• Journal of Biosystems Engineering
• /
• v.29 no.3
• /
• pp.281-286
• /
• 2004

The end-effector is the one of the important factors on development of the cucumber robot to harvester a cucumber. Three end-effectors were designed the single blade end-effector with one blade, the double blade end-effector with two blades and the triple blade end-effector with three blades. Performance tests of the end-effector, the fully integrated system, were conducted to determine the cutting rate by using two different kinds of cucumber. The success rates of cucumber cutting ratio of single end-effector, double end-effector and triple end-effector in laboratory. were 61.7%, 95%, 86.7%, respectively. The cutting rate of single blade or double blade was a little difference with respect to the different diameters of cucumber stem. However, the success cutting rate of the end-effector with triple blade was 61.7% under 29mm diameter of a grabbing stem section. The triple end-effector was not suitable for harvesting a cucumber, but was considered to be suitable for harvesting a grape, an apple and a tomato. The success rate of cucumber cutting ratio of triple end-effectors in greenhouse was 84%. The failure cutting rate was 16% which are due to abnormal shape of cucumber fruit.

• Journal of Korea Multimedia Society
• /
• v.16 no.4
• /
• pp.464-471
• /
• 2013

With the advance of wireless sensor networks, WMSN(Wireless Multimedia Sensor Networks) researches to transfers multimedia data have been studied actively. First of all, three basic protocols, End-to-end, Hop-by-hop, and Unreliable transmission method, are compared to transfer multimedia data using simulations. This paper presents an efficient protocol to transfer multimedia data by overhearing messages of nodes and by transmitting next packets during the unused time interval. The proposed method is verified its performance by simulations and experiments. The results shows that the transmission rate of the proposed method 22% higher than that of End-to-end protocol with half of transmission time. Also the performance of the proposed method is compared with that of PIGAB(Packet Interval Gap based on Adaptive Backoff).

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.21 no.9
• /
• pp.2385-2401
• /
• 1996

In this paper, a management model and a path trace algorithm are proposed for the end-to-end path tracing management in ATM network. Proposedmodel is based on the TINA(Telecommunication Information Networking Architecture) and computational object(CO) of ODP(Open Distributed Processing).We related computational object for the path trace to another computational object which covers the ATM routing and established operational procedure according to this relationship. This procedure identifies the end-to-end by way of tracing the ATM connections then collect identification information on the that path. End-to-end trace is performed on the network management level. Broadcasting-with-synchronized-control and GTM(Global Ticket Method) are proposed as path trace algorithm considering the real time properties and data ingetrity. Computer simulations are also performed to evaluate the performance of the proposed algorithm and its resuls are shown in this paper.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.9
• /
• pp.3110-3125
• /
• 2014

Network coding improves TCP's performance in lossy wireless networks. However, the complex congestion window evolution of network coded TCP (TCP-NC) makes the analysis of end-to-end throughput challenging. This paper analyzes the evolutionary process of TCP-NC against lossy links. An analytic model is established by applying a two-dimensional Markov chain. With maximum window size, end-to-end erasure rate and redundancy parameter as input parameters, the analytic model can reflect window evolution and calculate end-to-end throughput of TCP-NC precisely. The key point of our model is that by the novel definition of the states of Markov chain, both the number of related states and the computation complexity are substantially reduced. Our work helps to understand the factors that affect TCP-NC's performance and lay the foundation of its optimization. Extensive simulations on NS2 show that the analytic model features fairly high accuracy.

• The Journal of the Acoustical Society of Korea
• /
• v.40 no.6
• /
• pp.578-586
• /
• 2021

We propose a semi-supervised learning method based on Variational AutoEncoder (VAE) and Unsupervised Data Augmentation (UDA) to improve the performance of an end-to-end speech recognizer. In the proposed method, first, the VAE-based augmentation model and the baseline end-to-end speech recognizer are trained using the original speech data. Then, the baseline end-to-end speech recognizer is trained again using data augmented from the learned augmentation model. Finally, the learned augmentation model and end-to-end speech recognizer are re-learned using the UDA-based semi-supervised learning method. As a result of the computer simulation, the augmentation model is shown to improve the Word Error Rate (WER) of the baseline end-to-end speech recognizer, and further improve its performance by combining it with the UDA-based learning method.

