• Journal of Industrial Technology
• /
• v.28 no.B
• /
• pp.143-148
• /
• 2008

A real-time face detection is to find human faces robustly under the cluttered background free from the effect of occlusion by other objects or various lightening conditions. We propose a face detection system for real-time applications using wavelet decomposition method based on Gabor features. Firstly, skin candidate regions are extracted from the given image by skin color filtering and projection method. Then Gabor-feature based template matching is performed to choose face cadidate from the skin candidate regions. The chosen face candidate region is transformed into 2-level wavelet decomposition images, from which feature vectors are extracted for classification. Based on the extracted feature vectors, the face candidate region is finally classified into either face or nonface class by the Levenberg-Marguardt back-propagation neural network.

• Computers and Concrete
• /
• v.21 no.5
• /
• pp.505-511
• /
• 2018

This paper investigates the potential of a hybrid model which combines the least squares support vector machine (LSSVM) and an improved particle swarm optimization (IMPSO) techniques for prediction of concrete compressive strength. A modified PSO algorithm is employed in determining the optimal values of LSSVM parameters to improve the forecasting accuracy. Experimental data on concrete compressive strength in the literature were used to validate and evaluate the performance of the proposed IMPSO-LSSVM model. Further, predictions from five models (the IMPSO-LSSVM, PSO-LSSVM, genetic algorithm (GA) based LSSVM, back propagation (BP) neural network, and a statistical model) were compared with the experimental data. The results show that the proposed IMPSO-LSSVM model is a feasible and efficient tool for predicting the concrete compressive strength with high accuracy.

• Proceedings of the Korean Information Science Society Conference
• /
• 2001.10b
• /
• pp.160-162
• /
• 2001

최근 대용량의 데이터베이스로부터 유용한 정보를 발견하고 데이터간에 존재하는 연관성을 탐색하고 분석하는 데이터 마이닝에 관한 많은 연구들이 진행되고 있다. 실제 응용분야에선 수집된 데이터는 시간이 지날수록 데이터의 양이 늘어나게 되고, 중복되는 속성과 잡음을 갖게 되어 마이닝 기법을 이용하는데 많은 시간과 비용이 소요된다. 또한 어느 속성이 중요한지 알 수 없어 중요한 속성이 중요하지 않은 속성에 의해 왜곡되거나 제대로 분석되지 않을 수 있다. 이 논문은 이러한 문제점들을 해결하기 위해, 대용량의 데이터에 적용할 수 있고 데이터에서 알려지지 않은 패턴을 발견할 뿐만 아니라, 사용자가 얻고자 하는 출력을 생성할 수 있는 혼합형 신경망 클러스터링 기법을 제안한다. 그리고 알고리즘의 타당성을 검증하기 위해 몇 가지 벤치마크데이터를 이용하여 본 논문의 타당성을 보인다.

• Proceedings of the KIEE Conference
• /
• 1997.07c
• /
• pp.1021-1024
• /
• 1997

This paper has presented a method for self-tuning tile PID controller using a BP method of multilayered NNs. The proposed controller employ input signal as a learning signal of PID control. The proposed controller is applied to load-frequency control of power system and it is investigated a dynamic characteristic. The simulation results shows that proposed NN STPID controller has the good dynamics responses against load disturbances.

• International Journal of Control, Automation, and Systems
• /
• v.4 no.5
• /
• pp.592-600
• /
• 2006

The pattern recognition of the complicated grasping operation based on electroencephalography (simply named as EEG) is very helpful on realtime control of the robotic hand. In the paper, a new spectral feature analysis method based on Band Pass Filter (simply named as BPF) and Power Spectral Analysis (simply named as PSA) is presented for discriminating the complicated grasping operations. By analyzing the spectral features of grasping operations with the use of the two-channel EEG measurement system and the pattern recognition of the BP neural network, the degree of recognition by the traditional spectral feature method based on FFT and the new spectral features method based on BPF and PSA could be compared. The results show that the proposed method provides highly improved performance than the traditional one because the new method has two obvious advantages such as high recognition capability and the fast learning speed.

