• KIPS Transactions on Software and Data Engineering
• /
• v.4 no.10
• /
• pp.473-478
• /
• 2015

Recently, automatic video monitoring of individual pigs is emerging as an important issue in the management of group-housed pigs. Although a rich variety of studies have been reported on video monitoring techniques in intensive pig farming, it still requires further elaboration. In particular, when there exist adjoining pigs in a crowd pig room, it is necessary to have a way of separating adjoining pigs from the perspective of an image processing technique. In this paper, we propose an efficient image segmentation solution using both spatio-temporal information and region growing method for the identification of individual pigs in video surveillance systems. The experimental results with the videos obtained from a pig farm located in Sejong illustrated the efficiency of the proposed method.

• KIPS Transactions on Software and Data Engineering
• /
• v.2 no.3
• /
• pp.195-198
• /
• 2013

Many libraries utilize RFID tags for checking in and out of books. However, the recognition rate of this automatic process may depend on the orientation of antennas and RFID tags. Therefore we need supplemental systems to improve the recognition rate. The proposed algorithm sets up the ROI of the book existing area from the input image and then performs Canny edge detection algorithm to extract edges of books. Finally Hough line transform algorithm allows to detect the number of books from the extracted edges. To evaluate the performance of the proposed method, we applied our method to 350 book images under various circumstances. We then analyzed the performance of proposed method from results using recognition and mismatch ratio. The experimental result gave us 97.1% accuracy in book counting.

• The KIPS Transactions:PartB
• /
• v.10B no.7
• /
• pp.815-822
• /
• 2003

Fingerprint image enhancement and minutiae matching are two key steps in an automatic fingerprint identification system. In this paper, we propose a fingerprint recognition technique by using minutiae linking information. Recognition process have three steps ; preprocessing, minutiae extraction, matching step based on minutiae pairing. After extracting minutiae of a fingerprint from its thinned image for accuracy, we introduce matching process using minutiae linking information. Introduction of linking information into the minutiae matching process is a simple but accurate way, which solves the problem of reference minutiae pair selection with low cost in comparison stage of two fingerprints. This algorithm is invariable to translation and rotation of fingerprint. The matching algorithm was tested on 500 images from the semiconductor chip style scanner, experimental result revealed the false acceptance rate is decreased and genuine acceptance rate is increased than existing method.

• Nuclear Engineering and Technology
• /
• v.53 no.11
• /
• pp.3790-3797
• /
• 2021

Although the individual channel readout method can improve the performance of PET detectors with pixelated photo-sensors, such as silicon photomultiplier (SiPM), this method leads to a significant increase in the number of readout channels. In this study, we proposed a novel multiplexing method that could effectively reduce the number of readout channels to reduce system complexity and development cost. The proposed multiplexing circuit was designed to generate bipolar pulses with different zero-crossing points by adjusting the time constant of the high-pass filter connected to each channel of a pixelated photo-sensor. The channel position of the detected gamma-ray was identified by estimating the width between the rising edge and the zero-crossing point of the bipolar pulse. In order to evaluate the performance of the proposed multiplexing circuit, four detector blocks, each consisting of a 4 × 4 array of 3 mm × 3 mm × 20 mm LYSO and a 4 × 4 SiPM array, were constructed. The average energy resolution was 13.2 ± 1.1% for all 64 crystal pixels and each pixel position was accurately identified. A coincidence timing resolution was 580 ± 12 ps. The experimental results indicated that the novel multiplexing method proposed in this study is able to effectively reduce the number of readout channels while maintaining accurate position identification with good energy and timing performance. In addition, it could be useful for the development of PET systems consisting of a large number of pixelated detectors.

