• Journal of KIISE:Software and Applications
• /
• v.30 no.11
• /
• pp.1054-1061
• /
• 2003

Q-Learning, a representative algorithm of reinforcement learning, experiences repeatedly until estimation values about all state-action pairs of state space converge and achieve optimal policies. When the state space is high dimensional or continuous, complex reinforcement learning tasks involve very large state space and suffer from storing all individual state values in a single table. We introduce Q-Map that is new function approximation method to get classified policies. As an agent learns on-line, Q-Map groups states of similar situations and adapts to new experiences repeatedly. State-action pairs necessary for fine control are treated in the form of rule. As a result of experiment in maze environment and mountain car problem, we can achieve classified knowledge and extract easily rules from Q-Map

• International conference on construction engineering and project management
• /
• 2024.07a
• /
• pp.264-270
• /
• 2024

In the construction industry, steel structures are prominent due to their exceptional strength and high bearing capacity, making them resilient against natural calamities. However, the stability and overall structural integrity of these steel structures depend significantly on the precision of the individual steel members used. Presently, the dimensions of these steel members are typically measured manually using mechanical instruments such as steel tape and vernier calipers. This conventional approach is not only time-consuming but also highly vulnerable to human error. Consequently, there is a growing need for more accurate and reliable methods for assessing the dimensions of steel members. This paper aims to measure the dimensions of key checklists of the cross-section surface of the steel H-beams using Terrestrial Laser Scan (TLS) data. This study involves the automatic extraction of scan points associated with the cross-section surface of the H-beam members using RANSAC. By the end, an algorithm was developed to predict the actual edge points belonging to the boundary of the extracted surface and introduced an edge loss compensation model to compensate the losses occurred due to uncertainties. Experimental evaluations were conducted using various scan data collected from steel H-beam and the measured dimensions were subsequently compared with manual measurements and dimensions obtained through the previously proposed method, demonstrating that the measurements meet 1mm accuracy and are within the allowable tolerance range followed in industry. This research underscores the efficiency and reliability of the introduced approach, offering a promising solution to enhance the dimensional quality assessment of steel H-beams in the construction industry.

• The KIPS Transactions:PartB
• /
• v.18B no.3
• /
• pp.147-156
• /
• 2011

This paper proposes a 3-stage (preprocessing, feature extraction, and classification) fault detection and classification algorithm for induction motors. In the first stage, a low-pass filter is used to remove noise components in the fault signal. In the second stage, a discrete cosine transform (DCT) and a statistical method are used to extract features of the fault signal. Finally, a back propagation neural network (BPNN) method is applied to classify the fault signal. To evaluate the performance of the proposed algorithm, we used one second long normal/abnormal vibration signals of an induction motor sampled at 8kHz. Experimental results showed that the proposed algorithm achieves about 100% accuracy in fault classification, and it provides 50% improved accuracy when compared to the existing fault detection algorithm using a cross-covariance method. In a real-world data acquisition environment, unnecessary noise components are usually included to the real signal. Thus, we conducted an additional simulation to evaluate how well the proposed algorithm classifies the fault signals in a circumstance where a white Gaussian noise is inserted into the fault signals. The simulation results showed that the proposed algorithm achieves over 98% accuracy in fault classification. Moreover, we developed a testbed system including a TI's DSP (digital signal processor) to implement and verify the functionality of the proposed algorithm.

• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.1-16
• /
• 2022

Despite recent breakthroughs in deep learning and computer vision fields, the pixel-wise identification of tiny objects in high-resolution images with complex disturbances remains challenging. This study proposes a modified U-net for tiny crack segmentation in real-world steel-box-girder bridges. The modified U-net adopts the common U-net framework and a novel Self-Attention-Self-Adaption (SASA) neuron as the fundamental computing element. The Self-Attention module applies softmax and gate operations to obtain the attention vector. It enables the neuron to focus on the most significant receptive fields when processing large-scale feature maps. The Self-Adaption module consists of a multiplayer perceptron subnet and achieves deeper feature extraction inside a single neuron. For data augmentation, a grid-based crack random elastic deformation (CRED) algorithm is designed to enrich the diversities and irregular shapes of distributed cracks. Grid-based uniform control nodes are first set on both input images and binary labels, random offsets are then employed on these control nodes, and bilinear interpolation is performed for the rest pixels. The proposed SASA neuron and CRED algorithm are simultaneously deployed to train the modified U-net. 200 raw images with a high resolution of 4928 × 3264 are collected, 160 for training and the rest 40 for the test. 512 × 512 patches are generated from the original images by a sliding window with an overlap of 256 as inputs. Results show that the average IoU between the recognized and ground-truth cracks reaches 0.409, which is 29.8% higher than the regular U-net. A five-fold cross-validation study is performed to verify that the proposed method is robust to different training and test images. Ablation experiments further demonstrate the effectiveness of the proposed SASA neuron and CRED algorithm. Promotions of the average IoU individually utilizing the SASA and CRED module add up to the final promotion of the full model, indicating that the SASA and CRED modules contribute to the different stages of model and data in the training process.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.43 no.2 s.308
• /
• pp.11-19
• /
• 2006

