of cancer cells to heat, in order to damage or destroy them. Within the different ways of supplying heat to cancer cells and achieve their destruction or damage, the use of magnetic nanoparticles has attracted attention due to the capability of these particles to generate heat under the influence of an external magnetic field. In addition, these nanoparticles have a high surface area and sizes similar or even lower than biological entities, which allow their approaching and interaction with a specific region of interest. The most used magnetic nanoparticles for hyperthermia treatment are those based on iron oxides, mainly magnetite and maghemite, due to their biocompatibility, good magnetic properties and chemical stability. However, in order to fulfill more efficiently the requirements that demand the treatment of magnetic hyperthermia, there have been investigations using ferrites that incorporate different metallic ions, such as Mg, Mn, Co, Ca, Ni, Cu, Li, Gd, etc., in their structure. This paper reports the synthesis of nanosized Mg_xMn_1-xFe₂O₄ (x = 0.3 and 0.4) ferrites by sol-gel method and their evaluation in terms of heating capability and in vitro hemolysis to determine the potential use of these nanoparticles as thermoseeds for the treatment of cancer by magnetic hyperthermia. It was possible to obtain ferrites with nanometric sizes, a single crystalline phase with an inverse spinel structure and a behavior near to that of superparamagnetic materials. Additionally, at concentrations of 10 mg of magnetic material per mL of water, it was possible to reach a temperature of approximately 45°C, which is within the range of temperatures used for the treatment of hyperthermia. The results of the in vitro hemolysis assay showed that, at the concentrations tested, these nanoparticles are non-hemolytic, as their percentage of hemolysis is close to zero. Therefore, these materials can be used as thermoseeds for the treatment of cancer by magnetic hyperthermia.

Keywords: Ferrites, heating capability, hemolysis, nanoparticles, sol-gel.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 862

Authors: David M. Jenkins Jr.

Abstract:

Scale Time Offset Robust Modulation (STORM) [1]– [3] is a high bandwidth waveform design that adds time-scale to embedded reference modulations using only time-delay [4]. In an environment where each user has a specific delay and scale, identification of the user with the highest signal power and that user-s phase is facilitated by the STORM processor. Both of these parameters are required in an efficient multiuser detection algorithm. In this paper, the STORM modulation approach is evaluated with a direct sequence spread quadrature phase shift keying (DS-QPSK) system. A misconception of the STORM time scale modulation is that a fine temporal resolution is required at the receiver. STORM will be applied to a QPSK code division multiaccess (CDMA) system by modifying the spreading codes. Specifically, the in-phase code will use a typical spreading code, and the quadrature code will use a time-delayed and time-scaled version of the in-phase code. Subsequently, the same temporal resolution in the receiver is required before and after the application of STORM. In this paper, the bit error performance of STORM in a synchronous CDMA system is evaluated and compared to theory, and the bit error performance of STORM incorporated in a single user WCDMA downlink is presented to demonstrate the applicability of STORM in a modern communication system.

Keywords: Pseudonoise coded communication, Cyclic codes, Code division multiaccess

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1594

Authors: Jason Chien-Hsun Tseng

Abstract:

This paper evaluates performances of an adaptive noise cancelling (ANC) based target detection algorithm on a set of real test data supported by the Defense Evaluation Research Agency (DERA UK) for multi-target wideband active sonar echolocation system. The hybrid algorithm proposed is a combination of an adaptive ANC neuro-fuzzy scheme in the first instance and followed by an iterative optimum target motion estimation (TME) scheme. The neuro-fuzzy scheme is based on the adaptive noise cancelling concept with the core processor of ANFIS (adaptive neuro-fuzzy inference system) to provide an effective fine tuned signal. The resultant output is then sent as an input to the optimum TME scheme composed of twogauge trimmed-mean (TM) levelization, discrete wavelet denoising (WDeN), and optimal continuous wavelet transform (CWT) for further denosing and targets identification. Its aim is to recover the contact signals in an effective and efficient manner and then determine the Doppler motion (radial range, velocity and acceleration) at very low signal-to-noise ratio (SNR). Quantitative results have shown that the hybrid algorithm have excellent performance in predicting targets- Doppler motion within various target strength with the maximum false detection of 1.5%.