• International Journal of Computer Science & Network Security
• /
• v.23 no.1
• /
• pp.147-152
• /
• 2023

Mostly in motor fault detection the instantaneous values 3 axis vibration and 3phase current in time domain are acquired and converted to frequency domain. Vibrations are more useful in diagnosing the mechanical faults and motor current has remained more useful in electrical fault diagnosis. With having some experience and knowledge on the behavior of acquired data the electrical and mechanical faults are diagnosed through signal processing techniques or combine machine learning and signal processing techniques. In this paper, a single-layer LSTM based condition monitoring system is proposed in which the instantaneous values of three phased motor current are firstly acquired in simulated motor in in health and supply imbalance conditions in each of three stator currents. The acquired three phase current in time domain is then used to train a LSTM network, which can identify the type of fault in electrical supply of motor and phase in which the fault has occurred. Experimental results shows that the proposed single layer LSTM algorithm can identify the electrical supply faults and phase of fault with an average accuracy of 88% based on the three phase stator current as raw data without any processing or feature extraction.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.19 no.1
• /
• pp.309-316
• /
• 2024

Facial recognition technology is widely used in various fields but faces challenges due to its vulnerability to fraudulent activities such as photo spoofing. Extensive research has been conducted to overcome this challenge. Most of them, however, require the use of specialized equipment like multi-modal cameras or operation in high-performance environments. In this paper, we introduce LH-FAS v2 (: Lightweight Head-pose-based Face Anti-Spoofing v2), a system designed to operate on a commercial webcam without any specialized equipment, to address the issue of facial recognition spoofing. LH-FAS v2 utilizes FSA-Net for head pose estimation and ArcFace for facial recognition, effectively assessing changes in head pose and verifying facial identity. We developed the VD4PS dataset, incorporating photo spoofing scenarios to evaluate the model's performance. The experimental results show the model's balanced accuracy and speed, indicating that head pose estimation-based facial anti-spoofing technology can be effectively used to counteract photo spoofing.

• Journal of Korean Society of Forest Science
• /
• v.76 no.3
• /
• pp.243-248
• /
• 1987

The U.S. Forest Service administers the world's largest forestry research organization. From its modest beginning in 1876, some 30 years before the United States national forest system was established, the research branch has devoted its effort to meet current and future information needs of the forestry community of the United States, not just for the U.S. Forest Service. The research branch is one of three major administrative units of the U.S. Forest Service. The others being the National Forest System and State and Private Forestry. Currently the National Forest System comprises 155 national forests, 19 national grasslands, and 18 utilization projects located in 44 states. Puerto Rico, and the Virgin Islands. The National Forest System manages these areas for a large array of uses and benefits including timber, water, forage, wildlife, recreation, minerals, and wilderness. It is through the State and Private Forestry branch that the U.S. Forest Service cooperates and coordinates forestry activities and programs with state and local governments, forest industries, and private landowners. These activities include financial and technical assistance in disease, insect, and fire protection ; plan forestry programs ; improve harvesting and marketing practices ; and transfer forestry research results to user groups. Forestry research is carried out through eight regional Forest Experiment Stations and the Forest Product Laboratory. Studies are maintained at 70 administrative sites, and at 115 experimental forest and grasslands. All of the current sciences that composed modern forestry are included in the research program. These range from forest biology (i. e. silviculture, ecology, physiology, and genetics) to the physical, mathematical, engineering, managerial, and social sciences. The levels of research range from application, developmental, and basic research. Research planning and priority identification is an ongoing process with elements of the research program changing to meet short-term critical information needs(i. e. protection research) to long-term opportunities(i. e. biotechnology). Research planning and priority setting is done in cooperation with National Forest Systems, forest industries, universities, and individual groups such as environmental, wilderness, or wildlife organizations. There is an ongoing review process of research administration, organization, and science content to maintain quality of research. In the U.S. Forest Service the research responsibility is not completed until the new information is being applied by the various user group : I. e. technology transfer program. Research planning and development in the U.S. Forest Service is a dynamic activity. Porgrams for the year 2000 and beyond are now in the planning stage.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.28 no.6A
• /
• pp.799-808
• /
• 2008