• The Journal of the Acoustical Society of Korea
• /
• v.40 no.5
• /
• pp.530-536
• /
• 2021

Recently, due to the development of deep learning, end-to-end speech recognition, which directly maps graphemes to speech signals, shows good performance. Especially, among the end-to-end models, conformer shows the best performance. However end-to-end models only focuses on the probability of which grapheme will appear at the time. The decoding process uses a greedy search or beam search. This decoding method is easily affected by the final probability output by the model. In addition, the end-to-end models cannot use external pronunciation and language information due to structual problem. Therefore, in this paper conformer with lexicon transducer is proposed. We compare phoneme-based model with lexicon transducer and grapheme-based model with beam search. Test set is consist of words that do not appear in training data. The grapheme-based conformer with beam search shows 3.8 % of CER. The phoneme-based conformer with lexicon transducer shows 3.4 % of CER.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.11
• /
• pp.1486-1494
• /
• 2021

Character recognition is a technology required in various platforms, such as smart parking and text to speech, and many studies are being conducted to improve its performance through new attempts. However, with low-quality image used for character recognition, a difference in resolution of the training image and test image for character recognition occurs, resulting in poor accuracy. To solve this problem, this paper designed an end-to-end learning neural network that combines image super-resolution and character recognition so that the character recognition model performance is robust against various quality data, and implemented an alternative whole learning algorithm to learn the whole neural network. An alternative end-to-end learning and recognition performance test was conducted using the license plate image among various text images, and the effectiveness of the proposed algorithm was verified with the performance test.

• Proceedings of the KSRS Conference
• /
• v.1
• /
• pp.274-277
• /
• 2006

KARI is currently developing a geostationary ocean color imager (GOCI) for COMS. We report the progress in integrated optical modeling and analysis for stray light suppression and the end-to-end instrument performance verification including in-orbit calibration. The Sun is modeled as the emitting light source and the selected area around Korean peninsular as the observation target that scatters the sun light towards GOCI in orbit. The optical ray tracing employing active geometric scaling was then used for precise characterization of the spatial and radiometric performance at the instrument focal plane. The analysis results show positive reduction in the simulated stray light level with the design improvement including baffles. It also indicates that the ray traced in-orbit radiometric performances are effective tools for the independent assessment of more traditional linear and quadratic equation based estimation of water leaving radiance. The concept of integrated GOCI optical model and the computational method are presented.

• Journal of the Korean Society for Precision Engineering
• /
• v.31 no.5
• /
• pp.395-401
• /
• 2014

Development of five axis tool grinding machine and CAD/CAM systems increase tool design flexibility. In this research, investigated are cutting characteristics of ball-end mill with different helix angle. Special WC ball-end mills with $0^{\circ}$, $10^{\circ}$, $20^{\circ}$, $30^{\circ}$ helix angles are designed and used in various cutting tests. Machining performance according to helix angle variation is evaluated from cutting forces, surface roughness, tool wear, produced chip shape, and vibration characteristics. The ball-end mill with $10^{\circ}$ helix angle shows the best cutting performance due to appropriate chip load distribution and smooth chip flow. This research can be used for cutting edge geometry optimization and novel design of ball-end mill.

• Journal of the Architectural Institute of Korea Structure & Construction
• /
• v.36 no.4
• /
• pp.123-133
• /
• 2020

The purpose of this study is to analyze the method of retrofitting flexural strength and the flexural performance of retrofitted shear walls. There are various ways to reinforce the flexural strength of reinforced concrete shear wall structural systems that have already been built, in the case of that, the external force is increased, and the internal force is insufficient. However, there are various problems, such as excessive flexural stiffness after reinforcement and increasing the thickness and length of the wall. We have developed a retrofit method to solve these problems. The wall end is excavated to place the required vertical rebars, and concrete is poured after placing rebars. This is the same concept as creating wall end boundary elements later on. We also studied the anchorage method of reinforcement and the interaction method between the retrofitting end and the existing wall. The flexural test results for the reinforced concrete shear wall using the studied retrofit method can be predicted according to the sectional analysis and FEM analysis, and there are differences in the plastic hinge length, crack propagation, stiffness degradation and energy dissipation due to the bending depending on the vertical rebar ratio of wall end.