• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.3
• /
• pp.115-124
• /
• 2022

In this paper, YOLO and EAST models are tested to analyze their performance in text area detecting for real-world and normal text images. The earl ier YOLO models which include YOLOv3 have been known to underperform in detecting text areas for given images, but the recently released YOLOv4 and YOLOv5 achieved promising performances to detect text area included in various images. Experimental results show that both of YOLO v4 and v5 models are expected to be widely used for text detection in the filed of scene text recognition in the future.

• Journal of the Korean Society for Precision Engineering
• /
• v.28 no.6
• /
• pp.738-744
• /
• 2011

The reliability of blade root fixing section is required to endure the centrifugal force and vibration stress for the last stage blade of steam turbine in thermal power plant. Most of the domestic steam turbine last stage blades have finger type roots. The finger type blade is very complex, so the inspection had been performed only on the exposed fixing pin cross-section area due to the difficulty of inspection. But the centrifugal force and vibration stress are also applied at the blade root finger and the crack generates, so the inspection method for finger section is necessary. For the inspection of root finger, inspection points were decided by simulating ultra-sonic path with 3D modeling, curve-shape probe and fixing jig were invented, and the characteristics analysis method of ultrasonic reflection signal and defect signal disposition method were invented. This invented method was actually executed at site and prevented the blade liberation failure by detecting the cracks at the fingers. Also, the same type blades of the other turbines were inspected periodically and the reliability of the turbine increased.

• Journal of rehabilitation welfare engineering & assistive technology
• /
• v.10 no.2
• /
• pp.141-146
• /
• 2016

In this paper, we apply limited area mask operation and active contour model to accurately detect tongue area outline in tongue diagnosis system. To accurately analyze the properties of the tongue, first, the tongue area to be detected. Therefore an effective segmentation method for detecting the edge of tongue is very important. It experimented with tongue image DB consists of 20~30 students 30 people. Experiments on real tongue image show the good performance of this method. Experimental results show that the proposed method extracts object boundaries more accurately than existing methods without mask operation.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.2
• /
• pp.155-160
• /
• 2014

There are many image processing based algorithms and applications that fail when illumination change occurs. Therefore, the illumination change has to be detected then the illumination change occurred images need to be enhanced in order to keep the appropriate algorithm processing in a reality. In this paper, a new method for detecting illumination changes efficiently in a real time by using local region information and fuzzy logic is introduced. The effective way for detecting illumination changes in lighting area and the edge of the area was selected to analyze the mean and variance of the histogram of each area and to reflect the changing trends on previous frame's mean and variance for each area of the histogram. The ways are used as an input. The changes of mean and variance make different patterns w hen illumination change occurs. Fuzzy rules were defined based on the patterns of the input for detecting illumination changes. Proposed method was tested with different dataset through the evaluation metrics; in particular, the specificity, recall and precision showed high rates. An automatic parameter selection method was proposed for contrast limited adaptive histogram equalization method by using entropy of image through adaptive neural fuzzy inference system. The results showed that the contrast of images could be enhanced. The proposed algorithm is robust to detect global illumination change, and it is also computationally efficient in real applications.

• Korean Journal of Radiology
• /
• v.21 no.8
• /
• pp.978-986
• /
• 2020

Objective: To compare native and post-contrast T1 mapping with late gadolinium enhancement (LGE) imaging for detecting and measuring the microvascular obstruction (MVO) area in reperfused acute myocardial infarction (MI). Materials and Methods: This study included 20 patients with acute MI who had undergone 1.5T cardiovascular magnetic resonance imaging (CMR) after reperfusion therapy. CMR included cine imaging, LGE, and T1 mapping (modified look-locker inversion recovery). MI size was calculated from LGE by full-width at half-maximum technique. MVO was defined as an area with low signal intensity (LGE) or as a region of visually distinguishable T1 values (T1 maps) within infarcted myocardium. Regional T1 values were measured in MVO, infarcted, and remote myocardium on T1 maps. MVO area was measured on and compared among LGE, native, and post-contrast T1 maps. Results: The mean MI size was 27.1 ± 9.7% of the left ventricular mass. Of the 20 identified MVOs, 18 (90%) were detected on native T1 maps, while 10 (50%) were recognized on post-contrast T1 maps. The mean native T1 values of MVO, infarcted, and remote myocardium were 1013.5 ± 58.5, 1240.9 ± 55.8 (p < 0.001), and 1062.2 ± 55.8 ms (p = 0.169), respectively, while the mean post-contrast T1 values were 466.7 ± 26.8, 399.1 ± 21.3, and 585.2 ± 21.3 ms, respectively (p < 0.001). The mean MVO areas on LGE, native, and post-contrast T1 maps were 134.1 ± 81.2, 133.7 ± 80.4, and 117.1 ± 53.3 mm², respectively. The median (interquartile range) MVO areas on LGE, native, and post-contrast T1 maps were 128.0 (58.1-215.4), 110.5 (67.7-227.9), and 143.0 (76.7-155.3) mm², respectively (p = 0.002). Concordance correlation coefficients for the MVO area between LGE and native T1 maps, LGE and post-contrast T1 maps, and native and post-contrast T1 maps were 0.770, 0.375, and 0.565, respectively. Conclusion: MVO areas were accurately delineated on native T1 maps and showed high concordance with the areas measured on LGE. However, post-contrast T1 maps had low detection rates and underestimated MVO areas. Collectively, native T1 mapping is a useful tool for detecting MVO within the infarcted myocardium.

