• IEMEK Journal of Embedded Systems and Applications
• /
• v.19 no.2
• /
• pp.73-81
• /
• 2024

Recently, mobile manipulators have been utilized in agriculture industry for weed removal and harvest automation. This paper proposes a domain adaptive fruit detection method for harvest automation, by utilizing OWL-ViT model which is an open-vocabulary object detection model. The vision-language model can detect objects based on text prompt, and therefore, it can be extended to detect objects of undefined categories. In the development of deep learning models for real-world problems, constructing a large-scale labeled dataset is a time-consuming task and heavily relies on human effort. To reduce the labor-intensive workload, we utilized a large-scale public dataset as a source domain data and employed a domain adaptation method. Adversarial learning was conducted between a domain discriminator and feature extractor to reduce the gap between the distribution of feature vectors from the source domain and our target domain data. We collected a target domain dataset in a real-like environment and conducted experiments to demonstrate the effectiveness of the proposed method. In experiments, the domain adaptation method improved the AP50 metric from 38.88% to 78.59% for detecting objects within the range of 2m, and we achieved 81.7% of manipulation success rate.

• The Transactions of the Korean Institute of Electrical Engineers P
• /
• v.66 no.4
• /
• pp.247-251
• /
• 2017

This paper proposes a low power transmission technique for single-carrier modulation with frequency domain equalization. As time domain signals and frequency domain signals have unique corresponding functions, inserting zeros after each symbol causes a repetition in other domain, so maximal ratio combining technique using repetitive transmission can be applied in the frequency domain. In this paper, we configure transmit signals to insert zeros after each symbols for single-carrier modulation with frequency domain equalization and maximal ratio receive combining block in the receiver structures, propose a structure for transmitter and receiver and show that its performance is better than the traditional algorithm by simulations.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.30 no.6 s.249
• /
• pp.691-697
• /
• 2006

Metamodel, model of model, has been widely used to improve an efficiency of optimization process in engineering fields. However, global metamodels of constraints in a constrained optimization problem are required good accuracy around neighborhood of optimum point. To satisfy this requirement, more sampling points must be located around the boundary and inside of feasible region. Therefore, a new sampling strategy that is capable of identifying feasible domain should be applied to select sampling points for metamodels of constraints. In this research, we suggeste sequential feasible domain sampling that can locate sampling points likely within feasible domain by using penalty function method. To validate the excellence of feasible domain sampling, we compare the optimum results from the proposed method with those form conventional global space-filling sampling for a variety of optimization problems. The advantages of the feasible domain sampling are discussed further.

• The Journal of the Acoustical Society of Korea
• /
• v.28 no.1E
• /
• pp.29-34
• /
• 2009

MPEG4 High Efficiency Advanced Audio Coding (HE-AAC) and MPEG Surround are one of the most efficient combinations for low bit rate multi-channel audio coding. Based on the fact that these two codecs have identical quadrature mirror filter (QMF) analysis and synthesis structures, we propose a dual-domain connection scheme for the codecs. Specifically two time-domain connection methods are analyzed and compared to the QMF subband-domain connection method. Experimental results show that both the time-domain connection methods cause no subjective sound quality degradation compared to the QMF subband-domain connection method, which verifies that one can select either of them depending on application scenarios.

• Tunnel and Underground Space
• /
• v.15 no.5 s.58
• /
• pp.352-358
• /
• 2005

For non-destructive testing of concrete structures, time and frequency domain method were applied to detect cavity in underground model and pier model. To interpret the measured data, time domain method made use of tomography which was completed with first arrivaltime and inversion method. In this steady, frequency domain method using Fourier transform was tried. Maximum frequency in the frequency domain was analyzed to calculate location of cavity.

