• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.3
• /
• pp.123-128
• /
• 2019

This paper proposes the stock price prediction based on the artificial intelligence, where the model with recurrent convolution neural network (RCNN) layers is adopted. In the motivation of this prediction, long short-term memory model (LSTM)-based neural network can make the output of the time series prediction. On the other hand, the convolution neural network provides the data filtering, averaging, and augmentation. By combining the advantages mentioned above, the proposed technique predicts the estimated stock price of next day. In addition, in order to emphasize the recent time series, a custom weighted loss function is adopted. Moreover, stock data related to the stock price index are adopted to consider the market trends. In the experiments, the proposed stock price prediction reduces the test error by 3.19%, which is over other techniques by about 19%.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.49 no.11
• /
• pp.933-940
• /
• 2021

The space-borne SAR(Synthetic Aperture Radar) system radiates the microwave signal and receives the backscattered signal. The received signal is converted to digital at the Digital Receiver, which is implemented at the end of the SAR sensor receiving chain. The converted signal is formated after signal processing such as filtering and data compression. Two quantization are conducted in the Digital Receiver. One quantization is an analog to digital conversion at ADC(Analog-Digital Converter). Another quantization is the BAQ(Block Adaptive Quantization) for data compression. The quantization process is a conversion from a continuous or higher bit precision to a discrete or lower bit precision. As a result, a quantization noise is inevitably occurred. In this paper, the impact of two quantization processes are analyzed in a view of SNR degradation.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.1
• /
• pp.87-92
• /
• 2021

Bias, variance, error and learning are important factors for performance in modeling a big data based recommendation system. The recommendation model in this system must reduce complexity while maintaining the explanatory diagram. In addition, the sparsity of the dataset and the prediction of the system are more likely to be inversely proportional to each other. Therefore, a product recommendation model has been proposed through learning the similarity between products by using a factorization method of the sparsity of the dataset. In this paper, the generalization ability of the model is improved by applying the max-norm regularization as an optimization method for the loss function of this model. The solution is to apply a stochastic projection gradient descent method that projects a gradient. The sparser data became, it was confirmed that the propsed regularization method was relatively effective compared to the existing method through lots of experiment.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.8
• /
• pp.2068-2082
• /
• 2023

With the rapid development of deep learning, Depth Map Super-Resolution (DMSR) method has achieved more advanced performances. However, when the upsampling rate is very large, it is difficult to capture the structural consistency between color features and depth features by these DMSR methods. Therefore, we propose a color-image guided DMSR method based on iterative depth feature enhancement. Considering the feature difference between high-quality color features and low-quality depth features, we propose to decompose the depth features into High-Frequency (HF) and Low-Frequency (LF) components. Due to structural homogeneity of depth HF components and HF color features, only HF color features are used to enhance the depth HF features without using the LF color features. Before the HF and LF depth feature decomposition, the LF component of the previous depth decomposition and the updated HF component are combined together. After decomposing and reorganizing recursively-updated features, we combine all the depth LF features with the final updated depth HF features to obtain the enhanced-depth features. Next, the enhanced-depth features are input into the multistage depth map fusion reconstruction block, in which the cross enhancement module is introduced into the reconstruction block to fully mine the spatial correlation of depth map by interleaving various features between different convolution groups. Experimental results can show that the two objective assessments of root mean square error and mean absolute deviation of the proposed method are superior to those of many latest DMSR methods.

• The Journal of Korean Institute of Next Generation Computing
• /
• v.13 no.2
• /
• pp.35-48
• /
• 2017

There are many and various devices like sensors, displays, smart phone, etc. in smart environment. And contents can be provided by using these devices. Vast amounts of contents are provided to users, but in most environments, there are no regard for user or some simple elements like location and time are regarded. So there's a limit to provide meaningful contents to users. In this paper, I suggest the contents recommendation system that can recommend contents to users by reasoning context of users, devices and contents. The contents recommendation system suggested in this paper recommend the contents by calculating the user preferences using the situation reasoned with the contextual data acquired from various devices and the user profile received from the user directly. To organize this process, the method on how to model ontology with domain knowledge and how to design and develop the contents recommendation system are discussed in this paper. And an application of the contents recommendation system in Centum City, Busan is introduced. Then, the evaluation methods how the contents recommendation system is evaluated are explained. The evaluation result shows that the mean absolute error is 0.8730, which shows the excellent performance of the proposed contents recommendation system.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.115-117
• /
• 2021

Currently, as IoT technology develops, monitoring systems are being used in various fields, and image processing is being used in various forms. Image data causes noise due to various causes during the transmission and reception process, and if it is not removed, loss of image information or error propagation occurs. Therefore, denoising images is essential. Typical methods of eliminating Salt and Pepper noise in images include AF, MF, and A-TMF. However, existing methods have the disadvantage of being somewhat inadequate in high-density noise. Therefore, in this paper, we propose an algorithm for determining noise for Salt and Pepper denoising and replacing the central pixel with an original pixel if it is non-noise, and processing the filtering mask by segmenting and averaging it in eight directions. We evaluate the performance by comparing and analyzing the proposed algorithms with existing methods.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.28 no.3
• /
• pp.121-128
• /
• 2024