• International Journal of Advanced Culture Technology
• /
• v.4 no.1
• /
• pp.28-30
• /
• 2016

Due to the high increment in usage and built-in advanced technology of smartphones, human activity recognition relying on smartphone sensor data has become a focused research area. In order to reduce noise of collected data, most of previous studies assume that smartphones are fixed at certain positions. This strategy is impractical for real life applications. To overcome this issue, we here investigate a framework that allows detecting the status of a traveller as idle or moving regardless the position and the direction of smartphones. The application of our work is to estimate the total energy consumption of a traveller during a trip. A number of experiments have been carried out to show the effectiveness of our framework when travellers are not only walking but also using primitive vehicles like motorbikes.

• Journal of Advanced Marine Engineering and Technology
• /
• v.25 no.2
• /
• pp.338-344
• /
• 2001

According to the MARPOL 73/78 of Convention, all ships should have oil filtering equipment and 15 ppm bilge which satisfy Requirements of MARPOL 73/78. This study is concerned with designing and manufacturing a prototype Oil content Meter(OCM) used in machinery area of ship. The prototype OCM is composed of two parts which are oil content sensing module and data processing unit. The oil content sensing module consists of infra-red light source, photo-diode light receivers, and a glass tube for bilge water sample. The data processing unit has a micro-processor as hard core and peripheral devices. The experimental results of prototype OCM and analysis of collected data reveal linear property between transmitted light and scattered light as long as the bilge water has low level content of oil. And this linear property leads to a oil content detecting method which is programmed and loaded into the data processing unit. The performance of the prototype OCM is compared with that of the commercial OCM in the market.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.4
• /
• pp.72-80
• /
• 1996

In this paper, variable windowing mean filter to remove speckle noise and a measure to detect thin edge in ultrasonic images are proposed. Because ultrasonic images are corrupted by speckle noise showing a granular appearance, good edge detection is difficult. As a result, noise removing filter is needed in preprocessing stage. The speckle noise removing filter is based on mean filter whose window size is changed by the ratio of standard deviation to mean for image signal and noise signal in local area. And the measure expressed the difference of means between tow windows is used for detecting thin edge in filtered image. Results show that variable windowing mean filter removes speckle noise effectively, and proposed measure is useful in detecting thin edge.

• Proceedings of the KSME Conference
• /
• 2004.11a
• /
• pp.1231-1235
• /
• 2004

The needs of micro scale thermal detecting technique are increasing in biology and chemical industry. For example, Thermal finger print, Micro PCR(polymer chain reaction), ${\mu}TAS$ and so on. To satisfy these needs, we developed a DTSA(Diode Temperature Sensor Array) for detecting and controlling the temperature on small surface. The DTSA is fabricated by using VLSI technique. It consists of 32 ${\times}$ 32 array of diodes (1,024 diodes) for temperature detection and 8 heaters for temperature control on a 8mm ${\times}$ 8mm surface area. The working principle of temperature detection is that the forward voltage drop across a silicon diode is approximately proportional to the inverse of the absolute temperature of diode. And eight heaters ($1K{\Omega}$) made of poly-silicon are added onto a silicon wafer and controlled individually to maintain a uniform temperature distribution across the DTSA. Flip chip packaging used for easy connection of the DTSA. The circuitry for scanning and controlling DTSA are also developed

• Journal of Welding and Joining
• /
• v.21 no.1
• /
• pp.60-65
• /
• 2003

At present. welding of most pipes with large diameter is carried out by the manual process. Automation of the welding process is necessary f3r the sake of consistent weld quality and improvement in productivity. In this study, two vision sensors, based on the optical triangulation, were used to obtain the information for seam tracking and detecting the weld defects. Through utilization of the vision sensors, noises were removed, images and 3D information obtained and positions of the feature points detected. The aforementioned process provided the seam and leg position data, calculated the magnitude of the gap, fillet area and leg length and judged the weld defects by ISO 5817. Noises in the images were removed by using the gradient values of the laser stripe's coordinates and various feature points were detected by using an algorithm based on the iterative polygon approximation method. Since the process time is very important, all the aforementioned processes should be conducted during welding.