• Journal of Multimedia Information System
• /
• v.7 no.1
• /
• pp.33-40
• /
• 2020

Machine learning consists of supervised and unsupervised learning among which supervised learning is used for the speech recognition objectives. Supervised learning is the Data mining task of inferring a function from labeled training data. Speech recognition is the current trend that has gained focus over the decades. Most automation technologies use speech and speech recognition for various perspectives. This paper demonstrates an overview of major technological standpoint and gratitude of the elementary development of speech recognition and provides impression method has been developed in every stage of speech recognition using supervised learning. The project will use DNN to recognize speeches using magnitudes with large datasets.

• IEIE Transactions on Smart Processing and Computing
• /
• v.2 no.4
• /
• pp.208-215
• /
• 2013

Early iterations of the existing Global Positioning System (GPS)-based or radio lateration technique-based vehicle localization algorithms suffer from flip ambiguities, forged relative location information and location information exchange overhead, which affect the subsequent iterations. This, in turn, results in an erroneous neighbor-vehicle map. This paper proposes an extended information overlap measure (EIOM) algorithm to reduce the flip error rates by exchanging the neighbor-vehicle presence features in binary information. This algorithm shifts and associates three pieces of information in the Moore neighborhood format: 1) feature information of the neighboring vehicles from a vision-based environment sensor system; 2) cardinal locations of the neighboring vehicles in its Moore neighborhood; and 3) identification information (MAC/IP addresses). Simulations were conducted for multi-lane highway scenarios to compare the proposed algorithm with the existing algorithm. The results showed that the flip error rates were reduced by up to 50%.

• Preventive Nutrition and Food Science
• /
• v.8 no.2
• /
• pp.200-218
• /
• 2003

Phenolic compounds in food and plant materials belong to the simple phenols, phenolic acids, coumarins, flavonoids, stilbenes, tannins, lignans and lignins, all of which are considered as secondary plant metabolites. These compounds may be synthesized by plants during normal development or in response to stress conditions. Phenolics are not distributed uniformly in plants. Insoluble phenolics are components of cell walls while soluble ones are present in vacuoles. A cursory account of phenolics of cereals, beans, pulses, fruits, vegetables and oilseeds is provided in this overview. The information on the bioavailability and absorption of plant phenolics remains fragmentary and diverse. Pharmacological potentials of food phenolics ave extensively evaluated. However, there are many challenges that must be overcome in order to fully understand both the function of phenolics in plant as well as their health effects.

• Earthquakes and Structures
• /
• v.14 no.6
• /
• pp.551-565
• /
• 2018

Experimental testing is considered the most realistic approach to obtain a detailed representation of the nonlinear behaviour of masonry-infilled reinforced concrete (RC) structures. Among other applications, these tests can be used to calibrate the properties of numerical models such as simplified macro-models (e.g., strut-type models) representing the masonry infill behaviour. Since the significant cost of experimental tests limits their widespread use, alternative approaches need to be established to obtain adequate data to validate the referred simplified models. The proposed paper introduces a detailed finite element modelling strategy that can be used as an alternative to experimental tests to represent the behaviour of masonry-infilled RC frames under earthquake loading. Several examples of RC infilled frames with different infill configurations and properties subjected to cyclic loading are analysed using the proposed modelling approach. The comparison between numerical and experimental results shows that the numerical models capture the overall nonlinear behaviour of the physical specimens with adequate accuracy, predicting their monotonic stiffness, strength and several failure mechanisms.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.48 no.3
• /
• pp.313-320
• /
• 1999

The Lane-Curvature Method(LCM) presented in this paper is a new local obstacle avoidance method for indoor mobile robots. The method combines Curvature-Velocith Method(CVM) with a new directional method called the Lane Method. The Lane Method divides the environment into lanes taking the information on obstacles and desired heading of the robot into account ; then it chooses the best lane to follow to optimize travel along a desired heading. A local heading is then calculated for entering and following the best lane, and CVM uses this heading to determine the optimal translational and rotational velocity space methods, LCM yields safe collision-free motion as well as smooth motion taking the dynamics of the robot Xavier, show the efficiency of the proposed method.

