Search results for: measuring accuracy

Authors: Suzan Arslanturk, Mohammad-Reza Siadat, Theophilus Ogunyemi, Ananias Diokno

Abstract:

Missing data is a common challenge in statistical analyses of most clinical survey datasets. A variety of methods have been developed to enable analysis of survey data to deal with missing values. Imputation is the most commonly used among the above methods. However, in order to minimize the bias introduced due to imputation, one must choose the right imputation technique and apply it to the correct type of missing data. In this paper, we have identified different types of missing values: missing data due to skip pattern (SPMD), undetermined missing data (UMD), and genuine missing data (GMD) and applied rough set imputation on only the GMD portion of the missing data. We have used rough set imputation to evaluate the effect of such imputation on prediction by generating several simulation datasets based on an existing epidemiological dataset (MESA). To measure how well each dataset lends itself to the prediction model (logistic regression), we have used p-values from the Wald test. To evaluate the accuracy of the prediction, we have considered the width of 95% confidence interval for the probability of incontinence. Both imputed and non-imputed simulation datasets were fit to the prediction model, and they both turned out to be significant (p-value < 0.05). However, the Wald score shows a better fit for the imputed compared to non-imputed datasets (28.7 vs. 23.4). The average confidence interval width was decreased by 10.4% when the imputed dataset was used, meaning higher precision. The results show that using the rough set method for missing data imputation on GMD data improve the predictive capability of the logistic regression. Further studies are required to generalize this conclusion to other clinical survey datasets.

Keywords: rough set, imputation, clinical survey data simulation, genuine missing data, predictive index

Procedia PDF Downloads 166

Authors: Jacintha Menezes, Ramalingam Dharmalingam, Palaiahnakote Shivakumara

Abstract:

In the recent years, Omani acid lime cultivation and production has been affected by Citrus greening or Huanglongbing (HLB) disease. HLB disease is one of the most destructive diseases for citrus, with no remedies or countermeasures to stop the disease. Currently used Polymerase chain reaction (PCR) and enzyme-linked immunosorbent assay (ELISA) HLB detection tests require lengthy and labor-intensive laboratory procedures. Furthermore, the equipment and staff needed to carry out the laboratory procedures are frequently specialized hence making them a less optimal solution for the detection of the disease. The current research uses hyperspectral imaging technology for automatic detection of citrus trees with HLB disease. Omani citrus tree leaf images were captured through portable Specim IQ hyperspectral camera. The research considered healthy, nutrition deficient, and HLB infected leaf samples based on the Polymerase chain reaction (PCR) test. The highresolution image samples were sliced to into sub cubes. The sub cubes were further processed to obtain RGB images with spatial features. Similarly, RGB spectral slices were obtained through a moving window on the wavelength. The resized spectral-Spatial RGB images were given to Convolution Neural Networks for deep features extraction. The current research was able to classify a given sample to the appropriate class with 92.86% accuracy indicating the effectiveness of the proposed techniques. The significant bands with a difference in three types of leaves are found to be 560nm, 678nm, 726 nm and 750nm.

Keywords: huanglongbing (HLB), hyperspectral imaging (HSI), · omani citrus, CNN

Procedia PDF Downloads 73

Authors: D. Mehdizadeh, M. Rahimian, M. Eskandari-Ghadi

Abstract:

This article is concerned with the determination of the static interaction of a vertically loaded rigid circular disc embedded at the interface of a horizontal layer sandwiched in between two different transversely isotropic half-spaces called as tri-material full-space. The axes of symmetry of different regions are assumed to be normal to the horizontal interfaces and parallel to the movement direction. With the use of a potential function method, and by implementing Hankel integral transforms in the radial direction, the government partial differential equation for the solely scalar potential function is transformed to an ordinary 4th order differential equation, and the mixed boundary conditions are transformed into a pair of integral equations called dual integral equations, which can be reduced to a Fredholm integral equation of the second kind, which is solved analytically. Then, the displacements and stresses are given in the form of improper line integrals, which is due to inverse Hankel integral transforms. It is shown that the present solutions are in exact agreement with the existing solutions for a homogeneous full-space with transversely isotropic material. To confirm the accuracy of the numerical evaluation of the integrals involved, the numerical results are compared with the solutions exists for the homogeneous full-space. Then, some different cases with different degrees of material anisotropy are compared to portray the effect of degree of anisotropy.

Keywords: transversely isotropic, rigid disc, elasticity, dual integral equations, tri-material full-space

Procedia PDF Downloads 434

Authors: Dede Basturk, Metin Kaya, Halil Taskin, Nurtekin Erkmen

Abstract:

In total, 29 students studying in Selçuk University Physical Training and Sports School who are recreationally active participated voluntarilyin this study which was carried out in order to examine effects of Vertimax trainings on agility, quickness and acceleration. 3 groups took their parts in this study as Vertimax training group (N=10), Ordinary training group (N=10) and Control group (N=9). Measurements were carried out in performance laboratory of Selçuk University Physical Training and Sports School. A training program for quickness and agility was followed up for subjects 3 days a week (Monday, Wednesday, Friday) for 8 weeks. Subjects taking their parts in vertimax training group and ordinary training group participated in the training program for quickness and agility. Measurements were applied as pre-test and post-test. Subjects of vertimax training group followed the training program with vertimax device and subjects of ordinary training group followed the training program without vertimax device. As to control group who are recreationally active, they did not participate in any program. 4 gate photocells were used for measuring and measurement of distances was carried out in m. Furthermore, single gate photocell and honi were used for agility test. Measurements started with 15 minutes of warm-up. Acceleration, quickness and agility tests were applied on subjects. 3 measurements were made for each subject at 3 minutes resting intervals. The best rating of three measurements was recorded. 5 m quickness pre-test value of vertimax training groups has been determined as 1,11±0,06 s and post-test value has been determined as 1,06 ± 0,08 s (P<0,05). 5 m quickness pre-test value of ordinary training group has been determined as 1,11±0,06 s and post-test value has been determined as 1,07±0,07 s (P<0,05).5 m quickness pre-test value of control group has been determined as 1,13±0,08 s and post-test value has been determined as 1,10 ± 0,07 s (P>0,05). Upon examination of 10 m acceleration value before and after the training, 10 m acceleration pre-test value of vertimax training group has been determined as 1,82 ± 0,07 s and post-test value has been determined as 1,76±0,83 s (P>0,05). 10 m acceleration pre-test value of ordinary training group has been determined as 1,83±0,05 s and post-test value has been determined as 1,78 ± 0,08 s (P>0,05).10 m acceleration pre-test value of control group has been determined as 1,87±0,11 s and post-test value has been determined as 1,83 ± 0,09 s (P>0,05). Upon examination of 15 m acceleration value before and after the training, 15 m acceleration pre-test value of vertimax training group has been determined as 2,52±0,10 s and post-test value has been determined as 2,46 ± 0,11 s (P>0,05).15 m acceleration pre-test value of ordinary training group has been determined as 2,52±0,05 s and post-test value has been determined as 2,48 ± 0,06 s (P>0,05). 15 m acceleration pre-test value of control group has been determined as 2,55 ± 0,11 s and post-test value has been determined as 2,54 ± 0,08 s (P>0,05).Upon examination of agility performance before and after the training, agility pre-test value of vertimax training group has been determined as 9,50±0,47 s and post-test value has been determined as 9,66 ± 0,47 s (P>0,05). Agility pre-test value of ordinary training group has been determined as 9,99 ± 0,05 s and post-test value has been determined as 9,86 ± 0,40 s (P>0,05). Agility pre-test value of control group has been determined as 9,74 ± 0,45 s and post-test value has been determined as 9,92 ± 0,49 s (P>0,05). Consequently, it has been observed that quickness and acceleration features were developed significantly following 8 weeks of vertimax training program and agility features were not developed significantly. It is suggested that training practices used for the study may be used for situations which may require sudden moves and in order to attain the maximum speed in a short time. Nevertheless, it is also suggested that this training practice does not make contribution in development of moves which may require sudden direction changes. It is suggested that productiveness and innovation may come off in terms of training by using various practices of vertimax trainings.