Keywords: Wideband Active Sonar Echolocation, ANC Neuro-Fuzzy, Wavelet Denoise, CWT, Hybrid Algorithm.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1997

Authors: Rehab Abdullah Shendi

Abstract:

The use of customised soft-core processors in which instructions can be integrated into a system in application hardware is increasing in the Field Programmable Gate Array (FPGA) field. Specifically, the partial run-time reconfiguration of FPGAs in specialised processors for a particular domain can be very beneficial. In this report, the design and implementation for the customisation of a soft-core MIPS processor using an FPGA and partial reconfiguration (PR) of FPGA technology will be addressed to achieve efficient resource use. This can be achieved using a PR design flow that helps the design fit into a smaller device. Moreover, the impact of static power consumption could be reduced due to runtime reconfiguration. This will be done by configurable custom instructions implemented in the hardware as an extension on the MIPS CPU. The aim of this project is to investigate the PR of FPGAs for run-time adaptations of the instruction set of a soft-core CPU, including the integration of custom instructions and the exploration of the potential to use the MultiBoot feature available in Xilinx FPGAs to carry out the PR process. The system will be evaluated and tested on a Nexus 3 development board featuring a Xilinx Spartran-6 FPGA. The system will be able to load reconfigurable custom instructions dynamically into user programs with the help of the trap handler when the custom instruction is called by the MIPS CPU. The results of this experiment demonstrate that custom instructions in hardware can speed up a certain function and many instructions can be saved when compared to a software implementation of the same function. Implementing custom instructions in hardware is perfectly possible and worth exploring.

Keywords: Customisation, FPGA, MIPS, partial reconfiguration.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1144

Authors: Akim Borbuev, Francisco de León

Abstract:

Economic and population growth in densely-populated urban areas introduce major challenges to distribution system operators, planers, and designers. To supply added loads, utilities are frequently forced to invest in new distribution feeders. However, this is becoming increasingly more challenging due to space limitations and rising installation costs in urban settings. This paper proposes the conversion of critical alternating current (ac) distribution feeders into direct current (dc) feeders to increase the power transfer capacity by a factor as high as four. Current trends suggest that the return of dc transmission, distribution, and utilization are inevitable. Since a total system-level transformation to dc operation is not possible in a short period of time due to the needed huge investments and utility unreadiness, this paper recommends that feeders that are expected to exceed their limits in near future are converted to dc. The increase in power transfer capacity is achieved through several key differences between ac and dc power transmission systems. First, it is shown that underground cables can be operated at higher dc voltage than the ac voltage for the same dielectric stress in the insulation. Second, cable sheath losses, due to induced voltages yielding circulation currents, that can be as high as phase conductor losses under ac operation, are not present under dc. Finally, skin and proximity effects in conductors and sheaths do not exist in dc cables. The paper demonstrates that in addition to the increased power transfer capacity utilities substituting ac feeders by dc feeders could benefit from significant lower costs and reduced losses. Installing dc feeders is less expensive than installing new ac feeders even when new trenches are not needed. Case studies using the IEEE 342-Node Low Voltage Networked Test System quantify the technical and economic benefits of dc feeders.

Keywords: Dc power systems, distribution feeders, distribution networks, energy efficiency, power transfer capacity.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1002

Authors: Hazem M. El-Bakry

Abstract:

Neural processors have shown good results for detecting a certain character in a given input matrix. In this paper, a new idead to speed up the operation of neural processors for character detection is presented. Such processors are designed based on cross correlation in the frequency domain between the input matrix and the weights of neural networks. This approach is developed to reduce the computation steps required by these faster neural networks for the searching process. The principle of divide and conquer strategy is applied through image decomposition. Each image is divided into small in size sub-images and then each one is tested separately by using a single faster neural processor. Furthermore, faster character detection is obtained by using parallel processing techniques to test the resulting sub-images at the same time using the same number of faster neural networks. In contrast to using only faster neural processors, the speed up ratio is increased with the size of the input image when using faster neural processors and image decomposition. Moreover, the problem of local subimage normalization in the frequency domain is solved. The effect of image normalization on the speed up ratio of character detection is discussed. Simulation results show that local subimage normalization through weight normalization is faster than subimage normalization in the spatial domain. The overall speed up ratio of the detection process is increased as the normalization of weights is done off line.

