• Journal of The Korean Society of Agricultural Engineers
• /
• v.63 no.1
• /
• pp.11-25
• /
• 2021

The objective of this study was to analyze the impact of activation functions on flood forecasting model based on Artificial neural networks (ANNs). The traditional activation functions, the sigmoid and tanh functions, were compared with the functions which have been recently recommended for deep neural networks; the ReLU, leaky ReLU, and ELU functions. The flood forecasting model based on ANNs was designed to predict real-time runoff for 1 to 6-h lead time using the rainfall and runoff data of the past nine hours. The statistical measures such as R2, Nash-Sutcliffe Efficiency (NSE), Root Mean Squared Error (RMSE), the error of peak time (ETp), and the error of peak discharge (EQp) were used to evaluate the model accuracy. The tanh and ELU functions were most accurate with R2=0.97 and RMSE=30.1 (㎥/s) for 1-h lead time and R2=0.56 and RMSE=124.6~124.8 (㎥/s) for 6-h lead time. We also evaluated the learning speed by using the number of epochs that minimizes errors. The sigmoid function had the slowest learning speed due to the 'vanishing gradient problem' and the limited direction of weight update. The learning speed of the ELU function was 1.2 times faster than the tanh function. As a result, the ELU function most effectively improved the accuracy and speed of the ANNs model, so it was determined to be the best activation function for ANNs-based flood forecasting.

• Journal of Institute of Control, Robotics and Systems
• /
• v.16 no.12
• /
• pp.1150-1158
• /
• 2010

In this study, the Polynomial-based Radial Basis Function Neural Networks is proposed as one of the recognition part of overall face recognition system that consists of two parts such as the preprocessing part and recognition part. The design methodology and procedure of the proposed pRBFNNs are presented to obtain the solution to high-dimensional pattern recognition problem. First, in preprocessing part, we use a CCD camera to obtain a picture frame in real-time. By using histogram equalization method, we can partially enhance the distorted image influenced by natural as well as artificial illumination. We use an AdaBoost algorithm proposed by Viola and Jones, which is exploited for the detection of facial image area between face and non-facial image area. As the feature extraction algorithm, PCA method is used. In this study, the PCA method, which is a feature extraction algorithm, is used to carry out the dimension reduction of facial image area formed by high-dimensional information. Secondly, we use pRBFNNs to identify the ID by recognizing unique pattern of each person. The proposed pRBFNNs architecture consists of three functional modules such as the condition part, the conclusion part, and the inference part as fuzzy rules formed in 'If-then' format. In the condition part of fuzzy rules, input space is partitioned with Fuzzy C-Means clustering. In the conclusion part of rules, the connection weight of pRBFNNs is represented as three kinds of polynomials such as constant, linear, and quadratic. Coefficients of connection weight identified with back-propagation using gradient descent method. The output of pRBFNNs model is obtained by fuzzy inference method in the inference part of fuzzy rules. The essential design parameters (including learning rate, momentum coefficient and fuzzification coefficient) of the networks are optimized by means of the Particle Swarm Optimization. The proposed pRBFNNs are applied to real-time face recognition system and then demonstrated from the viewpoint of output performance and recognition rate.

• The Journal of Korea Robotics Society
• /
• v.15 no.3
• /
• pp.277-285
• /
• 2020

A differential drive wheeled robot is a kind of mobile robot suitable for indoor navigation. Model predictive control is an optimal control technique with various advantages and can achieve excellent performance. One of the main advantages of model predictive control is that it can easily handle constraints. Therefore, it deals with realistic constraints of the mobile robot and achieves admirable performance for trajectory tracking. In addition, the intention of the robot can be properly realized by adjusting the weight of the cost function component. This control technique is applied to the local planner of the navigation component so that the mobile robot can operate in real environment. Using the Robot Operating System (ROS), which has transcendent advantages in robot development, we have ensured that the algorithm works in the simulation and real experiment.

• The Journal of Korea Robotics Society
• /
• v.18 no.3
• /
• pp.337-345
• /
• 2023

In this paper, we propose a 3D LiDAR sensor-based costmap generation and path planning algorithm using it for reliable autonomous flight in complex indoor environments. 3D path planning is essential for reliable operation of UAVs. However, existing grid search-based or random sampling-based path planning algorithms in 3D space require a large amount of computation, and UAVs with weight constraints require reliable path planning results in real time. To solve this problem, we propose a method that divides a 3D space into several 2D spaces and a path planning algorithm that considers the distance to obstacles within each space. Among the paths generated in each space, the final path (Best path) that the UAV will follow is determined through the proposed objective function, and for this purpose, we consider the rotation angle of the 2D space, the path length, and the previous best path information. The proposed methods have been verified through autonomous flight of UAVs in real environments, and shows reliable obstacle avoidance performance in various complex environments.