There are growing concerns that the recently implemented Earthquake Early Warning service is overestimating the rapidly provided earthquake magnitudes (M). As a result, the predicted damages unnecessarily activate earthquake protection systems for critical facilities and lifeline infrastructures that are far away. This study is conducted to improve the estimation accuracy of M by incorporating the observed S-wave seismograms in the near source region after removing the site effects of the seismograms in real time by filtering in the time domain. The ensemble of horizontal S-wave spectra from at least five seismograms without site effects is calculated and normalized to a hypocentric target distance (21.54 km) by using the distance attenuation model of Q(f)=348f^0.52 and a cross-over distance of 50 km. The natural logarithmic mean of the S-wave ensemble spectra is then fitted to Brune's source spectrum to obtain the best estimates for M and stress drop (SD) with the fitting weight of 1/standard deviation. The proposed methodology was tested on the 18 recent inland earthquakes in South Korea, and the condition of at least five records for the near-source region is sufficiently fulfilled at an epicentral distance of 30 km. The natural logarithmic standard deviation of the observed S-wave spectra of the ensemble was calculated to be 0.53 using records near the source for 1~10 Hz, compared to 0.42 using whole records. The result shows that the root-mean-square error of M and ln(SD) is approximately 0.17 and 0.6, respectively. This accuracy can provide a confidence interval of 0.4~2.3 of Peak Ground Acceleration values in the distant range.

• Journal of Advanced Navigation Technology
• /
• v.28 no.3
• /
• pp.262-271
• /
• 2024

In this paper, by applying deep learning, one of the A.I. techniques, through angle information, which is optical observation data generated when observing satellites at observatories, distance information from observatories is learned to predict range data, thereby increasing the precision of satellite's orbit determination. To this end, we generated observational data from GMAT, reduced the learning data error of deep learning through preprocessing of the generated observational data, and conducted deep learning through MATLAB. Based on the predicted distance information from learning, trajectory determination was performed using an extended Kalman filter, one of the filtering techniques for trajectory determination, through GMAT. The reliability of the model was verified by comparing and analyzing the orbital determination with angular information without distance information and the orbital determination result with predicted distance information from the model.

• Journal of Broadcast Engineering
• /
• v.1 no.2
• /
• pp.176-187
• /
• 1996

This paper deals with the accurate estimation of 3- D pose (position and orientation) of a moving object with reference to the world frame (or robot base frame), based on a sequence of stereo images taken by cameras mounted on the end - effector of a robot manipulator. This work is an extension of the previous work[1]. Emphasis is given to the 3-D pose estimation relative to the world (or robot base) frame under the presence of not only the measurement noise in 2 - D images[ 1] but also the camera position errors due to the random noise involved in joint angles of a robot manipulator. To this end, a new set of discrete linear Kalman filter equations is derived, based on the following: 1) the orientation error of the object frame due to measurement noise in 2 - D images is modeled with reference to the camera frame by analyzing the noise propagation through 3- D reconstruction; 2) an extended Jacobian matrix is formulated by combining the result of 1) and the orientation error of the end-effector frame due to joint angle errors through robot differential kinematics; and 3) the rotational motion of an object, which is nonlinear in nature, is linearized based on quaternions. Motion parameters are computed from the estimated quaternions based on the iterated least-squares method. Simulation results show the significant reduction of estimation errors and also demonstrate an accurate convergence of the actual motion parameters to the true values.

• The Journal of the Acoustical Society of Korea
• /
• v.18 no.5
• /
• pp.12-20
• /
• 1999

In this paper, we compared speaker verification performance of the speech data collected in clean environment and in channel environment. For the improvement of the performance of speaker verification gathered in channel, we have studied on the efficient feature parameters in channel environment and on the preprocessing. Speech DB for experiment is consisted of Korean doublet of numbers, considering the text-prompted system. Speech features including LPCC(Linear Predictive Cepstral Coefficient), MFCC(Mel Frequency Cepstral Coefficient), PLP(Perceptually Linear Prediction), LSP(Line Spectrum Pair) are analyzed. Also, the preprocessing of filtering to remove channel noise is studied. To remove or compensate for the channel effect from the extracted features, cepstral weighting, CMS(Cepstral Mean Subtraction), RASTA(RelAtive SpecTrAl) are applied. Also by presenting the speech recognition performance on each features and the processing, we compared speech recognition performance and speaker verification performance. For the evaluation of the applied speech features and processing methods, HTK(HMM Tool Kit) 2.0 is used. Giving different threshold according to male or female speaker, we compare EER(Equal Error Rate) on the clean speech data and channel data. Our simulation results show that, removing low band and high band channel noise by applying band pass filter(150～3800Hz) in preprocessing procedure, and extracting MFCC from the filtered speech, the best speaker verification performance was achieved from the view point of EER measurement.