• Journal of the Korea Society of Computer and Information
• /
• v.26 no.4
• /
• pp.29-37
• /
• 2021

In this paper, we propose a deep learning-based retinal vessel segmentation model for handling multi-scale information of fundus images. we integrate the selective kernel convolution into U-Net-based convolutional neural network. The proposed model extracts and segment features information with various shapes and sizes of retinal blood vessels, which is important information for diagnosing eye-related diseases from fundus images. The proposed model consists of standard convolutions and selective kernel convolutions. While the standard convolutional layer extracts information through the same size kernel size, The selective kernel convolution extracts information from branches with various kernel sizes and combines them by adaptively adjusting them through split-attention. To evaluate the performance of the proposed model, we used the DRIVE and CHASE DB1 datasets and the proposed model showed F1 score of 82.91% and 81.71% on both datasets respectively, confirming that the proposed model is effective in segmenting retinal blood vessels.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.10a
• /
• pp.731-734
• /
• 2009

In this paper, we propose the method the vehicle color recognizing in the image including a vehicle. In an image, the color feature vector of a vehicle is extracted and by using the backpropagation learning algorithm, that is the multi-layer perceptron, the recognized vehicle color. By using the RGB and HSI color model the feature vector used as the input of the backpropagation learning algorithm is the feature of the color used as the input of the neural network. The color of a vehicle recognizes as the white, the silver color, the black, the red, the yellow, the blue, and the green among the color of the vehicle most very much found out as 7 colors. By using the image including a vehicle for the performance evaluation of the method proposing, the color recognition performance was experimented.

• Journal of IKEEE
• /
• v.26 no.3
• /
• pp.456-461
• /
• 2022

In this paper, a two-stage single-ended power amplifier (PA) with broadband gain characteristics was presented by utilizing a radio frequency (RF) transformer (TF), which is essential for a differential amplifier. The bandwidth of a PA can be improved by designing TF to have broadband characteristics and then applying it to the inter-stage matching network (IMN) of a PA. For broadband gain characteristics while maintaining the performance and area of the existing PA, an IMN was implemented on an monolithic microwave integrated circuit (MMIC) and a multi-layer printed circuit board (PCB), and the simulation results were compared. As a result of simulating the PA module designed using InGaP/GaAs HBT model, it has been confirmed that the PA employing the proposed design method has an improved fractional bandwidth of 19.8% at a center frequency of 3.3GHz, while the conventional PA showed that of 11.2%.

• Nuclear Engineering and Technology
• /
• v.53 no.1
• /
• pp.148-163
• /
• 2021

Timely fault identification is important for safe and reliable operation of the electric valve system. Many research works have utilized different data-driven approach for fault diagnosis in complex systems. However, they do not consider specific characteristics of critical control components such as electric valves. This work presents an integrated shallow-deep fault diagnostic model, developed based on signals extracted from DN50 electric valve. First, the local optimal issue of particle swarm optimization algorithm is solved by optimizing the weight search capability, the particle speed, and position update strategy. Then, to develop a shallow diagnostic model, the modified particle swarm algorithm is combined with support vector machine to form a hybrid improved particle swarm-support vector machine (IPs-SVM). To decouple the influence of the background noise, the wavelet packet transform method is used to reconstruct the vibration signal. Thereafter, the IPs-SVM is used to classify phase imbalance and damaged valve faults, and the performance was evaluated against other models developed using the conventional SVM and particle swarm optimized SVM. Secondly, three different deep belief network (DBN) models are developed, using different acoustic signal structures: raw signal, wavelet transformed signal and time-series (sequential) signal. The models are developed to estimate internal leakage sizes in the electric valve. The predictive performance of the DBN and the evaluation results of the proposed IPs-SVM are also presented in this paper.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.2
• /
• pp.149-162
• /
• 2005

This paper suggest a way to detect a specific wanted figure in public places such as subway stations and banks by comparing color face images extracted from the real time CCTV with the face images of designated specific figures. Assuming that the characteristic of the surveillance camera allows the face information in screens to change arbitrarily and to contain information on numerous faces, the accurate detection of the face area was focused. To solve this problem, the normalization work using subsampling with $20{\times}20$ pixels on arbitrary face images, which is based on the Perceptron Neural Network model suggested by R. Rosenblatt, created the effect of recogning the whole face. The optimal linear filter and the histogram shaper technique were employed to minimize the outside interference such as lightings and light. The addition operation of the egg-shaped masks was added to the pre-treatment process to minimize unnecessary work. The images finished with the pre-treatment process were divided into three reception fields and the information on the specific location of eyes, nose, and mouths was determined through the neural network. Furthermore, the precision of results was improved by constructing the three single-set network system with different initial values in a row.

• Journal of the Korea Society of Computer and Information
• /
• v.5 no.2
• /
• pp.30-36
• /
• 2000

The hidden node plays a role of the functional units that classifies the features of input pattern in the given question. Therefore, a neural network that consists of the number of a suitable optimum hidden node has be on the rise as a factor that has an important effect upon a result. However there is a problem that decides the number of hidden nodes based on back-propagation learning algorithm. If the number of hidden nodes is designated very small perfect learning is not done because the input pattern given cannot be classified enough. On the other hand, if designated a lot, overfitting occurs due to the unnecessary execution of operation and extravagance of memory point. So, the recognition rate is been law and the generality is fallen. Therefore, this paper suggests a method that decides the number of neural network node with feature information consisted of the parameter of learning algorithm. It excludes a node in the Pruning target, that has a maximum value among the feature value obtained and compares the average of the rest of hidden node feature value with the feature value of each hidden node, and then would like to improve the learning speed of neural network deciding the optimum structure of the multi-layer neural network as pruning the hidden node that has the feature value smaller than the average.

