• Journal of the Korea Society of Computer and Information
• /
• v.23 no.10
• /
• pp.11-21
• /
• 2018

In this paper we propose a novel log template discovery algorithm which achieves high quality of discovered log templates through iterative log filtering technique. Log templates are the static string pattern of logs that are used to produce actual logs by inserting variable values during runtime. Identifying individual logs into their template category correctly enables us to conduct automated analysis using state-of-the-art machine learning techniques. Our technique looks at the group of logs column-wise and filters the logs that have the value of the highest proportion. We repeat this process per each column until we are left with highly homogeneous set of logs that most likely belong to the same log template category. Then, we determine which column is the static part and which is the variable part by vertically comparing all the logs in the group. This process repeats until we have discovered all the templates from given logs. Also, during this process we discover the custom patterns such as ID formats that are unique to the application. This information helps us quickly identify such strings in the logs as variable parts thereby further increasing the accuracy of the discovered log templates. Existing solutions suffer from log templates being too general or too specific because of the inability to detect custom patterns. Through extensive evaluations we have learned that our proposed method achieves 2 to 20 times better accuracy.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.11 no.18
• /
• pp.7-15
• /
• 1988

This study is concerned with selecting mutually dependent quality improvement alternatives with resource constraints. These qualify improvement alternatives art different fro the tradition at alternatives which are independent from each other. In other words, selection of any improvement alternative requires other related specific improvement. Also the overall product quality in a multi stage manufacturing process is characterized by a complex multiplication method rather than a simple addition method which dose not allow to solve a linear knapsack problem despite its popularity in the traditional study. This study suggests a non-linear integer programming model for selecting mutually dependent quality improvement alternatives in multi-stage manufacturing process. In order to apply the model to selecting alternatives. This study also suggests a heuristic mode1 based on a dynamic programming model which is more practical than the non-linear integer programming model. The logic of the heuristic model enables 1) to estimate improvement effectiveness values on all improvement alternatives specifically defined for this study. 2) to arrange the effectiveness values in a descending order, and 3) to select the best one among the alternatives based on their forward and backward linkage relationships. This process repeats to selects other best alternatives within the resource constraints. This process is presented in a Computer programming in Appendix A. Alsc a numerical example of model application is presented in Chapter 4.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2013.10a
• /
• pp.338-345
• /
• 2013

Induction motors play an increasing importance in industrial manufacturing. Therefore, the state monitoring systems also have been considering as the key in dealing with their negative effect by absorbing faulty symptoms in motors. There are numerous proposed systems in literature, in which, several kinds of signals are utilized as the input. To solve the multiple faults problem of induction motors, like the proposed system, the vibration signals is good candidate. In this study, a new signal processing scheme was utilized, which transforms the time domain vibration signal into the spatial domain as an image. Then the spatial features of converted image then have been extracted by applying the dominant neighbourhood structure (DNS) algorithm. In addition, these feature vectors were evaluated to obtain the fruitful dimensions, which support to discriminate between states of motors. Because of reliability, the conventional one-against-all (OAA) multi-class support vector machines (MCSVM) have been utilized in the proposed system as classifier module. Even though examined in severity levels of signal-to-noise ratio (SNR), up to 15dB, the proposed system still reliable in term of two criteria: true positive (TF) and false positive (FP). Furthermore, it also offers better performance than five state-of-the-art systems.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.12
• /
• pp.4776-4794
• /
• 2020

We propose a rapid adaptive learning framework for streaming object detection, called EER-ASSL. The method combines the expected error reduction (EER) dependent rollback learning and the active semi-supervised learning (ASSL) for a rapid adaptive CNN detector. Most CNN object detectors are built on the assumption of static data distribution. However, images are often noisy and biased, and the data distribution is imbalanced in a real world environment. The proposed method consists of collaborative sampling and EER-ASSL. The EER-ASSL utilizes the active learning (AL) and rollback based semi-supervised learning (SSL). The AL allows us to select more informative and representative samples measuring uncertainty and diversity. The SSL divides the selected streaming image samples into the bins and each bin repeatedly transfers the discriminative knowledge of the EER and CNN models to the next bin until convergence and incorporation with the EER rollback learning algorithm is achieved. The EER models provide a rapid short-term myopic adaptation and the CNN models an incremental long-term performance improvement. EER-ASSL can overcome noisy and biased labels in varying data distribution. Extensive experiments shows that EER-ASSL obtained 70.9 mAP compared to state-of-the-art technology such as Faster RCNN, SSD300, and YOLOv2.

• ETRI Journal
• /
• v.44 no.2
• /
• pp.286-299
• /
• 2022

Human activity recognition in real time is a challenging task. Recently, a plethora of studies has been proposed using deep learning architectures. The implementation of these architectures requires the high computing power of the machine and a massive database. However, handcrafted features-based machine learning models need less computing power and very accurate where features are effectively extracted. In this study, we propose a handcrafted model based on three-dimensional sequential skeleton data. The human body skeleton movement over a frame is computed through joint positions in a frame. The joints of these skeletal frames are projected into two-dimensional space, forming a "movement polygon." These polygons are further transformed into a one-dimensional space by computing amplitudes at different angles from the centroid of polygons. The feature vector is formed by the sampling of these amplitudes at different angles. The performance of the algorithm is evaluated using a support vector machine on four public datasets: MSR Action3D, Berkeley MHAD, TST Fall Detection, and NTU-RGB+D, and the highest accuracies achieved on these datasets are 94.13%, 93.34%, 95.7%, and 86.8%, respectively. These accuracies are compared with similar state-of-the-art and show superior performance.