Keywords: vertimax, training, quickness, agility, acceleration

Procedia PDF Downloads 488

Authors: Arvind Srivastav, Tarun Kanti Bhattacharyya

Abstract:

The north-eastern, Himalayan, and Eastern Ghats Belt of India comprise of earthquake-prone, remote, and hilly terrains. Earthquakes have caused enormous damages in these regions in the past. A wireless sensor network based earthquake early warning system (EEWS) is being developed to mitigate the damages caused by earthquakes. It consists of sensor nodes, distributed over the region, that perform majority voting of the output of the seismic sensors in the vicinity, and relay a message to a base station to alert the residents when an earthquake is detected. At the heart of the EEWS is a low-power two-stage seismic sensor that continuously tracks seismic events from incoming three-axis accelerometer signal at the first-stage, and, in the presence of a seismic event, triggers the second-stage P-wave detector that detects the onset of P-wave in an earthquake event. The parameters of the P-wave detector have been optimized for minimizing detection time and maximizing the accuracy of detection.Working of the sensor scheme has been verified with seven earthquakes data retrieved from IRIS. In all test cases, the scheme detected the onset of P-wave accurately. Also, it has been established that the P-wave onset detection time reduces linearly with the sampling rate. It has been verified with test data; the detection time for data sampled at 10Hz was around 2 seconds which reduced to 0.3 second for the data sampled at 100Hz.

Keywords: earthquake early warning system, EEWS, STA/LTA, polarization, wavelet, event detector, P-wave detector

Procedia PDF Downloads 174

Authors: Song Zhang, Hui Wang, Johannes Henrich Schleifenbaum

Abstract:

Due to the increasing demand for quality assurance and reliability for additive manufacturing, the development of an advanced in-situ monitoring system is required to monitor the process anomalies as input for further process control. Optical-based monitoring systems, such as CMOS cameras and NIR cameras, are proved as effective ways to monitor the geometrical distortion and exceptional thermal distribution. Therefore, many studies and applications are focusing on the availability of the optical-based monitoring system for detecting varied types of defects. However, the capability of the monitoring setup is not quantified. In this study, a quantification model to evaluate the capability of the monitoring setups for the LPBF machine based on acquired monitoring data of a designed test artifact is presented, while the design of the relevant test artifacts is discussed. The monitoring setup is evaluated based on its hardware properties, location of the integration, and light condition. Methodology of data processing to quantify the capacity for each aspect is discussed. The minimal capability of the detectable size of the monitoring set up in the application is estimated by quantifying its resolution and accuracy. The quantification model is validated using a CCD camera-based monitoring system for LPBF machines in the laboratory with different setups. The result shows the model to quantify the monitoring system's performance, which makes the evaluation of monitoring systems with the same concept but different setups possible for the LPBF process and provides the direction to improve the setups.

Keywords: data processing, in-situ monitoring, LPBF process, optical system, quantization model, test artifact

Procedia PDF Downloads 194

Authors: Siti Nur Syazni Mohd Zuki, Tan Ling Ling, Nina Suhaity Azmi, Chong Kwok Feng, Lee Yook Heng

Abstract:

Nitrite (NO2-) ion is used prevalently as a preservative in processed meat. Elevated levels of nitrite also found in edible bird’s nests (EBNs). Consumption of NO2- ion at levels above the health-based risk may cause cancer in humans. Spectrophotometric Griess test is the simplest established standard method for NO2- ion detection, however, it requires careful control of pH of each reaction step and susceptible to strong oxidants and dyeing interferences. Other traditional methods rely on the use of laboratory-scale instruments such as GC-MS, HPLC and ion chromatography, which cannot give real-time response. Therefore, it is of significant need for devices capable of measuring nitrite concentration in-situ, rapidly and without reagents, sample pretreatment or extraction step. Herein, we constructed a microspheres-based microbial optode for visual quantitation of NO2- ion. Raoutella planticola, the bacterium expressing NAD(P)H nitrite reductase (NiR) enzyme has been successfully extracted by microbial technique from EBN collected from local birdhouse. The whole cells and the lipophilic Nile Blue chromoionophore were physically absorbed on the photocurable poly(n-butyl acrylate-N-acryloxysuccinimide) [poly (nBA-NAS)] microspheres, whilst the reduced coenzyme NAD(P)H was covalently immobilized on the succinimide-functionalized acrylic microspheres to produce a reagentless biosensing system. Upon the NiR enzyme catalyzes the oxidation of NAD(P)H to NAD(P)+, NO2- ion is reduced to ammonium hydroxide, and that a colour change from blue to pink of the immobilized Nile Blue chromoionophore is perceived as a result of deprotonation reaction increasing the local pH in the microspheres membrane. The microspheres-based optosensor was optimized with a reflectance spectrophotometer at 639 nm and pH 8. The resulting microbial bio-optode membrane could quantify NO2- ion at 0.1 ppm and had a linear response up to 400 ppm. Due to the large surface area to mass ratio of the acrylic microspheres, it allows efficient solid state diffusional mass transfer of the substrate to the bio-recognition phase, and achieve the steady state response as fast as 5 min. The proposed optical microbial biosensor requires no sample pre-treatment step and possesses high stability as the whole cell biocatalyst provides protection to the enzymes from interfering substances, hence it is suitable for measurements in contaminated samples.