• International Journal of Naval Architecture and Ocean Engineering
• /
• v.3 no.1
• /
• pp.86-94
• /
• 2011

This paper compares frequency domain and time domain predictions from the ShipMo3D ship motion library with observed motions from model tests and sea trials. ShipMo3D evaluates hull radiation and diffraction forces using the frequency domain Green function for zero forward speed, which is a suitable approach for ships travelling at moderate speed (e.g., Froude numbers up to 0.4). Numerical predictions give generally good agreement with experiments. Frequency domain and linear time domain predictions are almost identical. Evaluation of nonlinear buoyancy and incident wave forces using the instantaneous wetted hull surface gives no improvement in numerical predictions. Consistent prediction of roll motions remains a challenge for seakeeping codes due to the associated viscous effects.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2002.11a
• /
• pp.209-212
• /
• 2002

The realization for code domain analyzing system on WCDMA is presented. This system is realized with the RF, FPGA and DSP processor. The FPGA firgures out the scambling code, and DSP produces the power of code domain using OVSF. The experiment results allow the code domain power accuracy to keep $\pm$1dB low. This system leads to split the text, sound and image signal.

• Journal of the Korea Society of Computer and Information
• /
• v.23 no.1
• /
• pp.17-24
• /
• 2018

With improvement of speech recognition and natural language processing, dialog systems are recently adapted to various service domains. It became possible to get desirable services by conversation through the dialog system, but it is still necessary to improve separate modules, such as domain detection, intention detection, named entity recognition, and out-of-domain detection, in order to achieve stable service offer. When it misclassifies an in-domain sentence of conversation as out-of-domain, it will result in poor customer satisfaction and finally lost business. As there have been relatively small number of studies related to the out-of-domain detection, in this paper, we introduce a new method using a hierarchical Dirichlet process and demonstrate the effectiveness of it by experimental results on Korean dataset.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.7
• /
• pp.2512-2529
• /
• 2015

Wireless mesh networks (WMNs) provide an efficient and flexible method to the field of wireless networking, but also bring many security issues. A mesh point may lose all of its available links during its movement. Thus, the mesh point needs to handover to a new mesh point in order to obtain access to the network again. For multi-domain WMNs, we proposed a new ID-based signcryption scheme and accordingly present a novel ID-based handover protocol for mesh points. The mutual authentication and key establishment of two mesh points which belong to different trust domains can be achieved by using a single one-round message exchange during the authentication phase. The authentication server is not involved in our handover authentication protocol so that mutual authentication can be completed directly by the mesh points. Meanwhile, the data transmitted between the two mesh points can be carried by the authentication messages. Moreover, there are no restrictions on the PKG system parameters in our proposed multi-domain ID-based signcryption scheme so our handover scheme can be easily applied to real WMNs circumstances. Security of the signcryption scheme is proved in the random oracle model. It shows that our protocol satisfies the basic security requirements and is resistant to existing attacks based on the security of the signcryption. The analysis of the performance demonstrates that the protocol is efficient and suitable for the multi-domain WMNs environment.

• Journal of Ocean Engineering and Technology
• /
• v.34 no.6
• /
• pp.394-405
• /
• 2020

To assess the station-keeping performance of floating structures in the Arctic region, the ice load should be considered along with other environmental loads induced by waves, wind, and currents. However, present methods for performance evaluation in the time domain are not effective in terms of time and cost. An ice load generation module is proposed based on the experimental data measured at the KRISO ice model basin. The developed module was applied to a time domain simulation. Using the results of a captive model test conducted in multiple directions, the statistical characteristics of ice loads were analyzed and processed so that an ice load corresponding to an arbitrary angle of the structure could be generated. The developed module is connected to commercial dynamic analysis software (OrcaFlex) as an external force input. Station-keeping simulation in the time domain was conducted for the same floating structure used in the model test. The mooring system was modeled and included to reflect the designed operation scenario. Simulation results show the effectiveness of the proposed ice generation module and its application to station-keeping performance evaluation. Considering the generated ice load, the designed structure can maintain a heading angle relative to ice up to 4°. Station-keeping performance is enhanced as the heading angle conforms to the drift direction. It is expected that the developed module will be used as a platform to verify station-keeping algorithms for Arctic floating structures with a dynamic positioning system.