• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.3
• /
• pp.113-121
• /
• 2013

In this paper, we propose a sub-pixel rendering algorithm using learning-based 2D FIR filters. The proposed algorithm consists of two stages: the learning and synthesis stages. At the learning stage, we produce the low-resolution synthesis information derived from a sufficient number of high/low resolution block pairs, and store the synthesis information into a so-called dictionary. At the synthesis stage, the best candidate block corresponding to each input high-resolution block is found in the dictionary. Next, we can finally obtain the low-resolution image by synthesizing the low-resolution block using the selected 2D FIR filter on a sub-pixel basis. On the other hand, we additionally enhance the sharpness of the output image by using pre-emphasis considering RGB stripe pattern of display. The simulation results show that the proposed algorithm can provide significantly sharper results than conventional down-sampling methods, without blur effects and aliasing.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.2
• /
• pp.365-380
• /
• 2022

The target detection algorithm based on supervised learning is the current mainstream algorithm for target detection. A high-quality dataset is the prerequisite for the target detection algorithm to obtain good detection performance. The larger the number and quality of the dataset, the stronger the generalization ability of the model, that is, the dataset determines the upper limit of the model learning. The convolutional neural network optimizes the network parameters in a strong supervision method. The error is calculated by comparing the predicted frame with the manually labeled real frame, and then the error is passed into the network for continuous optimization. Strongly supervised learning mainly relies on a large number of images as models for continuous learning, so the number and quality of images directly affect the results of learning. This paper proposes a dataset STAR-24K (meaning a dataset for Space TArget Recognition with more than 24,000 images) for detecting common targets in space. Since there is currently no publicly available dataset for space target detection, we extracted some pictures from a series of channels such as pictures and videos released by the official websites of NASA (National Aeronautics and Space Administration) and ESA (The European Space Agency) and expanded them to 24,451 pictures. We evaluate popular object detection algorithms to build a benchmark. Our STAR-24K dataset is publicly available at https://github.com/Zzz-zcy/STAR-24K.

• Journal of Broadcast Engineering
• /
• v.20 no.5
• /
• pp.676-686
• /
• 2015

A 4D light field image is represented in traditional 2D spatial domain and additional 2D angular domain. The 4D light field has a resolution limitation both in spatial and angular domains since 4D signals are captured by 2D CMOS sensor with limited resolution. In this paper, we propose a dictionary learning-based superresolution algorithm in 4D light field domain to overcome the resolution limitation. The proposed algorithm performs dictionary learning using a large number of extracted 4D light field patches. Then, a high resolution light field image is reconstructed from a low resolution input using the learned dictionary. In this paper, we reconstruct a 4D light field image to have double resolution both in spatial and angular domains. Experimental result shows that the proposed method outperforms the traditional method for the test images captured by a commercial light field camera, i.e. Lytro.

• Journal of Multimedia Information System
• /
• v.2 no.1
• /
• pp.179-186
• /
• 2015

This paper describes the application of a simulated annealing to a random signal-based learning. The simulated annealing is used to generate the reinforcement signal which is used in the random signal-based learning. Random signal-based learning is similar to the reinforcement learning of neural network. It is poor at hill-climbing, whereas simulated annealing has an ability of probabilistic hill-climbing. Therefore, hybridizing a random signal-based learning with the simulated annealing can produce better performance than before. The validity of the proposed algorithm is confirmed by applying it to two different examples. One is finding the minimum of the nonlinear function. And the other is the optimization of fuzzy control rules using inverted pendulum.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.16 no.1
• /
• pp.35-45
• /
• 1991

Multi layer neural network using backpropagation learning algorithm is used to achieve identification and orientation estimation of different classes of aircraft in the variety of 3-D orientations. In-plane distortion invarient$(L,\;{\Phi})$ feature was extracted from each aircraft image to be used for training neural network aircraft classifier. For aircraft identification the optimum structure of the neural network classifier is studied to obtain high classification performance. Effective reductioin of learning time was achieved by using modified backpropagation learning algorithm and varying, learning parameters.

• Journal of Korea Multimedia Society
• /
• v.21 no.2
• /
• pp.250-260
• /
• 2018

In order to control the quadcopter using reinforcement learning, hovering of 1-axial drones prototype is implemented through reinforcement learning. A complementary filter is used to measure the correct angle, and the range of angles is from -180 degrees to +180 degrees using modified complementary filter. The policy gradient method is used together with the REINFORCE algorithm for reinforcement learning. The prototype learned in this way confirmed the difference in performance depending on the length of the episode.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2018.07a
• /
• pp.207-208
• /
• 2018

본 논문에서는 기계학습의 하나인 Learning Vector Quantization 알고리즘을 이용하여 컨트롤러 순서를 정하는 모델을 제안하였다. 제안한 모델은 모든 컨트롤러 정보를 수집하여 Learning Vector Quantization의 LVQ1와 LVQ2 기법을 이용하여 컨트롤러의 순서를 정한다. 이를 통해, 효율적인 컨트롤러 동기화가 이뤄질 것으로 기대된다.

• Journal of IKEEE
• /
• v.2 no.1 s.2
• /
• pp.134-138
• /
• 1998

Because of the statistical fluctuations and the high 'time-variability' nature of the traffic, managing the resources of the network require highly dynamic techniques with minimal Intervention and reaction times, and adaptive and learning capabilities. The neural networks normalizes the ATM cell arrival rate and queue length and has the adaptive learning algorithm, and experimentally investigated the method to prevent the congestion generated in ATM networks.

• Journal of Institute of Control, Robotics and Systems
• /
• v.2 no.2
• /
• pp.108-114
• /
• 1996

We present proofs of the stability and convergence of Self-organizing feature map (SOFM) neural network with time-invarient learning rate and binary reinforcement function. One of the major problems in Self-organizing feature map neural network concerns with learning rate-"Kalman Filter" gain in stochsatic control field which is monotone decreasing function and converges to 0 for satisfying minimum variance property. In this paper, we show that the stability and convergence of Self-organizing feature map neural network with time-invariant learning rate. The analysis of the proposed algorithm shows that the stability and convergence is guranteed with exponentially stable and weak convergence properties as well.s as well.

• Journal of Korea Multimedia Society
• /
• v.23 no.2
• /
• pp.343-350
• /
• 2020

In the era of the Fourth Industrial Revolution in the digital revolution is taking place, various attempts have been made to provide various contents in a digital environment. In this paper, agent-recruitment service system based on collaborative deep learning is proposed for the intelligent head hunting service. The service system is improved from previous research [7] using collaborative deep learning for more reliable recommendation results. The Collaborative deep learning is a hybrid recommendation algorithm using "Recurrent Neural Network(RNN)" specialized for exponential calculation, "collaborative filtering" which is traditional recommendation filtering methods, and "KNN-Clustering" for similar user analysis. The proposed service system can expect more reliable recommendation results than previous research and showed high satisfaction in user survey for verification.