Keywords: acrylic microspheres, microbial bio-optode, nitrite ion, reflectometric

Procedia PDF Downloads 440

Authors: T. Nozu, K. Hibi, T. Nishiie

Abstract:

This paper discusses the applicability of the numerical model for a damage prediction method of the accidental hydrogen explosion occurring in a hydrogen facility. The numerical model was based on an unstructured finite volume method (FVM) code “NuFD/FrontFlowRed”. For simulating unsteady turbulent combustion of leaked hydrogen gas, a combination of Large Eddy Simulation (LES) and a combustion model were used. The combustion model was based on a two scalar flamelet approach, where a G-equation model and a conserved scalar model expressed a propagation of premixed flame surface and a diffusion combustion process, respectively. For validation of this numerical model, we have simulated the previous two types of hydrogen explosion tests. One is open-space explosion test, and the source was a prismatic 5.27 m3 volume with 30% of hydrogen-air mixture. A reinforced concrete wall was set 4 m away from the front surface of the source. The source was ignited at the bottom center by a spark. The other is vented enclosure explosion test, and the chamber was 4.6 m × 4.6 m × 3.0 m with a vent opening on one side. Vent area of 5.4 m2 was used. Test was performed with ignition at the center of the wall opposite the vent. Hydrogen-air mixtures with hydrogen concentrations close to 18% vol. were used in the tests. The results from the numerical simulations are compared with the previous experimental data for the accuracy of the numerical model, and we have verified that the simulated overpressures and flame time-of-arrival data were in good agreement with the results of the previous two explosion tests.

Keywords: deflagration, large eddy simulation, turbulent combustion, vented enclosure

Procedia PDF Downloads 237

Authors: Gashin Shahsavari, Gilles Arnaud-Fassetta, Alberto Campisano, Roberto Bertilotti, Fabien Riou

Abstract:

Sewer deposits have been identified as a major cause of dysfunctions in combined sewer systems regarding sewer management, which induces different negative consequents resulting in poor hydraulic conveyance, environmental damages as well as worker’s health. In order to overcome the problematics of sedimentation, flushing has been considered as the most operative and cost-effective way to minimize the sediments impacts and prevent such challenges. Flushing, by prompting turbulent wave effects, can modify the bed form depending on the hydraulic properties and geometrical characteristics of the conduit. So far, the dynamics of the bed-load during high-flow events in combined sewer systems as a complex environment is not well understood, mostly due to lack of measuring devices capable to work in the “hostile” in combined sewer system correctly. In this regards, a one-episode flushing issue from an opening gate valve with weir function was carried out in a trunk sewer in Paris to understanding its cleansing efficiency on the sediments (thickness: 0-30 cm). During more than 1h of flushing within 5 m distance in downstream of this flushing device, a maximum flowrate and a maximum level of water have been recorded at 5 m in downstream of the gate as 4.1 m3/s and 2.1 m respectively. This paper is aimed to evaluate the efficiency of this type of gate for around 1.1 km (from the point -50 m to +1050 m in downstream from the gate) by (i) determining bed grain-size distribution and sediments evolution through the sewer channel, as well as their organic matter content, and (ii) identifying sections that exhibit more changes in their texture after the flush. For the first one, two series of sampling were taken from the sewer length and then analyzed in laboratory, one before flushing and second after, at same points among the sewer channel. Hence, a non-intrusive sampling instrument has undertaken to extract the sediments smaller than the fine gravels. The comparison between sediments texture after the flush operation and the initial state, revealed the most modified zones by the flush effect, regarding the sewer invert slope and hydraulic parameters in the zone up to 400 m from the gate. At this distance, despite the increase of sediment grain-size rages, D50 (median grain-size) varies between 0.6 mm and 1.1 mm compared to 0.8 mm and 10 mm before and after flushing, respectively. Overall, regarding the sewer channel invert slope, results indicate that grains smaller than sands (< 2 mm) are more transported to downstream along about 400 m from the gate: in average 69% before against 38% after the flush with more dispersion of grain-sizes distributions. Furthermore, high effect of the channel bed irregularities on the bed material evolution has been observed after the flush.

Keywords: bed-load evolution, combined sewer systems, flushing efficiency, sediments transport

Procedia PDF Downloads 399

Authors: Paulo Sérgio Ribeiro de Araújo Bogas

Abstract:

Although several studies have assumed (at least implicitly) that learners' approaches to learning develop into deeper approaches to higher education, there appears to be no clear theoretical basis for this assumption and no empirical evidence. As a scientific contribution to this discussion, a pedagogical intervention of a quasi-experimental nature was developed, with a mixed methodology, evaluating the intervention within a single curricular unit of Marketing, using cases based on real challenges of brands, business simulation, and customer projects. Primary and secondary experiences were incorporated in the intervention: the primary experiences are the experiential activities themselves; the secondary experiences result from the primary experience, such as reflection and discussion in work teams. A diversified learning relationship was encouraged through the various connections between the different members of the learning community. The present study concludes that in the same context, the student's responses can be described as students who reinforce the initial deep approach, students who maintain the initial deep approach level, and others who change from an emphasis on the deep approach to one closer to superficial. This typology did not always confirm studies reported in the literature, namely, whether the initial level of deep processing would influence the superficial and the opposite. The result of this investigation points to the inclusion of pedagogical and didactic activities that integrate different motivations and initial strategies, leading to the possible adoption of deep approaches to learning since it revealed statistically significant differences in the difference in the scores of the deep/superficial approach and the experiential level. In the case of real challenges, the categories of “attribution of meaning and meaning of studied” and the possibility of “contact with an aspirational context” for their future professional stand out. In this category, the dimensions of autonomy that will be required of them were also revealed when comparing the classroom context of real cases and the future professional context and the impact they may have on the world. Regarding the simulated practice, two categories of response stand out: on the one hand, the motivation associated with the possibility of measuring the results of the decisions taken, an awareness of oneself, and, on the other hand, the additional effort that this practice required for some of the students.