Keywords: Fast Character Detection, Neural Processors, Cross Correlation, Image Normalization, Parallel Processing.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1492

Authors: Srinivasan Chandrasekaran, Thailammai Chithambaram, Shihas A. Khader

Abstract:

The early-stage damage detection in offshore structures requires continuous structural health monitoring and for the large area the position of sensors will also plays an important role in the efficient damage detection. Determining the dynamic behavior of offshore structures requires dense deployment of sensors. The wired Structural Health Monitoring (SHM) systems are highly expensive and always needs larger installation space to deploy. Wireless sensor networks can enhance the SHM system by deployment of scalable sensor network, which consumes lesser space. This paper presents the results of wireless sensor network based Structural Health Monitoring method applied to a scaled experimental model of offshore structure that underwent wave loading. This method determines the serviceability of the offshore structure which is subjected to various environment loads. Wired and wireless sensors were installed in the model and the response of the scaled BLSRP model under wave loading was recorded. The wireless system discussed in this study is the Raspberry pi board with Arm V6 processor which is programmed to transmit the data acquired by the sensor to the server using Wi-Fi adapter, the data is then hosted in the webpage. The data acquired from the wireless and wired SHM systems were compared and the design of the wireless system is verified.

Keywords: Condition assessment, damage detection, structural health monitoring, structural response, wireless sensor network.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2915

Authors: Mojiri M, Ghazi Khanloo Sani K, Moghim Beigi A

Abstract:

The aim of this study was to compare the sensitometric properties of commonly used radiographic films processed with chemical solutions in different workload hospitals. The effect of different processing conditions on induced densities on radiologic films was investigated. Two accessible double emulsions Fuji and Kodak films were exposed with 11-step wedge and processed with Champion and CPAC processing solutions. The mentioned films provided in both workloads centers, high and low. Our findings displays that the speed and contrast of Kodak filmscreen in both work load (high and low) is higher than Fuji filmscreen for both processing solutions. However there was significant differences in films contrast for both workloads when CPAC solution had been used (p=0.000 and 0.028). The results showed base plus fog density for Kodak film was lower than Fuji. Generally Champion processing solution caused more speed and contrast for investigated films in different conditions and there was significant differences in 95% confidence level between two used processing solutions (p=0.01). Low base plus fog density for Kodak films provide more visibility and accuracy and higher contrast results in using lower exposure factors to obtain better quality in resulting radiographs. In this study we found an economic advantages since Champion solution and Kodak film are used while it makes lower patient dose. Thus, in a radiologic facility any change in film processor/processing cycle or chemistry should be carefully investigated before radiological procedures of patients are acquired.

Keywords: Sensitometry, densitometry, Radiographic film, processing solution

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1585

Authors: Amira Abbes Kheriji, Faouzi Bouani, Mekki Ksouri, Mohamed Ben Ahmed

Abstract:

Model Predictive Control (MPC) is increasingly being proposed for real time applications and embedded systems. However comparing to PID controller, the implementation of the MPC in miniaturized devices like Field Programmable Gate Arrays (FPGA) and microcontrollers has historically been very small scale due to its complexity in implementation and its computation time requirement. At the same time, such embedded technologies have become an enabler for future manufacturing enterprises as well as a transformer of organizations and markets. Recently, advances in microelectronics and software allow such technique to be implemented in embedded systems. In this work, we take advantage of these recent advances in this area in the deployment of one of the most studied and applied control technique in the industrial engineering. In fact in this paper, we propose an efficient framework for implementation of Generalized Predictive Control (GPC) in the performed STM32 microcontroller. The STM32 keil starter kit based on a JTAG interface and the STM32 board was used to implement the proposed GPC firmware. Besides the GPC, the PID anti windup algorithm was also implemented using Keil development tools designed for ARM processor-based microcontroller devices and working with C/Cµ langage. A performances comparison study was done between both firmwares. This performances study show good execution speed and low computational burden. These results encourage to develop simple predictive algorithms to be programmed in industrial standard hardware. The main features of the proposed framework are illustrated through two examples and compared with the anti windup PID controller.

