• Journal of Electrical Engineering and Technology
• /
• v.2 no.1
• /
• pp.118-122
• /
• 2007

In this work, experimental investigation is carried out in order to design and fabricate the UHF sensor that is able to detect the partial discharges produced from 10 artificial defects introduced into the real scale 70kV GIS mock-up under high voltage within a well shielded room. As well, in order to verify the on-site applicability of our method, the newly proposed CAPD (chaos analysis of partial discharge) is combined with spectral analysis for identifying the nature of 10 artificial defects under investigation. The PD pattern recognition of each defect has been fulfilled by applying our ANN software. The result indicates that the recognition rate reaches up to 80% by the newly proposed method while the traditional PRPD analysis method allows us to obtain 41%. In consequence, it can be pointed out that the proposed method seems likely to be applicable to the real GIS at the site.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.7
• /
• pp.1873-1880
• /
• 1997

It is an important factor to distinguish the kind of the character for increasing recognition rate before the character recognition in the document recognition system composed of the multi-font and multi-letters. All the letters of each country have a various unique characteristic in the each composition. In this paper, we used the stroke density as a method to distinguish the letter, and it has been adopted only Korean and English character. Input data is processed by the normalization to adopt multi-font document. Proposed method has been proved by the results of experiment the fact that the distinction probability of the Korean and English is more than 90%.

• Smart Media Journal
• /
• v.5 no.3
• /
• pp.25-29
• /
• 2016

This paper presents research for non-contact gesture recognition system using EPS(Electronic Potential Sensor) for measuring the human body of electromagnetic fields. It implemented a signal acquisition and signal processing system for designing a system suitable for motion recognition using the data coming from the sensors. we transform AC-type data into DC-type data by applying a 10Hz LPF considering H/W sampling rate. in addition, we extract 2-dimensional movement information by taking difference value between two cross-diagonal deployed sensor.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2017.10a
• /
• pp.166-169
• /
• 2017

Currently, studies using Deep Learning are actively carried out showing good results in many fields. A template matching method is mainly used to recognize parts mounted on PCB(Printed Circuit Board). However, template matching should have multiple templates depending on the shape, orientation and brightness. And it takes long time to perform matching because it searches for the entire image. And there is also a disadvantage that the recognition rate is considerably low. In this paper, we use the Faster-RCNN method for recognizing PCB components as machine learning for classifying several objects in one image. This method performs better than the template matching method, execution time and recognition.

• Proceedings of the KSPS conference
• /
• 2005.11a
• /
• pp.215-218
• /
• 2005

This paper reports the current status of development of the Korean speech recognition platform (ECHOS). We implement new modules including ETSI feature extraction, backward search with trigram, and utterance verification. The ETSI feature extraction module is implemented by converting the public software to an object-oriented program. We show that trigram language modeling in the backward search pass reduces the word error rate from 23.5% to 22% on a large vocabulary continuous speech recognition task. We confirm the utterance verification module by examining word graphs with confidence score.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.11 no.1
• /
• pp.102-107
• /
• 2007

In this paper presents a gesture based driven approach for video editing. Given a lecture video, we adopt novel approaches to automatically detect and synchronize its content with electronic slides. The gestures in each synchronized topic (or shot) are then tracked and recognized continuously. By registering shots and slides md recovering their transformation, the regions where the gestures take place can be known. Based on the recognized gestures and their registered positions, the information in slides can be seamlessly extracted not only to assist video editing, but also to enhance the quality of original lecture video. In experiment with two videos, the proposed system showd each gesture recognition rate 95.5%,96.4%.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.10a
• /
• pp.723-726
• /
• 2009

In this research, We simulated performances of error reduction algorithm for the speech signal based on the microphone array-based beamforming method in speech recognition system and analyzed its performance. Also, we processed speech signal adopted from microphone array and maximum signal to noise ratio for each channel, and then compared them with signal to noise ratio of speech signal. Speech recognition rate is improved from 54.2% to 61.4% in case 1 and is improved from 41.2% to 50.5% in case 2 of the lower signal to noise ratio. Therefore the average reduction rates are showed 15.7% in case 1.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.14 no.1
• /
• pp.251-256
• /
• 2019

This paper presents an algorithm for recognizing and tracking a vehicle at a distance using a monocular camera installed at the center of the windshield of a vehicle to operate an autonomous vehicle in a racing. The vehicle is detected using the Haar feature, and the size and position of the vehicle are determined by detecting the shadows at the bottom of the vehicle. The region around the recognized vehicle is determined as ROI (Region Of Interest) and the vehicle shadow within the ROI is found and tracked in the next frame. Then the position, relative speed and direction of the vehicle are predicted. Experimental results show that the vehicle is recognized with a recognition rate of over 90% at a distance of more than 100 meters.

• The Journal of the Acoustical Society of Korea
• /
• v.38 no.1
• /
• pp.47-50
• /
• 2019

This paper proposes an acoustic model adaptation method for effective speech recognition in noisy environments. In the proposed algorithm, the noise corruption function is estimated employing DNN (Deep Neural Network), and the function is applied to the model parameter estimation. The experimental results using the Aurora 2.0 framework and database demonstrate that the proposed model adaptation method shows more effective in known and unknown noisy environments compared to the conventional methods. In particular, the experiments of the unknown environments show 15.87 % of relative improvement in the average of WER (Word Error Rate).

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.10a
• /
• pp.721-724
• /
• 2014

Ultrasonic sensor is used widely for many applications because low cost, simple structure, and low restriction. There are many difficulties to recognize an object by use an ultrasonic sensor, because of low resolution, poor direction, and measurement error. To improve the these problem, we use the various kinds of sensor arrangement methods, large amount of sensor, and change the arrangement pattern of sensor. In this paper, to obtain the most basic parameters for pattern recognition such as distance, dimension of the object, an angle of the object, we get the improved results by use the intelligent calculation algorithm based on Neuro-Fuzzy. This method use the multifarious output voltage of ultrasonic sensor by simple electronic circuit.