Keywords: experiential learning, higher education, mixed methods, reflective learning, marketing

Procedia PDF Downloads 81

Authors: R. O. Osaseri, A. R. Usiobaifo

Abstract:

The power transformer which is responsible for the voltage transformation is of great relevance in the power system and oil-immerse transformer is widely used all over the world. A prompt and proper maintenance of the transformer is of utmost importance. The dissolved gasses content in power transformer, oil is of enormous importance in detecting incipient fault of the transformer. There is a need for accurate prediction of the incipient fault in transformer oil in order to facilitate the prompt maintenance and reducing the cost and error minimization. Study on fault prediction and diagnostic has been the center of many researchers and many previous works have been reported on the use of artificial intelligence to predict incipient failure of transformer faults. In this study machine learning technique was employed by using gradient decent algorithms and Support Vector Machine (SVM) in predicting incipient fault diagnosis of transformer. The method focuses on creating a system that improves its performance on previous result and historical data. The system design approach is basically in two phases; training and testing phase. The gradient decent algorithm is trained with a training dataset while the learned algorithm is applied to a set of new data. This two dataset is used to prove the accuracy of the proposed model. In this study a transformer fault diagnostic model based on Support Vector Machine (SVM) and gradient decent algorithms has been presented with a satisfactory diagnostic capability with high percentage in predicting incipient failure of transformer faults than existing diagnostic methods.

Keywords: diagnostic model, gradient decent, machine learning, support vector machine (SVM), transformer fault

Procedia PDF Downloads 317

Authors: Parker Murray, Marci Johnson, Tyson S. Burnham, Alina K. Fong, Mark D. Allen, Bruce McIff

Abstract:

Purpose: Pathological dysregulation of Neurovascular Coupling (NVC) caused by mild traumatic brain injury (mTBI) is the predominant source of chronic post-concussion syndrome (PCS) symptomology. fNCI has the ability to localize dysregulation in NVC by measuring blood-oxygen-level-dependent (BOLD) signaling during the performance of fMRI-adapted neuropsychological evaluations. With fNCI, 57 brain areas consistently affected by concussion were identified as PCS neural markers, which were validated on large samples of concussion patients and healthy controls. These neuromarkers provide the basis for a computation of PCS severity which is referred to as the Severity Index Score (SIS). The SIS has proven valuable in making pre-treatment decisions, monitoring treatment efficiency, and assessing long-term stability of outcomes. Methods and Materials: After being scanned while performing various cognitive tasks, 476 concussed patients received an SIS score based on the neural dysregulation of the 57 previously identified brain regions. These scans provide an objective measurement of attentional, subcortical, visual processing, language processing, and executive functioning abilities, which were used as biomarkers for post-concussive neural dysregulation. Initial SIS scores were used to develop individualized therapy incorporating cognitive, occupational, and neuromuscular modalities. These scores were also used to establish pre-treatment benchmarks and measure post-treatment improvement. Results: Changes in SIS were calculated in percent change from pre- to post-treatment. Patients showed a mean improvement of 76.5 percent (σ= 23.3), and 75.7 percent of patients showed at least 60 percent improvement. Longitudinal reassessment of 24 of the patients, measured an average of 7.6 months post-treatment, shows that SIS improvement is maintained and improved, with an average of 90.6 percent improvement from their original scan. Conclusions: fNCI provides a reliable measurement of NVC allowing for identification of concussion pathology. Additionally, fNCI derived SIS scores direct tailored therapy to restore NVC, subsequently resolving chronic PCS resulting from mTBI.

Keywords: concussion, functional magnetic resonance imaging (fMRI), neurovascular coupling (NVC), post-concussion syndrome (PCS)

Procedia PDF Downloads 347

Authors: Nubwimana Rachel, Mugabowindekwe Maurice

Abstract:

Trees play a crucial role in maintaining biodiversity and providing various ecological services. Traditional methods of counting trees are time-consuming, and there is a need for more efficient techniques. However, deep learning makes it feasible to identify the multi-scale elements hidden in aerial imagery. This research focuses on the application of deep learning techniques for tree detection and counting in both forest and non-forest areas through the exploration of the deep learning application for automated tree detection and counting using satellite imagery. The objective is to identify the most effective model for automated tree counting. We used different deep learning models such as YOLOV7, SSD, and UNET, along with Generative Adversarial Networks to generate synthetic samples for training and other augmentation techniques, including Random Resized Crop, AutoAugment, and Linear Contrast Enhancement. These models were trained and fine-tuned using satellite imagery to identify and count trees. The performance of the models was assessed through multiple trials; after training and fine-tuning the models, UNET demonstrated the best performance with a validation loss of 0.1211, validation accuracy of 0.9509, and validation precision of 0.9799. This research showcases the success of deep learning in accurate tree counting through remote sensing, particularly with the UNET model. It represents a significant contribution to the field by offering an efficient and precise alternative to conventional tree-counting methods.

Keywords: remote sensing, deep learning, tree counting, image segmentation, object detection, visualization

Procedia PDF Downloads 67

Authors: Udeh Nneka Evelyn

Abstract:

The shortcomings of traditional costing system, in terms of validity, accuracy, consistency and relevance increased the need for modern management accounting system. ABC (Activity-Based Costing) can be used as a modern tool for planning, control and decision making for management. Past studies on activity-based costing (ABC) system have focused on manufacturing firms thereby making the studies on service firms scanty to some extent. This paper reviewed the application and relevance of activity-based costing techniques in service oriented business organisations by employing a qualitative research method which relied heavily on literature review of past and current relevant articles focusing on activity-based costing (ABC). Findings suggest that ABC is not only appropriate for use in a manufacturing environment; it is also most appropriate for service organizations such as financial institutions, the healthcare industry, and government organizations. In fact, some banking and financial institutions have been applying the concept for years under other names. One of them is unit costing, which is used to calculate the cost of banking services by determining the cost and consumption of each unit of output of functions required to deliver the service. ABC in very basic terms may provide very good payback for businesses. Some of the benefits that relate directly to the financial services industry are: Identification of the most profitable customers; more accurate product and service pricing; increase product profitability; well-organized process costs.