• Journal of Radiation Protection and Research
• /
• v.40 no.1
• /
• pp.55-64
• /
• 2015

Dose calculations which are a crucial requirement for radiotherapy treatment planning systems require accuracy and rapid calculations. The conventional radiotherapy treatment planning dose algorithms are rapid but lack precision. Monte Carlo methods are time consuming but the most accurate. The new combined system that Monte Carlo methods calculate part of interesting domain and the rest is calculated by neural can calculate the dose distribution rapidly and accurately. The preliminary study showed that neural networks can map functions which contain discontinuous points and inflection points which the dose distributions in inhomogeneous media also have. Performance results between scaled conjugated gradient algorithm and Levenberg-Marquardt algorithm which are used for training the neural network with a different number of neurons were compared. Finally, the dose distributions of homogeneous phantom calculated by a commercialized treatment planning system were used as training data of the neural network. In the case of homogeneous phantom;the mean squared error of percent depth dose was 0.00214. Further works are programmed to develop the neural network model for 3-dimensinal dose calculations in homogeneous phantoms and inhomogeneous phantoms.

• Journal of Internet of Things and Convergence
• /
• v.10 no.2
• /
• pp.1-15
• /
• 2024

The construction of smart communities is a new method and important measure to ensure the security of residential areas. In order to solve the problem of low accuracy in face recognition caused by distorting facial features due to monitoring camera angles and other external factors, this paper proposes the following optimization strategies in designing a face recognition network: firstly, a global graph convolution module is designed to encode facial features as graph nodes, and a multi-scale feature enhancement residual module is designed to extract facial keypoint features in conjunction with the global graph convolution module. Secondly, after obtaining facial keypoints, they are constructed as a directed graph structure, and graph attention mechanisms are used to enhance the representation power of graph features. Finally, tensor computations are performed on the graph features of two faces, and the aggregated features are extracted and discriminated by a fully connected layer to determine whether the individuals' identities are the same. Through various experimental tests, the network designed in this paper achieves an AUC index of 85.65% for facial keypoint localization on the 300W public dataset and 88.92% on a self-built dataset. In terms of face recognition accuracy, the proposed network achieves an accuracy of 83.41% on the IBUG public dataset and 96.74% on a self-built dataset. Experimental results demonstrate that the network designed in this paper exhibits high detection and recognition accuracy for faces in surveillance videos.

• The Journal of Engineering Geology
• /
• v.30 no.4
• /
• pp.457-468
• /
• 2020

Problems arising during pile design works for plant construction, civil and architecture work are mostly come from uncertainty of geotechnical characteristics. In particular, obtaining the N-value measured through the Standard Penetration Test (SPT) is the most important data. However, it is difficult to obtain N-value by drilling investigation throughout the all target area. There are many constraints such as licensing, time, cost, equipment access and residential complaints etc. it is impossible to obtain geotechnical characteristics through drilling investigation within a short bidding period in overseas. The geotechnical characteristics at non-drilling investigation points are usually determined by the engineer's empirical judgment, which can leads to errors in pile design and quantity calculation causing construction delay and cost increase. It would be possible to overcome this problem if N-value could be predicted at the non-drilling investigation points using limited minimum drilling investigation data. This study was conducted to predicted the N-value using an Artificial Neural Network (ANN) which one of the Artificial intelligence (AI) method. An Artificial Neural Network treats a limited amount of geotechnical characteristics as a biological logic process, providing more reliable results for input variables. The purpose of this study is to predict N-value at the non-drilling investigation points through patterns which is studied by multi-layer perceptron and error back-propagation algorithms using the minimum geotechnical data. It has been reviewed the reliability of the values that predicted by AI method compared to the measured values, and we were able to confirm the high reliability as a result. To solving geotechnical uncertainty, we will perform sensitivity analysis of input variables to increase learning effect in next steps and it may need some technical update of program. We hope that our study will be helpful to design works in the future.

• Journal of Korea Water Resources Association
• /
• v.55 no.spc1
• /
• pp.1283-1293
• /
• 2022

The purpose of this study is to predict residual chlorine in order to maintain stable residual chlorine concentration in sedimentation basin by using artificial intelligence algorithms in water treatment process employing pre-chlorination. Available water quantity and quality data are collected and analyzed statistically to apply into mathematical multiple regression and artificial intelligence models including multi-layer perceptron neural network, random forest, long short term memory (LSTM) algorithms. Water temperature, turbidity, pH, conductivity, flow rate, alkalinity and pre-chlorination dosage data are used as the input parameters to develop prediction models. As results, it is presented that the random forest algorithm shows the most moderate prediction result among four cases, which are long short term memory, multi-layer perceptron, multiple regression including random forest. Especially, it is result that the multiple regression model can not represent the residual chlorine with the input parameters which varies independently with seasonal change, numerical scale and dimension difference between quantity and quality. For this reason, random forest model is more appropriate for predict water qualities than other algorithms, which is classified into decision tree type algorithm. Also, it is expected that real time prediction by artificial intelligence models can play role of the stable operation of residual chlorine in water treatment plant including pre-chlorination process.