Keywords: Embedded systems, Model Predictive Control, microcontroller, Keil tool.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 5447

Authors: C. W. Hsu, C. N. Liang, L. Y. Ke, F. Y. Huang

Abstract:

Many accidents were happened because of fast driving, habitual working overtime or tired spirit. This paper presents a solution of remote warning for vehicles collision avoidance using vehicular communication. The development system integrates dedicated short range communication (DSRC) and global position system (GPS) with embedded system into a powerful remote warning system. To transmit the vehicular information and broadcast vehicle position; DSRC communication technology is adopt as the bridge. The proposed system is divided into two parts of the positioning andvehicular units in a vehicle. The positioning unit is used to provide the position and heading information from GPS module, and furthermore the vehicular unit is used to receive the break, throttle, and othersignals via controller area network (CAN) interface connected to each mechanism. The mobile hardware are built with an embedded system using X86 processor in Linux system. A vehicle is communicated with other vehicles via DSRC in non-addressed protocol with wireless access in vehicular environments (WAVE) short message protocol. From the position data and vehicular information, this paper provided a conflict detection algorithm to do time separation and remote warning with error bubble consideration. And the warning information is on-line displayed in the screen. This system is able to enhance driver assistance service and realize critical safety by using vehicular information from the neighbor vehicles.KeywordsDedicated short range communication, GPS, Control area network, Collision avoidance warning system.

Keywords: Dedicated short range communication, GPS, Control area network, Collision avoidance warning system.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2154

Authors: Asnate Kirse, Daina Karklina, Sandra Muizniece-Brasava, Ruta Galoburda

Abstract:

Pulses are high in plant protein and dietary fiber, and contain slowly digestible starches. Innovative products from pulses could increase their consumption and benefit consumer health. This study was conducted to evaluate physicochemical stability of processed cowpea (Vigna unguiculata (L.) Walp. cv. Fradel) and maple pea (Pisum sativum var. arvense L. cv. Bruno) spreads at 5 °C temperature during 62-day storage. Physicochemical stability of pulse spreads was compared after sous vide treatment (80 °C/15 min) and high pressure processing (700 MPa/10 min/20 °C). Pulse spreads were made by homogenizing cooked pulses in a food processor together with salt, citric acid, oil, and bruschetta seasoning. A total of four different pulse spreads were studied: Cowpea spread without and with seasoning, maple pea spread without and with seasoning. Transparent PA/PE and light proof PET/ALU/PA/PP film pouches were used for packaging of pulse spreads under vacuum. The parameters investigated were pH, water activity and mass losses. Pulse spreads were tested on days 0, 15, 29, 42, 50, 57 and 62. The results showed that sous-vide treatment and high pressure processing had an insignificant influence on pH, water activity and mass losses after processing, irrespective of packaging material did not change (p>0.1). pH and water activity of sous-vide treated and high pressure processed pulse spreads in different packaging materials proved to be stable throughout the storage. Mass losses during storage accounted to 0.1% losses. Chosen sous-vide treatment and high pressure processing regimes and packaging materials are suitable to maintain consistent physicochemical quality of the new products during 62-day storage.

Keywords: Cowpea, flexible packaging, maple pea, pH, water activity.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1247

Authors: Dor Oppenheim, Yael Edan, Guy Shani

Abstract:

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

Keywords: Agricultural engineering, computer vision, image processing, flower detection.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2309

Authors: Jaeyoung Lee

Abstract:

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

Keywords: Edge network, embedded network, MMA, matrix multiplication accelerator and semantic segmentation network.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 410

Authors: Moustafa Osman Mohammed

Abstract:

This paper introduces Luhmann’s autopoietic social systems starting with the original concept of autopoiesis by biologists and scientists, including the modification of general systems based on socialized medicine. A specific type of autopoietic system is explained in the three existing groups of the ecological phenomena: interaction, social and medical sciences. This hypothesis model, nevertheless, has a nonlinear interaction with its natural environment ‘interactional cycle’ for the exchange of photon energy with molecular without any changes in topology. The external forces in the systems environment might be concomitant with the natural fluctuations’ influence (e.g. radioactive radiation, electromagnetic waves). The cantilever sensor deploys insights to the future chip processor for prevention of social metabolic systems. Thus, the circuits with resonant electric and optical properties are prototyped on board as an intra–chip inter–chip transmission for producing electromagnetic energy approximately ranges from 1.7 mA at 3.3 V to service the detection in locomotion with the least significant power losses. Nowadays, therapeutic systems are assimilated materials from embryonic stem cells to aggregate multiple functions of the vessels nature de-cellular structure for replenishment. While, the interior actuators deploy base-pair complementarity of nucleotides for the symmetric arrangement in particular bacterial nanonetworks of the sequence cycle creating double-stranded DNA strings. The DNA strands must be sequenced, assembled, and decoded in order to reconstruct the original source reliably. The design of exterior actuators have the ability in sensing different variations in the corresponding patterns regarding beat-to-beat heart rate variability (HRV) for spatial autocorrelation of molecular communication, which consists of human electromagnetic, piezoelectric, electrostatic and electrothermal energy to monitor and transfer the dynamic changes of all the cantilevers simultaneously in real-time workspace with high precision. A prototype-enabled dynamic energy sensor has been investigated in the laboratory for inclusion of nanoscale devices in the architecture with a fuzzy logic control for detection of thermal and electrostatic changes with optoelectronic devices to interpret uncertainty associated with signal interference. Ultimately, the controversial aspect of molecular frictional properties is adjusted to each other and forms its unique spatial structure modules for providing the environment mutual contribution in the investigation of mass temperature changes due to pathogenic archival architecture of clusters.