Keywords: profitability, activity-based costing (ABC), management accounting, manufacture

Procedia PDF Downloads 575

Authors: Seyedeh Fatemeh Nabavi, Hamid Dalir, Anooshiravan Farshidianfar

Abstract:

Laser welding is pivotal in modern manufacturing, offering unmatched precision, speed, and efficiency. Its versatility in minimizing heat-affected zones, seamlessly joining dissimilar materials, and working with various metals makes it indispensable for crafting intricate automotive components. Integration into automated systems ensures consistent delivery of high-quality welds, thereby enhancing overall production efficiency. Noteworthy are the safety benefits of laser welding, including reduced fumes and consumable materials, which align with industry standards and environmental sustainability goals. As the automotive sector increasingly demands advanced materials and stringent safety and quality standards, laser welding emerges as a cornerstone technology. A comprehensive model encompassing thermal dynamic and characteristics models accurately predicts geometrical, metallurgical, and mechanical aspects of the laser beam welding process. Notably, Model 2 showcases exceptional accuracy, achieving remarkably low error rates in predicting primary and secondary dendrite arm spacing (PDAS and SDAS). These findings underscore the model's reliability and effectiveness, providing invaluable insights and predictive capabilities crucial for optimizing welding processes and ensuring superior productivity, efficiency, and quality in the automotive industry.

Keywords: laser welding process, geometrical characteristics, mechanical characteristics, metallurgical characteristics, comprehensive model, thermal dynamic

Procedia PDF Downloads 46

Authors: Kwei Huan Liw, Sashi B. Darshan

Abstract:

Background: Double eyelid surgery has become one of the most sought-after aesthetic procedures among Asians. Many surgeons perform surgical blepharoplasty and various other methods of non-incisional blepharoplasty. Face-Q is a validated method of measuring patient satisfaction for facial aesthetic procedures. Here we have analyzed the overall eye satisfaction score, the upper eyelid appraisal score and the adverse effect on eyes score Methods: 274 patients (548 eyes), aged between 18 to 40 years old, were recruited from 2015-2018. Each patient underwent a non-incisional double-eyelid blepharoplasty using a single-knotted continuous buried suture. 3 – 5 stab incisions were made depending on the upper eyelid size. A needle loaded with 7-0 nylon is passed from the lateral most wound through the dermis and the conjunctiva in an alternate fashion into the remaining stab wounds. The suture is then tunneled back laterally in the deeper dermis and knotted securely with the suture end. The knot is then buried within the orbicularis oculi muscle. Each patient was required to fill the Face-Q questionnaire before the procedure and 2 weeks post procedure. The results are described based on the percentage of the maximum achievable score. Patients were reviewed after 12 to 18 months to assess the long-term outcome. Results: The overall eye satisfaction score demonstrated a high level of post-operative satisfaction (97.85%), compared to 27.32% pre-operatively. The appraisal of upper eyelid scores showed drastic improvement in perception post-operatively (95.31%) compared to 21.44% pre-operatively. Adverse effect on eyes score showed a very low post-operative complication rate (0.4%) The long-term follow-up showed 6 cases that had developed asymmetrical folds. Only 1 patient agreed for revision surgery. The other 5 patients were still satisfied with the outcome and were not keen for revision surgery. None of the cases had loosening of knots. Conclusion: Modified single-knot continuous buried suture technique is a simple and non-invasive method to create aesthetically pleasing non-surgical double-eyelids, which has long-term effects. Proper patient selection is crucial and good surgical technique is required to achieve a desirable outcome.

Keywords: blepharoplasty, double-eyelid, face-Q, non-incisional

Procedia PDF Downloads 117

Authors: Yun Yu Stephanie Law

Abstract:

Sociometer Theory is well tested among young adults in western population, however, limited research is found for other age groups, like adolescent and middle-adulthood in Asia population. Thus, one of the main purposes of this study is to verify the validity of Sociometer Theory in different age groups among Asian. To be specific, we hypothesized that an increase in one’s perceived social rejection is associated to a decrease in his/her state self-esteem among all age groups in Asian population. And we expected that this association can be found among all age groups including adolescent, young adults and middle-adults group in our first study. In this way, we can verify the validity of Sociometer Theory across different age groups as well as its significance in Asian population. Furthermore, those participants who received rejection about ‘mate-role’ would also receive some negative feedbacks regarding their current/future capacity of being a good mate. Results suggested that participants’ state self-esteem sensitivity for mating-capacity rejection is higher when comparing to that of friend-capacity rejection, i.e. greater drop in state self-esteem when receiving mating-capacity feedbacks then receiving friend-capacity feedbacks. These results, however, is just applicable on young adults. Thus, the main purpose of study two would be testing the state self-esteem sensitivity towards social rejection in different domains among three age groups. We hypothesized that group differences would be found for three age groups regarding state self-esteem sensitivity. Research question 1: perceived social rejection is associated to decrease in state self-esteem, is applicable among different age groups in Asia population. Research question 2: there are significant group differences for three age groups regarding state self-esteem sensitivity. Methods: 300 subjects are divided into three age groups, adolescents group, young adult group and middle-adult group, with 100 subjects in each group. Two questionnaires were used in testing this fundamental concept. Subjects were then asked to rate themselves on questionnaire in measuring their current state self-esteem in order to obtain the baseline measurements for later comparison. In order to avoid demand characteristics from subjects, other unrelated tasks like word matching were also given after the first test. Results: A positive correlation between scores in questionnaire 1 and questionnaire 2 among all age groups. Conclusion: State self-esteem decrease to both imagined social rejection (study1) and experienced social rejection (study2). Moreover, level of decrease in state self-esteem vary when receiving different domains of social rejection. Implications: a better understanding of self-esteem development for various age group might bring insights for education systems and policies for teaching approaches and learning methods among different age groups.

Keywords: state self-esteem, social rejection, stage theory, self-feelings

Procedia PDF Downloads 227

Authors: Hung-Jen Lin, Chien-Hao Wang, Chien-Peng Huang, Yu-Sheng Tseng, En-Cheng Yang, Joe-Air Jiang

Abstract:

Bees’ incoming and outgoing behavior is a decisive index which can indicate the health condition of a colony. Traditional methods for monitoring the behavior of honey bees (Apis mellifera) take too much time and are highly labor-intensive, and the lack of automation and synchronization disables researchers and beekeepers from obtaining real-time information of beehives. To solve these problems, this study proposes to use an Internet of Things (IoT)-based system for counting honey bees’ incoming and outgoing activities using an infrared interruption technique, while environmental factors are recorded simultaneously. The accuracy of the established system is verified by comparing the counting results with the outcomes of manual counting. Moreover, this highly -accurate device is appropriate for providing quantitative information regarding honey bees’ incoming and outgoing behavior. Different statistical analysis methods, including one-way ANOVA and two-way ANOVA, are used to investigate the influence of environmental factors, such as temperature, humidity, illumination and ambient pressure, on bees’ incoming and outgoing behavior. With the real-time data, a standard model is established using the outcomes from analyzing the relationship between environmental factors and bees’ incoming and outgoing behavior. In the future, smart control systems, such as a temperature control system, can also be combined with the proposed system to create an appropriate colony environment. It is expected that the proposed system will make a considerable contribution to the apiculture and researchers.