In this paper, we propose a method to estimate camera extrinsic parameter based on projective and permutation invariance point features. Because feature informations in previous research is variant to c.:men viewpoint, extraction of correspondent point is difficult. Therefore, in this paper, we propose the extracting method of invariant point features, and new matching method using similarity evaluation function and Graham search method for reducing time complexity and finding correspondent points accurately. In the calculation of camera extrinsic parameter stage, we also propose two-stage motion parameter estimation method for enhancing convergent degree of LM algorithm. In the experiment, we compare and analyse the proposed method with existing method by using various indoor images to demonstrate the superiority of the proposed algorithms.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09a
• /
• pp.260-263
• /
• 2003

EEC is an electrical signal, which occurs during information processing in the brain. These EEG signals has been used clinically, but nowadays we are mainly studying Brain-Computer Interface(BCI) such as interfacing with a computer through the EEG controlling the machine through the EEG The ultimate purpose of BCI study is specifying the EEG at various mental states so as to control the computer and machine. A BCI has to perform two tasks, the parameter estimation task, which attemps to describe the properties of the EEG signal and the classification task, which separates the different EEC patterns based on the estimated parameters. First, we have to do parameter estimation of EEG to embody BCI system. It is important to improve performance of classifier, But, It is not easy to do parameter estimation by reason of EEG is sensitivity and undergo various influences. Therefore, this research should do parameter estimation and classification of the EEG to use various analysis algorithm.

• Journal of the Korean Society for Precision Engineering
• /
• v.23 no.1 s.178
• /
• pp.64-71
• /
• 2006

Ultraprecision machining and MEMS technology have been taken more and more important position in machining of microparts. Micro endmilling is one of the prominent technology that has wide spectrum of application field ranging from macro parts to micro products. Also, the method of micro-grooving using micro endmill is used widely owing to many merit, but has problems of precision and quality of products due to tool wear and tool fracture. This investigation deals with state monitoring using acoustic emission(AE) signal in the micro-grooving. Characteristic evaluation of AE raw signal, AE hit and frequency analysis for condition monitoring is presented. Also, the feature extraction of AE signal directly related to machining process is executed. Then, the distinctive micro endmill state according to the each tool condition is classified by the fuzzy C-means algorithm.

• Journal of Electrical Engineering and Technology
• /
• v.10 no.4
• /
• pp.1899-1909
• /
• 2015

Glaucoma is one of the most common causes of blindness which is caused by increase of fluid pressure in the eye which damages the optic nerve and eventually causing vision loss. An automated technique to diagnose glaucoma disease can reduce the physicians’ effort in screening of Glaucoma in a person through the fundal retinal images. In this paper, optimal hyper analytic wavelet transform for Glaucoma detection technique from fundal retinal images is proposed. The optimal coefficients for transformation process are found out using the hybrid GSO-Cuckoo search algorithm. This technique consists of pre-processing module, optimal transformation module, feature extraction module and classification module. The implementation is carried out with MATLAB and the evaluation metrics employed are accuracy, sensitivity and specificity. Comparative analysis is carried out by comparing the hybrid GSO with the conventional GSO. The results reported in our paper show that the proposed technique has performed well and has achieved good evaluation metric values. Two 10- fold cross validated test runs are performed, yielding an average fitness of 91.13% and 96.2% accuracy with CGD-BPN (Conjugate Gradient Descent- Back Propagation Network) and Support Vector Machines (SVM) respectively. The techniques also gives high sensitivity and specificity values. The attained high evaluation metric values show the efficiency of detecting Glaucoma by the proposed technique.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.67 no.11
• /
• pp.1496-1505
• /
• 2018

The technology of emotion recognition is necessary for human computer interaction communication. There are many cases where one cannot communicate without considering one's emotion. As such, emotional recognition technology is an essential element in the field of communication. n this regard, it is highly utilized in various fields. Various bio-sensor sensors are used for human emotional recognition and can be used to measure emotions. This paper proposes a system for recognizing human emotions using two physiological sensors. For emotional classification, two-dimensional Russell's emotional model was used, and a method of classification based on personality was proposed by extracting sensor-specific characteristics. In addition, the emotional model was divided into four emotions using the Support Vector Machine classification algorithm. Finally, the proposed emotional recognition system was evaluated through a practical experiment.

• The Journal of the Korea Contents Association
• /
• v.7 no.8
• /
• pp.41-49
• /
• 2007

BCI system integrates control or telecommunication system with generating electric signals in scalp itself after signal acquisition. This system detect a movement of EEG at real time, can control an electron equipment using a generated signal through EEG movement or software-based processor. In this paper, we deal with removing and separating artifacts induceced from measurement when brain-computer interface system that analyzes recognizes EEG signals occurred from various mental states. In this paper, we proposed a method of EEG classification and an artifact interval detection using bisection mathematical modeling in the EEG classification process for BCI system implementation.