Keywords: Autopoiesis, quantum photonics, portable energy, photonic structure, photodynamic therapeutic system.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 801

Authors: Liliia N. Butymova, Vladimir Ya Modorskii

Abstract:

To ensure the gas transmittal GCU's efficient operation, leakages through the labyrinth packings (LP) should be minimized. Leakages can be minimized by decreasing the LP gap, which in turn depends on thermal processes and possible rotor vibrations and is designed to ensure absence of mechanical contact. Vibration mitigation allows to minimize the LP gap. It is advantageous to research influence of processes in the dynamic gas-structure system on LP vibrations. This paper considers influence of rotor vibrations on LP gas dynamics and influence of the latter on the rotor structure within the FSI unidirectional dynamical coupled problem. Dependences of nonstationary parameters of gas-dynamic process in LP on rotor vibrations under various gas speeds and pressures, shaft rotation speeds and vibration amplitudes, and working medium features were studied. The programmed multi-processor ANSYS CFX was chosen as a numerical computation tool. The problem was solved using PNRPU high-capacity computer complex. Deformed shaft vibrations are replaced with an unyielding profile that moves in the fixed annulus "up-and-down" according to set harmonic rule. This solves a nonstationary gas-dynamic problem and determines time dependence of total gas-dynamic force value influencing the shaft. Pressure increase from 0.1 to 10 MPa causes growth of gas-dynamic force oscillation amplitude and frequency. The phase shift angle between gas-dynamic force oscillations and those of shaft displacement decreases from 3π/4 to π/2. Damping constant has maximum value under 1 MPa pressure in the gap. Increase of shaft oscillation frequency from 50 to 150 Hz under P=10 MPa causes growth of gas-dynamic force oscillation amplitude. Damping constant has maximum value at 50 Hz equaling 1.012. Increase of shaft vibration amplitude from 20 to 80 µm under P=10 MPa causes the rise of gas-dynamic force amplitude up to 20 times. Damping constant increases from 0.092 to 0.251. Calculations for various working substances (methane, perfect gas, air at 25 ˚С) prove the minimum gas-dynamic force persistent oscillating amplitude under P=0.1 MPa being observed in methane, and maximum in the air. Frequency remains almost unchanged and the phase shift in the air changes from 3π/4 to π/2. Calculations for various working substances (methane, perfect gas, air at 25 ˚С) prove the maximum gas-dynamic force oscillating amplitude under P=10 MPa being observed in methane, and minimum in the air. Air demonstrates surging. Increase of leakage speed from 0 to 20 m/s through LP under P=0.1 MPa causes the gas-dynamic force oscillating amplitude to decrease by 3 orders and oscillation frequency and the phase shift to increase 2 times and stabilize. Increase of leakage speed from 0 to 20 m/s in LP under P=1 MPa causes gas-dynamic force oscillating amplitude to decrease by almost 4 orders. The phase shift angle increases from π/72 to π/2. Oscillations become persistent. Flow rate proved to influence greatly on pressure oscillations amplitude and a phase shift angle. Work medium influence depends on operation conditions. At pressure growth, vibrations are mostly affected in methane (of working substances list considered), and at pressure decrease, in the air at 25 ˚С.

Keywords: Aeroelasticity, labyrinth packings, oscillation phase shift, vibration.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1525