Keywords: ANOVA, environmental factors, honey bee, incoming and outgoing behavior

Procedia PDF Downloads 362

Authors: Habtamu Tkubet Ebuy

Abstract:

Purpose Conventional dynamic energy simulation tools go beyond the static dimension of simplified methods by providing better and more accurate prediction of building performance. However, their ability to forecast actual performance is undermined by a low representation of human interactions. The purpose of this study is to examine the potential benefits of incorporating information on occupant diversity into occupant behavior models used to simulate building performance. The co-simulation of the stochastic behavior of the occupants substantially increases the accuracy of the simulation. Design/methodology/approach In this article, probabilistic models of the "opening and closing" behavior of the window of inhabitants have been developed in a separate multi-agent platform, SimOcc, and implemented in the building simulation, TRNSYS, in such a way that the behavior of the window with the interconnectivity can be reflected in the simulation analysis of the building. Findings The results of the study prove that the application of complex behaviors is important to research in predicting actual building performance. The results aid in the identification of the gap between reality and existing simulation methods. We hope this study and its results will serve as a guide for researchers interested in investigating occupant behavior in the future. Research limitations/implications Further case studies involving multi-user behavior for complex commercial buildings need to more understand the impact of the occupant behavior on building performance. Originality/value This study is considered as a good opportunity to achieve the national strategy by showing a suitable tool to help stakeholders in the design phase of new or retrofitted buildings to improve the performance of office buildings.

Keywords: occupant behavior, co-simulation, energy consumption, thermal comfort

Procedia PDF Downloads 97

Authors: M A. Habit, S. A. Syed-Sahil, A. Bahari

Abstract:

Abstract—Laser cutting is a method of manufacturing process that uses laser in order to cut materials. It provides and ensures extreme accuracy which has a clean cut effect, CO2 laser dominate this application due to their good- quality beam combined with high output power. It comes with a small scale and it has a limitation in cutting sizes of materials, therefore it is more appropriate for custom- made products. The same laser cutting machine is also capable in cutting fine material such as fine silk, cotton, leather, polyester, etc. Lack of explorations and knowledge besides being unaware about this technology had caused many of the designers not to use this laser cutting method in their collections. The objectives of this study are: 1) To identify the potential of laser cutting technique in Custom-Made Garments for men’s casual wear: 2) To experiment the laser cutting technique in custom made garments: 3) To offer guidelines and formula for men’s custom- made casualwear designs with aesthetic value. In order to achieve the objectives, this research has been conducted by using mixed methods which are interviews with two (2) local experts in the apparel manufacturing industries and interviews via telephone with five (5) local respondents who are local emerging fashion designers, the questionnaires were distributed to one hundred (100) respondents around Klang Valley, in order to gain the information about their understanding and awareness regarding laser cutting technology. The experiment was conducted by using natural and man- made fibers. As a conclusion, all of the objectives had been achieved in producing custom-made men’s casualwear and with the production of these attires it will help to educate and enhance the innovation in fine technology. Therefore, there will be a good linkage and collaboration between the design experts and the manufacturing companies.

Keywords: custom-made, fashion, laser cut, men’s wear

Procedia PDF Downloads 438

Authors: Muhammad Ameer Nawaz Akram

Abstract:

The 21st century has been revealed that many individuals around the world are living in urban settlements than in rural zones. The evolution of numerous cities in emerging and newly developed countries is accompanied by the rise of slums. The precise definition of a slum varies countries to countries, but the universal harmony is that slums are dilapidated settlements facing severe poverty and have lacked access to sanitation, water, electricity, good living styles, and land tenure. The slum settlements always vary in unique patterns within and among the countries and cities. The core objective of this study is the spatial identification and classification of slums in Gujrat city Pakistan from very high-resolution GeoEye-1 (0.41m) satellite imagery. Slums were first identified using GPS for sample site identification and ground-truthing; through this process, 425 slums were identified. Then Object-Oriented Analysis (OOA) was applied to classify slums on digital image. Spatial analysis softwares, e.g., ArcGIS 10.3, Erdas Imagine 9.3, and Envi 5.1, were used for processing data and performing the analysis. Results show that OOA provides up to 90% accuracy for the identification of slums. Jalal Cheema and Allah Ho colonies are severely affected by slum settlements. The ratio of criminal activities is also higher here than in other areas. Slums are increasing with the passage of time in urban areas, and they will be like a hazardous problem in coming future. So now, the executive bodies need to make effective policies and move towards the amelioration process of the city.

Keywords: slums, GPS, satellite imagery, object oriented analysis, zonal change detection

Procedia PDF Downloads 132

Authors: Zeynep Genel

Abstract:

The concept of reputation in corporate development comes to the fore as one of the most frequently discussed topics in recent years. Many organizations, which make worldwide investments, make effort in order to adapt themselves to the topics within the scope of this concept and to promote the name of the organization through the values that might become prominent. The stakeholder groups are considered as the most important actors determining the reputation. Even, the effect of stakeholders is not evaluated as a direct factor; it is signed as indirect effects of their perception are a very strong on ultimate reputation. It is foreseen that the parallelism between the projected reputation and the perceived c reputation, which is established as a result of communication experiences perceived by the stakeholders, has an important effect on achieving these objectives. In assessing the efficiency of these efforts, the opinions of stakeholders are widely utilized. In other words, the projected reputation, in which the positive and/or negative reflections of corporate communication play effective role, is measured through how the stakeholders perceptively position the organization. From this perspective, it is thought that the interaction and cooperation of corporate communication professionals with different stakeholder groups during the reputation positioning efforts play significant role in achieving the targeted reputation or in sustainability of this value. The governmental stakeholders having intense communication with mass stakeholder groups are within the most effective stakeholder groups of organization. The most important reason of this is that the organizations, regarding which the governmental stakeholders have positive perception, inspire more confidence to the mass stakeholders. At this point, the organizations carrying out joint projects with governmental stakeholders in parallel with sustainable communication approach come to the fore as the organizations having strong reputation, whereas the reputation of organizations, which fall behind in this regard or which cannot establish the efficiency from this aspect, is thought to be perceived as weak. Similarly, the social responsibility campaigns, in which the governmental stakeholders are involved and which play efficient role in strengthening the reputation, are thought to draw more attention. From this perspective, the role and effect of governmental stakeholders on the reputation positioning is discussed in this study. In parallel with this objective, it is aimed to reveal perspectives of seven governmental stakeholders towards the cooperation in reputation positioning. The sample group representing the governmental stakeholders is examined under the lights of results obtained from in-depth interviews with the executives of different ministries. It is asserted that this study, which aims to express the importance of stakeholder participation in corporate reputation positioning especially in Turkey and the effective role of governmental stakeholders in strong reputation, might provide a new perspective on measuring the corporate reputation, as well as establishing an important source to contribute to the studies in both academic and practical domains.