• The Pure and Applied Mathematics
• /
• v.15 no.2
• /
• pp.121-134
• /
• 2008

A pebbling step on a graph consists of removing two pebbles from one vertex and placing one pebble on an adjacent vertex. The covering cover pebbling number of a graph is the smallest number of pebbles, such that, however the pebbles are initially placed on the vertices of the graph, after a sequence of pebbling moves, the set of vertices with pebbles forms a covering of G. In this paper we find the covering cover pebbling number of n-cube and diameter two graphs. Finally we give an upperbound for the covering cover pebbling number of graphs of diameter d.

• The Journal of Asian Finance, Economics and Business
• /
• v.8 no.11
• /
• pp.31-39
• /
• 2021

This study empirically examines the proposition that the domestic fundamentals of a nation can emerge as absorptive capacity factors to reap the benefits of inward FDI. The study is contextualized in Asia, set from1982 to 2017, and data is grouped into low-income and lower-middle-income economies, in comparison to high-income and upper-middle-income economies, catering to different geographical regions within Asia. The investigation is based on a series of absorptive capacity factors such as infrastructure, human capital, domestic credit, and health indicator. The methodological analysis is premised on dynamic panel structure and employs the Generalized Method of Moments (GMM) estimation technique. The empirical findings suggest that that the infrastructure variable appears to be the major absorptive capacity factor for both groups of countries. The health indicator, on the other hand, can help reap the benefits of inward FDI, but only if the threshold level is met. The selected economies must achieve this threshold level to reap the benefits of FDI. To absorb the benefits of inward FDI, countries must be proactive in providing sound infrastructure and implementing proper healthcare measures.

• International Journal of Computer Science & Network Security
• /
• v.21 no.6
• /
• pp.294-303
• /
• 2021

VANET is an upcoming technology with an encouraging prospect as well as great challenges, specifically in its security. This paper intends to survey such probable attacks and the correlating detection mechanisms that are introduced in the literature. Accordingly, administering security and protecting the owner's privacy has become a primary argument in VANETs. To furnish stronger security and preserve privacy, one should recognize the various probable attacks on the network and the essence of their behavior. This paper presents a comprehensive survey on diversified attacks and the recommended unfolding by the various researchers which concentrate on security services and the corresponding countermeasures to make VANET communications more secure.

• The Journal of Korea Robotics Society
• /
• v.17 no.2
• /
• pp.133-141
• /
• 2022

This paper introduces model compression algorithms which make a deep neural network smaller and faster for embedded systems. The model compression algorithms can be largely categorized into pruning, quantization and knowledge distillation. In this study, gradual pruning, quantization aware training, and knowledge distillation which learns the activation boundary in the hidden layer of the teacher neural network are integrated. As a large deep neural network is compressed and accelerated by these algorithms, embedded computing boards can run the deep neural network much faster with less memory usage while preserving the reasonable accuracy. To evaluate the performance of the compressed neural networks, we evaluate the size, latency and accuracy of the deep neural network, DenseNet201, for image classification with CIFAR-10 dataset on the NVIDIA Jetson Xavier.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.18 no.3
• /
• pp.101-108
• /
• 2023

Optimizing the performance of deep neural networks on embedded systems is a challenging task that requires efficient compilers and runtime systems. We propose a TVM-based approach that consists of three steps: quantization, auto-scheduling, and ahead-of-time compilation. Our approach reduces the computational complexity of models without significant loss of accuracy, and generates optimized code for various hardware platforms. We evaluate our approach on three representative CNNs using ImageNet Dataset on the NVIDIA Jetson AGX Xavier board and show that it outperforms baseline methods in terms of processing speed.