• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1992.10a
• /
• pp.47-51
• /
• 1992

Automatic detection of tool breakage during NC machining is a key issue not only for improving productivity but to implement the unattended manufacturing system. In this paper, we develop a vibration sensor-based tool breakage detection system for NC milling processes. The system obtains the time-domain vibration signal from the sensor attached on the spindle bracket of our CNC machine and declares tool failures through the on-line monitoring schemes. For on-line detection, our approach is to use the PSC(statistical process control) methods being increasingly used for on-line process control. The main thrust of this paper is to propose and compare the performance of SPC methods including : a) X-bar control scheme, b) S control scheme, c)EWMA (exponentially weighted moving average) scheme, and d) AEWMA (adaptive exponentially weighted moving average) scheme. The performance of the control schemes are compared in terms of the type 1 and 2 error calculated from the experiment data.

• Journal of Communications and Networks
• /
• v.14 no.3
• /
• pp.310-318
• /
• 2012

Detecting intrusion attacks accurately and rapidly in wireless networks is one of the most challenging security problems. Intrusion attacks of various types can be detected by the change in traffic flow that they induce. Wireless industrial networks based on the wireless networks for industrial automation-process automation (WIA-PA) standard use a superframe to schedule network communications. We propose an intrusion detection system for WIA-PA networks. After modeling and analyzing traffic flow data by time-sequence techniques, we propose a data traffic prediction model based on autoregressive moving average (ARMA) using the time series data. The model can quickly and precisely predict network traffic. We initialized the model with data traffic measurements taken by a 16-channel analyzer. Test results show that our scheme can effectively detect intrusion attacks, improve the overall network performance, and prolong the network lifetime.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.8
• /
• pp.2964-2983
• /
• 2015

As the Android operating system has become a key target for malware authors, Android protection has become a thriving research area. Beside the proved importance of system permissions for malware analysis, there is a lot of overlapping in permissions between malware apps and goodware apps. The exploitation of them effectively in malware detection is still an open issue. In this paper, to investigate the feasibility of neuro-fuzzy techniques to Android protection based on system permissions, we introduce a self-adaptive neuro-fuzzy inference system to classify the Android apps into malware and goodware. According to the framework introduced, the most significant permissions that characterize optimally malware apps are identified using Information Gain Ratio method and encapsulated into patterns of features. The patterns of features data is used to train and test the system using stratified cross-validation methodologies. The experiments conducted conclude that the proposed classifier can be effective in Android protection. The results also underline that the neuro-fuzzy techniques are feasible to employ in the field.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.6 no.4
• /
• pp.989-1005
• /
• 2012

In this paper, throughput of IEEE 802.11 carrier-sense multiple access (CSMA) with collision-avoidance (CA) protocols in non-saturated traffic conditions is presented taking into account the impact of imperfect channel sensing. The imperfect channel sensing includes both missed-detection and false alarm and their impact on the utilization of IEEE 802.11 analyzed and expressed as a closed form. To include the imperfect channel sensing at the physical layer, we modified the state transition probabilities of well-known two state Markov process model. Simulation results closely match the theoretical expressions confirming the effectiveness of the proposed model. Based on both theoretical and simulated results, the choice of the best probability detection while maintaining probability of false alarm is less than 0.5 is a key factor for maximizing utilization of IEEE 802.11.

• Journal of Microbiology
• /
• v.45 no.4
• /
• pp.297-304
• /
• 2007

The detection of virulence factors of Aeromonas is a key component in determining potential pathogenicity because these factors act multifunctionally and multifactorially. In this study water samples were collected from a trout farm on a seasonal basis, and diseased fish and Aeromonas species were isolated and identified. For rapid detection of six virulence factors of isolated Aeromonas, a hexaplex-polymerase chain reaction (hexaplex-PCR) assay was used. The detected virulence factors include aerolysin (aer), GCAT (gcat), serine protease (ser), nuclease (nuc) lipase (lip) and lateral flagella (laf). The dominant strain found in our isolates was Aeromonas sobria, and the dominant virulence factors were aer and nuc for all seasons. We confirmed that A. sobria and two of the virulence genes (aer and nuc) are related. We proposed a method by which one can identify the major strains of Aeromonas: A. hydrophila, A. sobria, A. caviae, and A. veronii, using hexaplex-PCR.

