• Journal of the Korea Society of Computer and Information
• /
• v.23 no.10
• /
• pp.23-31
• /
• 2018

Recently, there have been many studies on machine learning. Among them, studies on reinforcement learning are actively worked. In this study, we propose a controller to control bicycle using DDPG (Deep Deterministic Policy Gradient) algorithm which is the latest deep reinforcement learning method. In this paper, we redefine the compensation function of bicycle dynamics and neural network to learn agents. When using the proposed method for data learning and control, it is possible to perform the function of not allowing the bicycle to fall over and reach the further given destination unlike the existing method. For the performance evaluation, we have experimented that the proposed algorithm works in various environments such as fixed speed, random, target point, and not determined. Finally, as a result, it is confirmed that the proposed algorithm shows better performance than the conventional neural network algorithms NAF and PPO.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.25 no.3B
• /
• pp.557-564
• /
• 2000

This paper presents a watermarking algorithm for embedding visually recognizable pattern (Mark, Logo, Symbol, stamping or signature) into the image. first, the color image(RGB model)is transformed in YCbCr model and then the Y component is transformed into 3-level wavelet transform. Next, the values are assembled with pattern watermark. PN(pseudo noise) code at spread spectrum communication method and mutilevel watermark weights. This values are inserted into discrete wavelet domain. In our scheme, new calculating method is designed to calculate wavelet transform with integer value in considering the quantization error. and we used the color conversion with fixed-point arithmetic to be easy to make the hardware hereafter. Also, we made the new solution using mutilevel threshold to robust to common signal distortions and malicious attack, and to enhance quality of image in considering the human visual system. the experimental results showed that the proposed watermarking algorithm was superior to other similar water marking algorithm. We showed what it was robust to common signal processing and geometric transform such as brightness. contrast, filtering. scaling. JPEG lossy compression and geometric deformation.

• International journal of advanced smart convergence
• /
• v.11 no.4
• /
• pp.81-87
• /
• 2022

In this paper, we propose a watershed algorithm that applies a high-frequency enhancement filter to emphasize the boundary and a local adaptive threshold to search for minimum points. The previous method causes the problem of over-segmentation, and over- segmentation appears around the boundary of the object, creating an inaccurate boundary of the region. The proposed method applies a high-frequency enhancement filter that emphasizes the high-frequency region while preserving the low-frequency region, and performs a minimum point search to consider local characteristics. When merging regions, a fixed threshold is applied. As a result of the experiment, the proposed method reduced the number of segmented regions by about 58% while preserving the boundaries of the regions compared to when high frequency emphasis filters were not used.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.9 no.6
• /
• pp.335-344
• /
• 2014

As a kind of 2D spatial coordinate transform, image warping is a basic image processing technique utilized in various applications. Though image warping algorithm is composed of relatively simple operations such as memory accesses and computations of weighted average, real-time implementations on embedded vision systems suffer from limited computational power because the simple operations are iterated as many times as the number of pixels. This paper presents a real-time implementation of a look-up table(LUT)-based image warping using an FPGA. In order to ensure sufficient data transfer rate from memories storing mapping LUT and image data, appropriate memory devices are selected by analyzing memory access patterns in an LUT-based image warping using backward mapping. In addition, hardware structure of a parallel and pipelined architecture is proposed for fast computation of bilinear interpolation using fixed-point operations. Accuracy of the implemented hardware is verified using a synthesized test image, and an application to real-time lens distortion correction is exemplified.

• Proceedings of the KIEE Conference
• /
• 1996.07b
• /
• pp.1295-1297
• /
• 1996

CCITT G.721, G.723 standard ADPCM algorithm is implemented by using TI's fixed point DSP start kit (DSK). ADPCM can be implemented on a various rates, such as 16K, 24K, 32K and 40K. The ADPCM is sample based compression technique and its complexity is not so high as the other speech compression techniques such as CELP, VSELP and GSM, etc. ADPCM is widely applicable to most of the low cost speech compression application and they are tapeless answering machine, simultaneous voice and fax modem, digital phone, etc. TMS320C50 DSP is a low cost fixed point DSP chip and C50 DSK system has an AIC (analog interface chip) which operates as a single chip A/D and D/A converter with 14 bit resolution, C50 DSP chip with on-chip memory of 10K and RS232C interface module. ADPCM C code is compiled by TI C50 C-compiler and implemented on the DSK on-chip memory. Speech signal input is converted into 14 bit linear PCM data and encoded into ADPCM data and the data is sent to PC through RS232C. The ADPCM data on PC is received by the DSK through RS232C and then decoded to generate the 14 bit linear PCM data and converted into the speech signal. The DSK system has audio in/out jack and we can input and out the speech signal.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.43 no.1 s.307
• /
• pp.91-100
• /
• 2006

