• International Journal of Advanced Culture Technology
• /
• v.10 no.1
• /
• pp.230-235
• /
• 2022

This study is to select an optimal object detection algorithm for designing a self-checkout counter to improve the inconvenience of payment systems for products without existing barcodes. To this end, a performance comparison analysis of YOLO v2, Tiny YOLO v2, and the latest YOLO v5 among deep learning-based object detection algorithms was performed to derive results. In this paper, performance comparison was conducted by forming learning data as an example of 'donut' in a bakery store, and the performance result of YOLO v5 was the highest at 96.9% of mAP. Therefore, YOLO v5 was selected as the artificial intelligence object detection algorithm to be applied in this paper. As a result of performance analysis, when the optimal threshold was set for each donut, the precision and reproduction rate of all donuts exceeded 0.85, and the majority of donuts showed excellent recognition performance of 0.90 or more. We expect that the results of this paper will be helpful as the fundamental data for the development of an automatic payment system using AI self-service technology that is highly usable in the non-face-to-face era.

• Journal of Satellite, Information and Communications
• /
• v.8 no.3
• /
• pp.6-9
• /
• 2013

Cognitive Radio is an intelligent interference avoidance idle spectrum communication system for your environment and actively determine the frequency bandwidth recycling by sharing the spectrum in a way that maximizes the efficiency of radio resource technology. At this time, this does not cause interference to the PU spectrum sensing technology is important. If you can choose, depending on the state of the channel spectrum sensing algorithm will be more efficient sensing. Matched filter detection system model that reflects the individual sensing through a fuzzy controller in this paper, energy detection and self-correlation detection are proposed and analyzed.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.25 no.9
• /
• pp.727-731
• /
• 2012

This paper dealt with the measurement and analysis of electromagnetic waves radiated from a partial discharge (PD) source in insulation oil to apply condition monitoring of oil-immersed transformers. Two types of narrow-band monopole antennas with the resonant frequency of a 500 MHz and a 1 GHz were designed and fabricated. Also, a needle-plane electrode system was manufactured to simulate PDs and the curvature radius of the needle is 10 ${\mu}m$ and the diameter of the plane is 60 mm. Electromagnetic wave was measured by the PD measurement system with the monopole antennas. Detection sensitivity of the fabricated antenna was compared for the same PD magnitude; 620 $mV_{peak}$ for the 500 MHz antenna and 960 $mV_{peak}$ for the 1 GHz antenna to the PD magnitude of 74 pC. Consequently, the 1 GHz monopole antenna is more effective to detect PDs in oil-immersed transformers.

• Journal of the Optical Society of Korea
• /
• v.8 no.4
• /
• pp.174-181
• /
• 2004

Various kinds of systems, that can do target recognition and position detection simultaneously by using infrared sensing detectors, have been developed. In this paper, the detection system TRSS (Thermal target Recognition by Spiral Scanning) adopts linear array shaped uncooled IR detector and uses spiral type fast scanning method for relative position detection of target objects, which radiate an IR region wavelength spectrum. It can detect thermal energy radiating from a 9 m-size target object as far as 200 m distance. And the maximum field of a detector is fully filled with the same size of target object at the minimum approaching distance 50 m. We investigate two types of lens systems. One is a singlet lens and the other is a doublet lens system. Every system includes one aspheric surface and free positioned aperture stop. Many designs of F/1.5 system with ${\pm}5.2^{\circ}$ field at the Efl=20, 30 mm conditions for single element and double elements lens system respectively are compared in their resolution performance [MTF] according to the aspheric surface and stop position changing on their optimization process. Optimum design is established including mechanical boundary conditions and manufacturing considerations.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.22 no.3
• /
• pp.222-229
• /
• 2009

