• International journal of advanced smart convergence
• /
• v.10 no.3
• /
• pp.51-58
• /
• 2021

In this paper, we propose a deep learning classifier for estimating the number of layers in the Earth's structure. When installing a grounding system, knowledge of the subsurface in the area is absolutely necessary. The subsurface structure can be modeled by the earth parameters. Knowing the exact number of layers can significantly reduce the amount of computation to estimate these parameters. The classifier consists of a feedforward neural network. Apparent resistivity curves were used to train the deep learning classifier. The apparent resistivity at 20 equally spaced log points in each curve are used as the features for the input of the deep learning classifier. Apparent resistivity curve data sets are collected either by theoretical calculations or by Wenner's measurement method. Deep learning classifiers are coded by Keras, an open source neural network library written in Python. This model has been shown to converge with close to 100% accuracy.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.1
• /
• pp.141-144
• /
• 2020

Convolutional neural networks (CNNs) show high performance in computer vision tasks including object detection, but a lot of weight storage and computation is required. In this paper, a pruning scheme is applied to CNNs for object detection, which can remove much amount of weights with a negligible performance degradation. Contrary to the previous ones, the pruning scheme applied in this paper considers the base accelerator architecture. With the consideration, the pruned CNNs can be efficiently performed on an ASIC or FPGA accelerator. Even with the constrained pruning, the resulting CNN shows a negligible degradation of detection performance, less-than-1% point degradation of mAP on VOD0712 test set. With the proposed scheme, CNNs can be applied to objection dtection efficiently.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.6
• /
• pp.2444-2463
• /
• 2020

Significant detection of video can more rationally allocate computing resources and reduce the amount of computation to improve accuracy. Deep learning can extract the edge features of the image, providing technical support for video saliency. This paper proposes a new detection method. We combine the Convolutional Neural Network (CNN) and the Deep Bidirectional LSTM Network (DB-LSTM) to learn the spatio-temporal features by exploring the object motion information and object motion information to generate video. A continuous frame of significant images. We also analyzed the sample database and found that human attention and significant conversion are time-dependent, so we also considered the significance detection of video cross-frame. Finally, experiments show that our method is superior to other advanced methods.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.7
• /
• pp.2191-2208
• /
• 2022

Motion blur in PET (Positron emission tomography) images induced by respiratory motion will reduce the quality of imaging. Although exiting methods have positive performance for respiratory motion correction in medical practice, there are still many aspects that can be improved. In this paper, an improved 3D unsupervised framework, Res-Voxel based on U-Net network was proposed for the motion correction. The Res-Voxel with multiple residual structure may improve the ability of predicting deformation field, and use a smaller convolution kernel to reduce the parameters of the model and decrease the amount of computation required. The proposed is tested on the simulated PET imaging data and the clinical data. Experimental results demonstrate that the proposed achieved Dice indices 93.81%, 81.75% and 75.10% on the simulated geometric phantom data, voxel phantom data and the clinical data respectively. It is demonstrated that the proposed method can improve the registration and correction performance of PET image.

• Journal of the Korean Society for Precision Engineering
• /
• v.12 no.8
• /
• pp.156-167
• /
• 1995

In this paper, we propose a 3-D range finding method for situation that stereo camera has small translational motion. Binocular stereo generally tends to produce stereo correspondence errors and needs huge amount of computation. The former drawback is because the additional constraints to regularize the correspondence problem are not always true for every scene. The latter drawback is because they use either correlation or optimization to find correct disparity. We present a method which overcomes these drawbacks by moving the stereo camera actively. The method utilized a motion parallax acquired by monocular motion stereo to restrict the search range of binocular disparity. Using only the uniqueness of disparity makes it possible to find reliable binocular disparity. Experimental results with real scene are presented to demonstrate the effectiveness of this method.

• Journal of the Korean Society for Precision Engineering
• /
• v.12 no.6
• /
• pp.128-136
• /
• 1995

The problem addressed in this paper is that of the on-line computational burden of time-optimal control laws for quick, strongly nonlinear systems like revolute robots. It will be demonstrated that a large amount of off-line computation can be substituted for most of the on-line burden in cases of time optimization with constrained inputs if differential point-to- point specifications can be relaxed to cell-to-cell transitions. These cells result from a coarse discretization of likely swaths of state space into a set of nonuniform, contiguous volumes of relatively simple shapes. The cell boundaries approximate stream surfaces of the phase fluid and surfaces of equal transit times. Once the cells have been designed, the bang- bang schedules for the inputs are determined for all likely starting cells and terminating cells. The scheduling process is completed by treating all cells into which the trajectories might unex- pectedly stray as additional starting cells. Then an efficient-to-compute control law can be based on the resulting table of optimal strategies.

• The Journal of Korea Robotics Society
• /
• v.18 no.1
• /
• pp.66-71
• /
• 2023

As cameras have become primary sensors for mobile robots, vision based Simultaneous Localization and Mapping (SLAM) has achieved impressive results with the recent development of computer vision and deep learning. However, vision information has a disadvantage in that a lot of information disappears in a low-light environment. To overcome the problem, we propose an image enhancement method to perform visual SLAM in a low-light environment. Using the deep generative adversarial models and modified gamma correction, the quality of low-light images were improved. The proposed method is less sharp than the existing method, but it can be applied to ORB-SLAM in real time by dramatically reducing the amount of computation. The experimental results were able to prove the validity of the proposed method by applying to public Dataset TUM and VIVID++.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.630-631
• /
• 2021

In this paper, we propose a high-resolution fringe pattern generation technique using deep learning networks. Generating a hologram using a computer requires a very large amount of computation. Therefore, in order to replace this, it was shown that it can be replaced through deep learning, but there was a limitation in the resolution of the output fringe pattern. To improve this, we propose an algorithm for generating a high-resolution fringe pattern.

• Journal of Electrical Engineering and information Science
• /
• v.1 no.2
• /
• pp.67-76
• /
• 1996

In the present paper we propose two new improved iterative restoration algorithms. One is to accelerate convergence of the steepest descent method using the improved search directions, while the other accelerates convergence by using preconditioners. It is also shown that the proposed preconditioned algorithm can accelerate iteration-adaptive iterative image restoration algorithm. The preconditioner in the proposed algorithm can be implemented by using the FIR filter structure, so it can be applied to practical application with manageable amount of computation. Experimental results of the proposed methods show good perfomance improvement in the sense of both convergence speed and quality of the restored image. Although the proposed methods cannot be directly included in spatially-adaptive restoration, they can be used as pre-processing for iteration-adaptive algorithms.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.27 no.3
• /
• pp.346-354
• /
• 2024

Deep learning models require high computing power due to a substantial amount of computation. It is difficult to use them in devices with limited computing environments, such as coastal surveillance equipments. In this study, a lightweight model is constructed by analyzing the weight changes of the convolutional layers during the training process based on MobileNet and then pruning the layers that affects the model less. The performance comparison results show that the lightweight model maintains performance while reducing computational load, parameters, model size, and data processing speed. As a result of this study, an effective pruning method for constructing lightweight deep learning models and the possibility of using equipment resources efficiently through lightweight models in limited computing environments such as coastal surveillance equipments are presented.