• Structural Engineering and Mechanics
• /
• v.85 no.1
• /
• pp.119-133
• /
• 2023

Impact event is the key factor influencing the operational state of the mechanical equipment. Additionally, nonlinear factors existing in the complex mechanical equipment which are currently attracting more and more attention. Therefore, this paper proposes a novel hybrid-separate identification strategy to solve the force identification problem of the nonlinear structure under impact excitation. The 'hybrid' means that the identification strategy contains both l1-norm (sparse) and l2-norm regularization methods. The 'separate' means that the nonlinear response part only generated by nonlinear force needs to be separated from measured response. First, the state-of-the-art two-step iterative shrinkage/thresholding (TwIST) algorithm and sparse representation with the cubic B-spline function are developed to solve established normalized sparse regularization model to identify the accurate impact force and accurate peak value of the nonlinear force. Then, the identified impact force is substituted into the nonlinear response separation equation to obtain the nonlinear response part. Finally, a reduced transfer equation is established and solved by the classical Tikhonove regularization method to obtain the wave profile (variation trend) of the nonlinear force. Numerical and experimental identification results demonstrate that the novel hybrid-separate strategy can accurately and efficiently obtain the nonlinear force and impact force for the nonlinear structure.

• International Journal of Computer Science & Network Security
• /
• v.23 no.3
• /
• pp.10-16
• /
• 2023

The ever-increasing amount of data generated by various industries and systems has led to the development of data mining techniques as a means to extract valuable insights and knowledge from such data. The electrical energy industry is no exception, with the large amounts of data generated by SCADA systems. This study focuses on the analysis of historical data recorded in the SCADA database of the Libyan Electricity Company. The database, spanned from January 1st, 2013, to December 31st, 2022, contains records of daily date and hour, energy production, temperature, humidity, wind speed, and energy consumption levels. The data was pre-processed and analyzed using the WEKA tool and the Apriori algorithm, a supervised machine learning technique. The aim of the study was to extract association rules that would assist decision-makers in making informed decisions with greater efficiency and reduced costs. The results obtained from the study were evaluated in terms of accuracy and production time, and the conclusion of the study shows that the results are promising and encouraging for future use in the Libyan Electricity Company. The study highlights the importance of data mining and the benefits of utilizing machine learning technology in decision-making processes.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.11a
• /
• pp.113-117
• /
• 2020

Images were taken under various weather such as rain, haze, snow often show low visibility, which can dramatically decrease accuracy of some tasks in computer vision: object detection, segmentation. Besides, previous work to enhance image usually downsample the image to receive consistency features but have not yet good upsample algorithm to recover original size. So, in this research, we jointly implement removal streak in heavy rain image and super resolution using a deep network. We put forth a 2-stage network: a multi-model network followed by a refinement network. The first stage using rain formula in the single image and two operation layers (addition, multiplication) removes rain streak and noise to get clean image in low resolution. The second stage uses refinement network to recover damaged background information as well as upsample, and receive high resolution image. Our method improves visual quality image, gains accuracy in human action recognition task in datasets. Extensive experiments show that our network outperforms the state of the art (SoTA) methods.

• Journal of Broadcast Engineering
• /
• v.15 no.2
• /
• pp.265-279
• /
• 2010

A color image requires at least three color channels to obtain the full color image. However the image sensor obtains only the intensity of the brightness, that is, three image sensors are required for every pixel to capture the full color image. Since the image sensor is quiet expensive, most of digital still cameras adopt single image sensor array with color filter array (CFA) to reduce the size and the cost. Since the image obtained using single sensor array has only one color component per pixel, we need to reconstruct the missing two color components to obtain the full color image. We call this process as color filter interpolation or demosaicking. In this paper, demosaicking algorithm composed of two large step is proposed. Proposed algorithm is combined with several different algorithms such as Edge-directed demosaicking, Second-order gradients as correction terms, Smooth hue transition Interpolation, etc. The simulation results show that the proposed algorithm performs much better than the state-of-the-art demosaicking algorithms in terms of both subjective and objective qualities.

• Journal of KIISE:Computing Practices and Letters
• /
• v.10 no.4
• /
• pp.309-318
• /
• 2004

This paper proposes the H.264-based selective fine granular scalable (FGS) coding scheme that selectively uses the temporal prediction data in the enhancement layer. The base layer of the proposed scheme is basically coded by the H.264 (MPEG-4 Part 10 AVC) visual coding scheme that is the state-of-art in codig efficiency. The enhancement layer is basically coded by the same bitplane-based algorithm of the MPEG-4 (Part 2) fine granular scalable coding scheme. In this paper, we introduce a new algorithm that uses the temproal prediction mechanism inside the enhancement layer and the effective selection mechanism to decide whether the temporally-predicted data would be sent to the decoder or not. Whenever applying the temporal prediction inside the enhancement layer, the temporal redundancies may be effectively reduced, however the drift problem would be severly occurred especially at the low bitrate transmission, due to the mismatch bewteen the encoder's and decoder's reference frame images. Proposed algorithm selectively uses the temporal-prediction data inside the enhancement layer only in case those data could siginificantly reduce the temporal redundancies, to minimize the drift error and thus to improve the overall coding efficiency. Simulation results, based on several test image sequences, show that the proposed scheme has 1∼3 dB higher coding efficiency than the H.264-based FGS coding scheme, even 3∼5 dB higher coding efficiency than the MPEG-4 FGS international standard.