A noble infrared sensor was studied for passenger conditional detection in vehicle, This research relates to uncooled infrared sensors for detecting the presence, type and temperature of occupants in vehicle. It sense that the occupants purpose to control the smart airbag for safety in the case of adult or child and to control the automatic air conditioning for convenience. This paper described the design and the fabrication of microbolometers which were composed of 2 by 8 elements using the surface micromachining technology. The characteristics of the array were investigated in the spectral region of $8{\sim}12{\mu}m$. The fabricated detectors exhibited the thermal mass of $7.05{\times}10^{-9}\;J/K$, the thermal conductance of $1.03{\times}10^{-6}\;W/K$, the thermal time constant of 6.8 ms, the responsivity of $2.96{\times}10^4\;V/W$ and the detectivity of $1.01{\times}10^9\;cmHz^{1/2}/W$, at the chopper frequency of 10 Hz and the bias current of $4.4{\mu}A$. We could successfully detect the human body condition in the divided zone. As a results, we concluded that microbolometer optimized in this research could be useful for the application of passenger conditional detection in vehicle.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.34 no.1
• /
• pp.44-52
• /
• 2014

Leakage is an important factor to be considered for the management of underground water supply pipelines in a smart water grid system, especially if the pipelines are aged and buried under the pavement or various structures of a highly populated city. Because the exact detection of the location of such leaks in pipelines is essential for their efficient operation, a new methodology for leak location detection based on frequency band variation, windowing filters, and probability is proposed in this paper. Because the exact detection of the leak location depends on the precision of estimation of time delay between sensor signals due to leak noise, some window functions that offer weightings at significant frequencies are applied for calculating the improved cross-correlation function. Experimental results obtained by applying this methodology to an actual buried water supply pipeline, ~ 253.9 m long and made of cast iron, revealed that the approach of frequency band variation with those windows and probability offers better performance for leak location detection.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.12
• /
• pp.77-84
• /
• 2022

In this paper, we propose a Label Reassignment Strategy to improve the performance of an object detection algorithm. Our approach involves two stages: an inference stage and an assignment stage. In the inference stage, we perform multi-scale inference with predefined scale sizes on a trained model and re-infer masked images to obtain robust classification results. In the assignment stage, we calculate the IoU between bounding boxes to remove duplicates. We also check box and class occurrence between the detection result and annotation label to re-assign the dominant class type. We trained the YOLOX-L model with the re-annotated dataset to validate our strategy. The model achieved a 3.9% improvement in mAP and 3x better performance on AP_S compared to the model trained with the original dataset. Our results demonstrate that the proposed Label Reassignment Strategy can effectively improve the performance of an object detection model.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.05a
• /
• pp.9-11
• /
• 2022

Community detection is a hot topic in social network analysis, and many existing studies use graph theory analysis methods to detect communities. This paper focuses on detecting absolute fair maximal balanced cliques in signed attributed social networks, which can satisfy ensuring the fairness of complex networks and break the bottleneck of the "information cocoon".

• Proceedings of the KIEE Conference
• /
• 1997.07c
• /
• pp.975-978
• /
• 1997

This Paper presents a high impedance fault (HIF) detection algorithm of distribution systems using wavelet transform. Two HIFs on dry soil and sandy soil were simulated on various load conditions in 22.9 kV distribution systems using EMTP, and the current wavelets were decomposed by wavelet transform. The current root mean square(rms) change, the index change rate and the relative amplitude change were used as the multi-criteria for a HIF detection. The index change rate and the relative amplitude were made using the wavelet coefficients.

• Proceedings of the KIEE Conference
• /
• 2004.05b
• /
• pp.97-100
• /
• 2004

HFPD(High Frequency Partial Discharge) measurement has been widely performed to diagnosis of cable system in the world although Korea just begin. The efficiency of measurement has been reported and proved by many researchers who are engaged in the diagnosis of high power apparatus. LG Cable has investigated the technology of diagnosis for cable system by means of PD measurement for 10 years. In this paper, PD measurement with portable device was performed to evaluate the quality of the 154kV transmission line and the 22.9kV XLPE cable line. Test results are shown that the HFPD detection technology is very attractive for the diagnosis of power cable system by high detection sensitivity at on-site.