Keywords: collaborative communications, reputation management, stakeholder engagement, ultimate reputation

Procedia PDF Downloads 222

Authors: Shih-Wen Hsiao, Rong-Qi Chen

Abstract:

In the field of fashion design, 3D Mannequin is a kind of assisting tool which could rapidly realize the design concepts. While the concept of 3D Mannequin is applied to the computer added fashion design, it will connect with the development and the application of design platform and system. Thus, the situation mentioned above revealed a truth that it is very critical to develop a module of 3D Mannequin which would correspond with the necessity of fashion design. This research proposes a concrete plan that developing and constructing a system of 3D Mannequin with Kinect. In the content, ergonomic measurements of objective human features could be attained real-time through the implement with depth camera of Kinect, and then the mesh morphing can be implemented through transformed the locations of the control-points on the model by inputting those ergonomic data to get an exclusive 3D mannequin model. In the proposed methodology, after the scanned points from the Kinect are revised for accuracy and smoothening, a complete human feature would be reconstructed by the ICP algorithm with the method of image processing. Also, the objective human feature could be recognized to analyze and get real measurements. Furthermore, the data of ergonomic measurements could be applied to shape morphing for the division of 3D Mannequin reconstructed by feature curves. Due to a standardized and customer-oriented 3D Mannequin would be generated by the implement of subdivision, the research could be applied to the fashion design or the presentation and display of 3D virtual clothes. In order to examine the practicality of research structure, a system of 3D Mannequin would be constructed with JAVA program in this study. Through the revision of experiments the practicability-contained research result would come out.

Keywords: 3D mannequin, kinect scanner, interactive closest point, shape morphing, subdivision

Procedia PDF Downloads 303

Authors: Devadrita Dey Sarkar

Abstract:

Regardless of the many technologic advances in the past decade, increased training and experience, and the obvious benefits of uniform standards, the false-negative rate in screening mammography remains unacceptably high .A computer aided neural network classification of regions of suspicion (ROS) on digitized mammograms is presented in this abstract which employs features extracted by a new technique based on independent component analysis. CAD is a concept established by taking into account equally the roles of physicians and computers, whereas automated computer diagnosis is a concept based on computer algorithms only. With CAD, the performance by computers does not have to be comparable to or better than that by physicians, but needs to be complementary to that by physicians. In fact, a large number of CAD systems have been employed for assisting physicians in the early detection of breast cancers on mammograms. A CAD scheme that makes use of lateral breast images has the potential to improve the overall performance in the detection of breast lumps. Because breast lumps can be detected reliably by computer on lateral breast mammographs, radiologists’ accuracy in the detection of breast lumps would be improved by the use of CAD, and thus early diagnosis of breast cancer would become possible. In the future, many CAD schemes could be assembled as packages and implemented as a part of PACS. For example, the package for breast CAD may include the computerized detection of breast nodules, as well as the computerized classification of benign and malignant nodules. In order to assist in the differential diagnosis, it would be possible to search for and retrieve images (or lesions) with these CAD systems, which would be reliable and useful method for quantifying the similarity of a pair of images for visual comparison by radiologists.

Keywords: CAD(computer-aided design), lesions, neural network, ROS(region of suspicion)

Procedia PDF Downloads 453

Authors: Seyed Sadegh Naseralavi, Sadegh Balaghi, Ehsan Khojastehfar

Abstract:

Time history seismic analysis is supposed to be the most accurate method to predict the seismic demand of structures. On the other hand, the required computational time of this method toward achieving the result is its main deficiency. While being applied in optimization process, in which the structure must be analyzed thousands of time, reducing the required computational time of seismic analysis of structures makes the optimization algorithms more practical. Apparently, the invented approximate methods produce some amount of errors in comparison with exact time history analysis but the recently proposed method namely, Complete Quadratic Combination (CQC) and Sum Root of the Sum of Squares (SRSS) drastically reduces the computational time by combination of peak responses in each mode. In the present research, the Basic Modal Displacement (BMD) method is introduced and applied towards estimation of seismic demand of main structure. Seismic demand of sampled structure is estimated by calculation of modal displacement of basic structure (in which the modal displacement has been calculated). Shear steel sampled structures are selected as case studies. The error applying the introduced method is calculated by comparison of the estimated seismic demands with exact time history dynamic analysis. The efficiency of the proposed method is demonstrated by application of three types of earthquakes (in view of time of peak ground acceleration).

Keywords: time history dynamic analysis, basic modal displacement, earthquake-induced demands, shear steel structures

Procedia PDF Downloads 351

Authors: Srinivas Peri, Siva Abhishek Sirivella, Tejaswini Kallakuri, Uzair Ahmad

Abstract:

Smart cities and intelligent transportation systems rely heavily on effective traffic management and infrastructure planning. This research tackles the data scarcity challenge by generating realistically synthetic traffic data from the PeMS-Bay dataset, enhancing predictive modeling accuracy and reliability. Advanced techniques like TimeGAN and GaussianCopula are utilized to create synthetic data that mimics the statistical and structural characteristics of real-world traffic. The future integration of Spatial-Temporal Generative Adversarial Networks (ST-GAN) is anticipated to capture both spatial and temporal correlations, further improving data quality and realism. Each synthetic data generation model's performance is evaluated against real-world data to identify the most effective models for accurately replicating traffic patterns. Long Short-Term Memory (LSTM) networks are employed to model and predict complex temporal dependencies within traffic patterns. This holistic approach aims to identify areas with low vehicle counts, reveal underlying traffic issues, and guide targeted infrastructure interventions. By combining GAN-based synthetic data generation with LSTM-based traffic modeling, this study facilitates data-driven decision-making that improves urban mobility, safety, and the overall efficiency of city planning initiatives.