• Annual Conference of KIPS
• /
• 2018.05a
• /
• pp.232-235
• /
• 2018

Convolutional Neural Network is arguably the most popular deep learning architecture that is one of the most attractive area of research since it has various applications including face detection and recognition. The cascaded CNN operates at multiple resolution and rejects the background regions in the fast low resolution stages. By considering that advantage, we carry out the study on accuracy of cascaded CNN for face detection applications. The key point for our study is to analysing and improving the accuracy of cascaded CNN by applying simulations of algorithm where by we used Google's Tensorflow GPU as deep learning framework.

• Journal of the Korea Society of Computer and Information
• /
• v.24 no.9
• /
• pp.21-27
• /
• 2019

In this paper, we propose a deep auto-encoder-based pipe leak detection (PLD) technique from time-series acoustic data collected by microphone sensor nodes. The key idea of the proposed technique is to learn representative features of the leak-free state using leak-free time-series acoustic data and the deep auto-encoder. The proposed technique can be used to create a PLD model that detects leaks in the pipeline in an unsupervised learning manner. This means that we only use leak-free data without labeling while training the deep auto-encoder. In addition, when compared to the previous supervised learning-based PLD method that uses image features, this technique does not require complex preprocessing of time-series acoustic data owing to the unsupervised feature extraction scheme. The experimental results show that the proposed PLD method using the deep auto-encoder can provide reliable PLD accuracy even considering unsupervised learning-based feature extraction.

• The Journal of Korea Robotics Society
• /
• v.17 no.4
• /
• pp.425-430
• /
• 2022

As robots are considered one of the mainstream digital transformations, robots with machine vision becomes a main area of study providing the ability to check what robots watch and make decisions based on it. However, it is difficult to find a small object in the image mainly due to the flaw of the most of visual recognition networks. Because visual recognition networks are mostly convolution neural network which usually consider local features. So, we make a model considering not only local feature, but also global feature. In this paper, we propose a detection method of a small marker on the object using deep learning and an algorithm that considers global features by combining Transformer's self-attention technique with a convolutional neural network. We suggest a self-attention model with new definition of Query, Key and Value for model to learn global feature and simplified equation by getting rid of position vector and classification token which cause the model to be heavy and slow. Finally, we show that our model achieves higher mAP than state of the art model YOLOr.

• Annual Conference of KIPS
• /
• 2008.05a
• /
• pp.92-94
• /
• 2008

In this paper we describe an approach for template based detection and tracking of objects by chamfer matching in real time video. Detecting and tracking of any objects is the key problem in computer vision. In our case we try for hand and head of human for detection and tracking by chamfer matching technique. Matching involves correlating the templates with the distance transformed scene and determining the locations where the mismatch is below a certain user defined threshold.

• International journal of advanced smart convergence
• /
• v.12 no.4
• /
• pp.190-201
• /
• 2023

Securing transportation safety infrastructure technology for Lv.4 connected autonomous driving is very important for the spread of autonomous vehicles, and the safe operation of level 4 autonomous vehicles in adverse weather has limitations due to the development of vehicle-only technology. We developed the radar-enabled AI convergence transportation entities detection system. This system is mounted on fixed and mobile supports on the road, and provides excellent autonomous driving situation recognition/determination results by converging transportation entities information collected from various monitoring sensors such as 60GHz radar and EO/IR based on artificial intelligence. By installing such a radar-enabled AI convergence transportation entities detection system on an autonomous road, it is possible to increase driving efficiency and ensure safety in adverse weather. To secure competitive technologies in the global market, the development of four key technologies such as ① AI-enabled transportation situation recognition/determination algorithm, ② 60GHz radar development technology, ③ multi-sensor data convergence technology, and ④ AI data framework technology is required.