• International journal of advanced smart convergence
• /
• v.8 no.4
• /
• pp.68-74
• /
• 2019

Along with the deeper architecture in the deep learning approaches, the need for the data becomes very big. In the real problem, to get huge data in some disciplines is very costly. Therefore, learning on limited data in the recent years turns to be a very appealing area. Meta-learning offers a new perspective to learn a model with this limitation. A state-of-the-art model that is made using a meta-learning framework, Meta-SGD, is proposed with a key idea of learning a hyperparameter or a learning rate of the fast adaptation stage in the outer update. However, this learning rate usually is set to be very small. In consequence, the objective function of SGD will give a little improvement to our weight parameters. In other words, the prior is being a key value of getting a good adaptation. As a goal of meta-learning approaches, learning using a single gradient step in the inner update may lead to a bad performance. Especially if the prior that we use is far from the expected one, or it works in the opposite way that it is very effective to adapt the model. By this reason, we propose to add a weight term to decrease, or increase in some conditions, the effect of this prior. The experiment on few-shot learning shows that emphasizing or weakening the prior can give better performance than using its original value.

• Journal of Korea Multimedia Society
• /
• v.5 no.6
• /
• pp.609-615
• /
• 2002

In this paper, we propose a contrast for the improvement of multi-gray scale image on display system. The proposed method distinguishes a maximum value and a minimum value in input fields or frames. By this judgement, the improvement degree of image quality is decided. This method does not require field and frame memory. Moreover, its lower hardware complexity than conventional methods make it easy to apply this method for flat panel display(FPD) which requires real-time processing. And the contrast of input gray level can be controled flexibly by varying the weight the weight value which controls the contrast range. The proposed method gives an image by controlling weighting slope selectively at intervals according to the brightness-control algorithm and the type of image in the look-up table. The function of the proposed method has been verified through Synopsys VHDL and computer simulation. And its results show that the proposed method can improve the quality of image.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2005.11a
• /
• pp.117-120
• /
• 2005

The business market of architecture has got a system that controls a deposit according to the price function. This system is written on a law of contract about countries. So the main body of construction has to make a reasonable contract. This study is written about a rate of numerical index on controling a deposit. We tried to fine problems and solutions of labor expenses, instrument costs and material costs which is so big and changable on the construction market Labor expenses are expressed according to the rate of construction scale between direct and indirect cost that applies ability of works. Instrument costs are expressed according to an output method of a unit price annually and a weight allowance of local instrument conditions and use frequence. The last material costs expressed according to a local weight allowance make a decision of the material cost index. They applies locally relative index more than absolute one on what uses the price rate of producers and importations. This solutions are not enough to apply to the real market, so it needs to exam and to be on the market after a feasibility study.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.4
• /
• pp.1760-1778
• /
• 2018

For image segmentation with intensity inhomogeneity, many region-based level set methods have been proposed. Some of them however can't get the relatively ideal segmentation results under the severe intensity inhomogeneity and weak edges, and without use of the image gradient information. To improve that, we propose a new level set method combined with local direction gradient in this paper. Firstly, based on two assumptions on intensity inhomogeneity to images, the relationships between segmentation objects and image gradients to local minimum and maximum around a pixel are presented, from which a new pixel classification method based on weight of Euclidian distance is introduced. Secondly, to implement the model, variational level set method combined with image spatial neighborhood information is used, which enhances the anti-noise capacity of the proposed gradient information based model. Thirdly, a new diffusion process with an edge indicator function is incorporated into the level set function to classify the pixels in homogeneous regions of the same segmentation object, and also to make the proposed method more insensitive to initial contours and stable numerical implementation. To verify our proposed method, different testing images including synthetic images, magnetic resonance imaging (MRI) and real-world images are introduced. The image segmentation results demonstrate that our method can deal with the relatively severe intensity inhomogeneity and obtain the comparatively ideal segmentation results efficiently.

• Journal of the Korea Computer Graphics Society
• /
• v.21 no.3
• /
• pp.27-35
• /
• 2015

This study presents a volume rendering method using ambient occlusion which is one of global illumination methods. By considering the volume density distribution as normal distribution, ambient occlusion can be calculated at real-time speed regardless of modification of opacity transfer function. We calculate and store the averages and standard deviations of densities in a block centered at each voxel in pre-processing time. In rendering process, we determine the illumination value by estimating the nearby opacity. We generalized theoretical model and generated better quality images improving our previous research. In detail, various shapes of transfer function can be used due to the proposed equation model. Moreover, we introduced a multi-range model to give nearer objects more weight. As the result, more realistic volume rendering image can be generated at real-time speed by mixing local and ambient occlusion shading.

• Journal of IKEEE
• /
• v.23 no.4
• /
• pp.1314-1320
• /
• 2019

In this paper, we propose a new method for optimizing bit operations and applying them to DNN(Deep Neural Network) in software environment. As a method for this, we propose a packing function for bitwise optimization and a masking matrix multiplication operation for application to DNN. The packing function converts 32-bit real value to 2-bit quantization value through threshold comparison operation. When this sequence is over, four 32-bit real values are changed to one 8-bit value. The masking matrix multiplication operation consists of a special operation for multiplying the packed weight value with the normal input value. And each operation was then processed in parallel using a GPU accelerator. As a result of this experiment, memory saved about 16 times than 32-bit DNN Model. Nevertheless, the accuracy was within 1%, similar to the 32-bit model.