• Proceedings of the KIEE Conference
• /
• 1994.11a
• /
• pp.21-23
• /
• 1994

The degree of forecast accuracy with BP-algorithm largely depends upon the neuron number in hidden layer. In order to construct the optimal structure, first, we prescribe the error bounds of learning procedure, and then, we provid the method of incrementing the number of hidden neurons by using the derivative of errors with respect to an output neuron weights. For the case study, we apply the proposed method to forecast the component-wise residential load, and compare this results to that of time series forecasting.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.10
• /
• pp.4865-4885
• /
• 2019

We propose a FPGA based design that performs real-time power-efficient analysis of heterogeneous sensor data using adaptive ANN on edge gateway of smart military wearables. In this work, four independent ANN classifiers are developed with optimum topologies. Out of which human activity, BP and toxic gas classifier are multiclass and ECG classifier is binary. These classifiers are later integrated into a single adaptive ANN hardware with a select line(s) that switches the hardware architecture as per the sensor type. Five versions of adaptive ANN with different precisions have been synthesized into IP cores. These IP cores are implemented and tested on Xilinx Artix-7 FPGA using Microblaze test system and LabVIEW based sensor simulators. The hardware analysis shows that the adaptive ANN even with 8-bit precision is the most efficient IP core in terms of hardware resource utilization and power consumption without compromising much on classification accuracy. This IP core requires only 31 microseconds for classification by consuming only 12 milliwatts of power. The proposed adaptive ANN design saves 61% to 97% of different FPGA resources and 44% of power as compared with the independent implementations. In addition, 96.87% to 98.75% of data throughput reduction is achieved by this edge gateway.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2010.05a
• /
• pp.205-208
• /
• 2010

We introduce a novel feature extraction scheme for online handwritten digit based on utilizing Zernike moment and angulation feature. The time sequential signal from mouse movement on the writing pad is described as a sequence of consecutive points on the x-y plane. So, we can create data-set which are successive and time-sequential pixel position data by preprocessing. Data preprocessed is used for Zernike moment and angulation feature extraction. this feature is scale-, translation-, and rotation-invariant. The extracted specific feature is fed to a BP(backpropagation) neural network, which in turn classifies it as one of the nine digits. In this paper, proposed method not noly show high recognition rate but also need less learning data for 200 handwritten digit data.

• Journal of Korean Society of Steel Construction
• /
• v.16 no.4 s.71
• /
• pp.443-452
• /
• 2004

Recently, the Artificial Neural Network(ANN) which can organize complex non-linear problems by effectively applying the parallel computational model that is similar to the human brain, was adopted in the wide department of technology and resulted in many successful applications. In this study, a more appropriate formal method is suggested for the preliminary structural design stage controlled merely by the designer's experience and intuition. To do so, this study proposes a multi-level ANN according to the general progressive structural design procedure, using Back-Propagation Algorithm (BP) and Genetic Algorithm (GA) for the ANN learning. The preliminary structural design of cable-stayed bridges was applied to illustrate the applicability of the study formulated as stated above, and the results of two different learning methods were compared.

• Journal of Sensor Science and Technology
• /
• v.7 no.6
• /
• pp.418-425
• /
• 1998

Stereo cameras are the most widely used sensing systems for automated machines including robots to interact with their three-dimensional(3D) working environments. The position of a target point in the 3D world coordinates can be measured by the use of stereo cameras and the camera calibration is an important preliminary step for the task. Existing camera calibration techniques can be classified into two large categories - linear and nonlinear techniques. While linear techniques are simple but somewhat inaccurate, the nonlinear ones require a modeling process to compensate for the lens distortion and a rather complicated procedure to solve the nonlinear equations. In this paper, a method employing a neural network for the calibration problem is described for tackling the problems arisen when existing techniques are applied and the results are reported. Particularly, it is shown experimentally that by utilizing the function approximation capability of multi-layer neural networks trained by the back-propagation(BP) algorithm to learn the error pattern of a linear technique, the measurement accuracy can be simply and efficiently increased.