Keywords: GAN, long short-term memory (LSTM), synthetic data generation, traffic management

Procedia PDF Downloads 8

Authors: Seyed Mohammad Hashemi, Shahrokh Barati

Abstract:

The problem of fault-tolerant control (FTC) by accommodation method has been studied in this paper. The fault occurs in any system components such as actuators, sensors or internal structure of the system and leads to loss of performance and instability of the system. When a fault occurs, the purpose of the fault-tolerant control is designate strategy that can keep the control loop stable and system performance as much as possible perform it without shutting down the system. Here, the section of fault detection and isolation (FDI) system has been evaluated with regard to actuator's fault. Designing a fault detection and isolation system for a multi input-multi output (MIMO) is done by an unknown input observer, so the system is divided to several subsystems as the effect of other inputs such as disturbing given system state equations. In this observer design method, the effect of these disturbances will weaken and the only fault is detected on specific input. The results of this approach simulation can confirm the ability of the fault detection and isolation system design. After fault detection and isolation, it is necessary to redesign controller based on a suitable modification. In this regard after the use of unknown input observer theory and obtain residual signal and evaluate it, PID controller parameters redesigned for iterative. Stability of the closed loop system has proved in the presence of this method. Also, In order to soften the volatility caused by Annie variations of the PID controller parameters, modifying Sigma as a way acceptable solution used. Finally, the simulation results of three tank popular example confirm the accuracy of performance.

Keywords: fault tolerant control, fault detection and isolation, actuator fault, unknown input observer

Procedia PDF Downloads 450

Authors: Lai Wenfang

Abstract:

Tthis study will describe how to use artificial intelligence (AI) technology to build the user-oriented platform for integrated archival service. The platform will be launched in 2020 by the National Archives Administration (NAA) in Taiwan. With the progression of information communication technology (ICT) the NAA has built many systems to provide archival service. In order to cope with new challenges, such as new ICT, artificial intelligence or blockchain etc. the NAA will try to use the natural language processing (NLP) and machine learning (ML) skill to build a training model and propose suggestions based on the data sent to the platform. NAA expects the platform not only can automatically inform the sending agencies’ staffs which records catalogues are against the transfer or destroy rules, but also can use the model to find the details hidden in the catalogues and suggest NAA’s staff whether the records should be or not to be, to shorten the auditing time. The platform keeps all the users’ browse trails; so that the platform can predict what kinds of archives user could be interested and recommend the search terms by visualization, moreover, inform them the new coming archives. In addition, according to the Archives Act, the NAA’s staff must spend a lot of time to mark or remove the personal data, classified data, etc. before archives provided. To upgrade the archives access service process, the platform will use some text recognition pattern to black out automatically, the staff only need to adjust the error and upload the correct one, when the platform has learned the accuracy will be getting higher. In short, the purpose of the platform is to deduct the government digital transformation and implement the vision of a service-oriented smart government.

Keywords: artificial intelligence, natural language processing, machine learning, visualization

Procedia PDF Downloads 168

Authors: Selen Unaldi, Emmanuel Richaud, Matthieu Gervais, Laurent Berthe

Abstract:

Aircrafts are painted with several layers to guarantee their protection from external attacks. For aluminum AA 2024-T3 (metallic structural part of the plane), a protective primer is applied to ensure its corrosion protection. On top of this layer, the top coat is applied for aesthetic aspects. During the lifetime of an aircraft, top coat stripping has an essential role which should be operated as an average of every four years. However, since conventional stripping processes create hazardous disposals and need long hours of labor work, alternative methods have been investigated. Amongst them, laser stripping appears as one of the most promising techniques not only because of the reasons mentioned above but also its controllable and monitorable aspects. The application of a laser beam from the coated side provides stripping, but the depth of the process should be well controlled in order to prevent damage to a substrate and the anticorrosion primer. Apart from that, thermal effects should be taken into account on the painted layers. As an alternative, we worked on developing a process that includes the usage of shock wave propagation to create the stripping via mechanical effects with the application of the beam from the substrate side (back face) of the samples. Laser stripping was applied on thickness-specified samples with a thickness deviation of 10-20%. First, the stripping threshold is determined as a function of power density which is the first flight off of the top coats. After obtaining threshold values, the same power densities were applied to specimens to create large stripping zones with a spot overlap of 10-40%. Layer characteristics were determined on specimens in terms of physicochemical properties and thickness range both before and after laser stripping in order to validate the substrate material health and coating properties. The substrate health is monitored by measuring the roughness of the laser-impacted zones and free surface energy tests (both before and after laser stripping). Also, Hugoniot Elastic Limit (HEL) is determined from VISAR diagnostic on AA 2024-T3 substrates (for the back face surface deformations). In addition, the coating properties are investigated as a function of adhesion levels and anticorrosion properties (neutral salt spray test). The influence of polyurethane top-coat thickness is studied in order to verify the laser stripping process window for industrial aircraft applications.

Keywords: aircraft coatings, laser stripping, laser adhesion tests, epoxy, polyurethane

Procedia PDF Downloads 74

Authors: Wei Zhang

Abstract:

With the rapid development of deep learning, neural network and deep learning algorithms play a significant role in various practical applications. Due to the high accuracy and good performance, Convolutional Neural Networks (CNNs) especially have become a research hot spot in the past few years. However, the size of the networks becomes increasingly large scale due to the demands of the practical applications, which poses a significant challenge to construct a high-performance implementation of deep learning neural networks. Meanwhile, many of these application scenarios also have strict requirements on the performance and low-power consumption of hardware devices. Therefore, it is particularly critical to choose a moderate computing platform for hardware acceleration of CNNs. This article aimed to survey the recent advance in Field Programmable Gate Array (FPGA)-based acceleration of CNNs. Various designs and implementations of the accelerator based on FPGA under different devices and network models are overviewed, and the versions of Graphic Processing Units (GPUs), Application Specific Integrated Circuits (ASICs) and Digital Signal Processors (DSPs) are compared to present our own critical analysis and comments. Finally, we give a discussion on different perspectives of these acceleration and optimization methods on FPGA platforms to further explore the opportunities and challenges for future research. More helpfully, we give a prospect for future development of the FPGA-based accelerator.

Keywords: deep learning, field programmable gate array, FPGA, hardware accelerator, convolutional neural networks, CNN

Procedia PDF Downloads 123