• 대한공업교육학회지
• /
• v.31 no.2
• /
• pp.364-380
• /
• 2006

This paper presents a procedure for designing command to maneuver flexible structure with very little residual vibration, even in the presence of modeling errors. For the open loop maneuver, the various shaped profiles using multiple step inputs delayed in time are considered for robustness and compared with the responses of rigid body and flexible body in virtue of simulations and experiments. Input shaping generates vibration-reducing shaped commands through convolution of an impulse sequence with the desired command. A flexible model with a cylindrical hub and four symmetric appendages is considered to examine the responses to real plant, and to illustrate the effectiveness of the proposed shapers. The appendages are long and flexible, leading to low frequency vibration under any control action. It is shown by a series of simulation that a properly designed feedback controller with input shaper performs well, as compared with open loop controller with input shaper. The control objective is to achieve a fast settling time of residual vibration to flexible structure and robustness (insensitivity)to plant uncertainty, to eliminate residual vibration.

• Journal of IKEEE
• /
• v.24 no.4
• /
• pp.1141-1147
• /
• 2020

The hornet species are so similar in shape that they are difficult for non-experts to classify, and because the size of the objects is small and move fast, it is more difficult to detect and classify the species in real time. In this paper, we developed a system that classifies hornets species in real time based on a deep learning algorithm using a boundary box. In order to minimize the background area included in the bounding box when labeling the training image, we propose a method of selecting only the head and body of the hornet. It also experimentally compares existing boundary box-based object recognition algorithms to find the best algorithms that can detect wasps in real time and classify their species. As a result of the experiment, when the mish function was applied as the activation function of the convolution layer and the hornet images were tested using the YOLOv4 model with the Spatial Attention Module (SAM) applied before the object detection block, the average precision was 97.89% and the average recall was 98.69%.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.2
• /
• pp.456-477
• /
• 2024

With information technology's rapid development, the Internet faces serious security problems. Studies have shown that malware has become a primary means of attacking the Internet. Therefore, adversarial samples have become a vital breakthrough point for studying malware. By studying adversarial samples, we can gain insights into the behavior and characteristics of malware, evaluate the performance of existing detectors in the face of deceptive samples, and help to discover vulnerabilities and improve detection methods for better performance. However, existing adversarial sample generation methods still need help regarding escape effectiveness and mobility. For instance, researchers have attempted to incorporate perturbation methods like Fast Gradient Sign Method (FGSM), Projected Gradient Descent (PGD), and others into adversarial samples to obfuscate detectors. However, these methods are only effective in specific environments and yield limited evasion effectiveness. To solve the above problems, this paper proposes a malware adversarial sample generation method (PixGAN) based on the pixel attention mechanism, which aims to improve adversarial samples' escape effect and mobility. The method transforms malware into grey-scale images and introduces the pixel attention mechanism in the Deep Convolution Generative Adversarial Networks (DCGAN) model to weigh the critical pixels in the grey-scale map, which improves the modeling ability of the generator and discriminator, thus enhancing the escape effect and mobility of the adversarial samples. The escape rate (ASR) is used as an evaluation index of the quality of the adversarial samples. The experimental results show that the adversarial samples generated by PixGAN achieve escape rates of 97%, 94%, 35%, 39%, and 43% on the Random Forest (RF), Support Vector Machine (SVM), Convolutional Neural Network (CNN), Convolutional Neural Network and Recurrent Neural Network (CNN_RNN), and Convolutional Neural Network and Long Short Term Memory (CNN_LSTM) algorithmic detectors, respectively.

• Journal of KIISE:Computing Practices and Letters
• /
• v.7 no.5
• /
• pp.510-521
• /
• 2001

Image retrieval system should be construct for searching fast, efficient image be extract the accurate feature information of image with more massive and more complex characteristics. Image retrieval system are essential differences between image databases and traditional databases. These differences lead to interesting new issues in searching of image, data modeling. So, cause us to consider new generation method of database, efficient retrieval method of image. In this paper, To extract feature information of edge using in searching from input image, we was performed to extract the edge by convolution Laplacian mask and input image, and we implemented the annotation-based and content-based image retrieval system for searching fast, efficient image by generation image database from extracting feature information of edge and metadata. We can improve the performance of the image contents retrieval, because the annotation-based and content-based image retrieval system is using image index which is made up of the content-based edge feature extract information represented in the low level of image and annotation-based edge feature information represented in the high level of image. As a conclusion, image retrieval system proposed in this paper is possible the accurate management of the accumulated information for the image contents and the information sharing and reuse of image because the proposed method do construct the image database by metadata.

• Korean Journal of Optics and Photonics
• /
• v.31 no.2
• /
• pp.59-70
• /
• 2020

We discuss a modified numerical model based on the Monte Carlo radiative transfer (MCRT) method, i.e., the MCRT matrix method, for the analysis of atmospheric scattering effects in three-dimensional flash LIDAR systems. Based on the MCRT method, the radiative transfer function for a LIDAR signal is constructed in a form of a matrix, which corresponds to the characteristic response. Exploiting the superposition and convolution of the characteristic response matrices under the paraxial approximation, an extended computer simulation model of an overall flash LIDAR system is developed. The MCRT matrix method substantially reduces the number of tracking signals, which may grow excessively in the case of conventional Monte Carlo methods. Consequently, it can readily yield fast acquisition of the signal response under various scattering conditions and LIDAR-system configurations. Using the computational model based on the MCRT matrix method, we carry out numerical simulations of a three-dimensional flash LIDAR system operating under different atmospheric conditions, varying the scattering coefficient in terms of visible distance. We numerically analyze various phenomena caused by scattering effects in this system, such as degradation of the signal-to-noise ratio, glitches, and spatiotemporal spread and time delay of the LIDAR signals. The MCRT matrix method is expected to be very effective in analyzing a variety of LIDAR systems, including flash LIDAR systems for autonomous driving.

• Geophysics and Geophysical Exploration
• /
• v.25 no.4
• /
• pp.177-188
• /
• 2022

Recently, the application of distributed acoustic sensors (DAS), which can replace geophones and seismometers, has significantly increased along with interest in micro-seismic monitoring technique, which is one of the CO₂ storage monitoring techniques. A significant amount of temporally and spatially continuous data is recorded in a DAS monitoring system, thereby necessitating fast and accurate data processing techniques. Because event detection and seismic phase picking are the most basic data processing techniques, they should be performed on all data. In this study, a machine learning-based P, S wave phase picking algorithm was developed to compensate for the limitations of conventional phase picking algorithms, and it was modified using a transfer learning technique for the application of DAS data consisting of a single component with a low signal-to-noise ratio. Our model was constructed by modifying the convolution-based EQTransformer, which performs well in phase picking, to the ResUNet structure. Not only the global earthquake dataset, STEAD but also the augmented dataset was used as training datasets to enhance the prediction performance on the unseen characteristics of the target dataset. The performance of the developed algorithm was verified using K-net and KiK-net data with characteristics different from the training data. Additionally, after modifying the trained model to suit DAS data using the transfer learning technique, the performance was verified by applying it to the DAS field data measured in the Pohang Janggi basin.