The analysis of vehicular loads reflecting the domestic traffic circumstances is necessary for the development of adequate design live load models in the analysis and design of cable-supported bridges or the development of fatigue load models to predict the remaining lifespan of the bridges. This study intends to develop an ANN(artificial neural network)-based Bridge WIM system and Influence line-based Bridge WIM system for obtaining information concerning the loads conditions of vehicles crossing bridge structures by exploiting the signals measured by strain gauges installed at the bottom surface of the bridge superstructure. This study relies on experimental data corresponding to the travelling of hundreds of random vehicles rather than on theoretical data generated through numerical simulations to secure data sets for the training and test of the ANN. In addition, data acquired from 3 types of vehicles weighed statically at measurement station and then crossing the bridge repeatedly are also exploited to examine the accuracy of the trained ANN. The results obtained through the proposed ANN-based analysis method, the influence line analysis method considering the local behavior of the bridge are compared for an example cable-stayed bridge. In view of the results related to the cable-stayed bridge, the cross beam ANN analysis method appears to provide more remarkable load analysis results than the cross beam influence line method.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.31 no.4
• /
• pp.351-359
• /
• 2011

In a structure, damage can occur at several scales from micro-cracking to corrosion or loose bolts. This makes the identification of damage difficult with one mode of sensing. Hence, a multi-mode actuated sensing system is proposed based on a self-sensing circuit using a piezoelectric sensor. In the self sensing-based multi-mode actuated sensing, one mode provides a wide frequency-band structural response from the self-sensed impedance measurement and the other mode provides a specific frequency-induced structural wavelet response from the self-sensed guided wave measurement. In this study, an experimental study on the pipeline system is carried out to verify the effectiveness and the robustness of the proposed structural health monitoring approach. Different types of structural damage are artificially inflicted on the pipeline system. To classify the multiple types of structural damage, a supervised learning-based statistical pattern recognition is implemented by composing a two-dimensional space using the damage indices extracted from the impedance and guided wave features. For more systematic damage classification, several control parameters to determine an optimal decision boundary for the supervised learning-based pattern recognition are optimized. Finally, further research issues will be discussed for real-world implementation of the proposed approach.

• Clinical and Experimental Reproductive Medicine
• /
• v.29 no.2
• /
• pp.117-127
• /
• 2002

Objective: This study was to investigate the generation of the functional neuron derived from human embryonic stem (hES, MB03) cells on in vitro neural cell differentiation system. Methods: For neural progenitor cell formation derived from hES cells, we produced embryoid bodies (EB: for 5 days, without mitogen) from hES cells and then neurospheres (for $7{\sim}10$ days, 20 ng/ml of bFGF added N2 medium) from EB. And then finally for the differentiation into mature neuron, neural progenitor cells were cultured in i) N2 medium only (without bFGF), ii) N2 supplemented with 20 ng/ml platelet derived growth factor-bb (PDGF-bb) or iii) N2 supplemented with 5 ng/ml brain derived neurotrophic factor (BDNF) for 2 weeks. Identification of neural cell differentiation was carried out by immunocytochemistry using $\beta_{III}$-tubulin (1:250), MAP-2 (1:100) and GFAP (1:500). Also, generation of functional neuron was identified using anti-glutamate (Sigma, 1:1000), anti-GABA (Sigma, 1:1000), anti-serotonin (Sigma, 1:1000) and anti-tyrosine hydroxylase (Sigma, 1:1000). Results: In vitro neural cell differentiation, neurotrophic factors (PDGF and BDNF) treated cell groups were high expressed MAP-2 and GFAP than non-treated cell group. The highest expression pattern of MAP-2 and $\beta_{III}$-tubulin was indicated in BDNF treated group. Also, in the presence of PDGF-bb or BDNF, most of the neural cells derived from hES cells were differentiated into glutamate and GABA neuron in vitro. Furthermore, we confirmed that there were a few serotonin and tyrosine hydroxylase positive neuron in the same culture environment. Conclusion: This results suggested that the generation of functional neuron derived from hES cells was increased by addition of neurotrophic factors such as PDGF-bb or BDNF in b-FGF induced neural cell differentiation system and especially glutamate and GABA neurons were mainly produced in the system.