A scheme is proposed for implementing the block LMS algorithm in a block floating point framework that permits processing of data over a wide dynamic range at a processor complexity and coat as low as that of a fixed point processor. The proposed scheme adopts appropriate formats for representing the filter coefficients and the data. Using these and a new upper bound on the step size, update relations for the filter weight mantissas and exponent are developed, taking care so that neither overflow occurs, nor are quantifies which are already very small multiplied directly. It is further shown how the mantissas of the filter coefficients and also the filter output can be evaluated faster by suitably modifying the approach of the fast block LMS algorithm

• Journal of Power Electronics
• /
• v.11 no.2
• /
• pp.218-227
• /
• 2011

The subject of variable step size maximum power point tracking (MPPT) algorithms has been addressed in the literature. However, most of the addressed algorithms tune the variable step size according to two variables: the photovoltaic (PV) array voltage ($V_{PV}$) and the PV array current ($I_{PV}$). Therefore, both the PV array current and voltage have to be measured. Recently, maximum power point trackers that arc based on a single variable ($I_{PV}$ or $V_{PV}$) have received a great deal of attention due to their simplicity and ease of implementation, when compared to other tracking techniques. In this paper, two methods have been proposed to design a variable step size MPPT algorithm using only a single current sensor for stand-alone battery storage PV systems. These methods utilize only the relationship between the PV array measured current and the converter duty cycle (D) to automatically adapt the step change in the duty cycle to reach the maximum power point (MPP) of the PV array. Detailed analyses and flowcharts of the proposed methods are included. Moreover, a comparison has been made between the proposed methods to investigate their performance in the transient and steady states. Finally, experimental results with field programmable gate arrays (FPGAs) are presented to verify the performance of the proposed methods.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.2
• /
• pp.552-557
• /
• 2000

In the paper, we proposed a face recognition method that uses GA-BP(Genetic Algorithm-Back propagation Network) that optimizes initial parameters such as bias values or weights. Each pixel in the picture is used for input of the neuralnetwork. The initial weights of neural network is consist of fixed-point real values and converted to bit string on purpose of using the individuals that arte expressed in the Genetic Algorithm. For the fitness value, we defined the value that shows the lowest error of neural network, which is evaluated using newly defined adaptive re-learning operator and built the optimized and most advanced neural network. Then we made experiments on the face recognition. In comparison with learning convergence speed, the proposed algorithm shows faster convergence speed than solo executed back propagation algorithm and provides better performance, about 2.9% in proposed method than solo executed back propagation algorithm.

• Journal of Advanced Navigation Technology
• /
• v.22 no.6
• /
• pp.623-629
• /
• 2018

In this paper, we propose a real-time obstacle detection and avoidance path generation algorithm for UAV. 2-D Lidar is used to detect obstacles, and the detected obstacle data is used to generate real-time histogram for local avoidance path and a 2-D SLAM map used for global avoidance path generation to the target point. The VFH algorithm for local avoidance path generation generates a real-time histogram of how much the obstacles are distributed in the vector direction and distance, and this histogram is used to generate the local avoidance path when detecting near fixed or dynamic obstacles. We propose an algorithm, called modified $RRT^*-Smart$, to overcome existing limitations. That generates global avoidance path to the target point by creating lower costs because nodes are checked whether or not straight path to a target point, and given arbitrary lengths and directionality to the target points when nodes are created. In this paper, we prove the efficient avoidance maneuvering through various simulation experiment environment by creating efficient avoidance paths.

• Kyungpook Mathematical Journal
• /
• v.59 no.4
• /
• pp.703-723
• /
• 2019

In this paper, we consider a common solution of three problems in real Hilbert spaces: the Ky Fan inequality problem, the variational inequality problem and the fixed point problem for demicontractive single-valued and quasi-nonexpansive multi-valued mappings. To find the solution we present a new iterative algorithm and prove a strong convergence theorem under mild conditions. Moreover, we provide a numerical example to illustrate the convergence behavior of the proposed iterative method.