• Journal of the Korean Society of Industry Convergence
• /
• v.22 no.3
• /
• pp.293-301
• /
• 2019

Based on accurate image processing technology, a system for measuring displacement in ${\mu}m$ for drive error (position error, straightness error, flatness error) at a distance using parallel light and image sensor is developed, and a system for applying this technology development to a large 3D rapid prototyping machine and compensating in real time is developed to dramatically reduce the range of measurement error and enable intelligent 3D production of high quality products.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.15 no.3 s.96
• /
• pp.329-333
• /
• 2005

This paper is concerned with the measurement of low-frequency vibrations of structures using the image processing method. To measure the vibrations visually, the measurement system consists of a camera, an image grabber board, and a computer. The specific target installed on the structure is used to calculate the vibration of structure. The captured image is then converted into a pixel-based data and then analyzed numerically. The limitation of the system depends on the image capturing speed and the size of image. In this paper, we propose the methodology for the vibration measurement using the image processing method. The method enables us to measure the displacement directly without any contact. The current resolution of the vibration measurement is limited to sub centimeter scale. However, the frequency bandwidth and resolution can be enhanced by a high-speed and high-resolution image processing system.

• Journal of information and communication convergence engineering
• /
• v.10 no.2
• /
• pp.194-199
• /
• 2012

This paper proposes a novel scheme of a gray scale fingerprint image for a high-accuracy capacitive sensor chip. The conventional grayscale image scheme uses a digital-to-analog converter (DAC) of a large-scale layout or charge-pump circuit with high power consumption and complexity by a global clock signal. A modified capacitive detection circuit for the charge sharing scheme is proposed, which uses a down literal circuit (DLC) with a floating-gate metal-oxide semiconductor transistor (FGMOS) based on a neuron model. The detection circuit is designed and simulated in a 3.3 V, 0.35 ${\mu}m$ standard CMOS process. Because the proposed circuit does not need a comparator and peripheral circuits, the pixel layout size can be reduced and the image resolution can be improved.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2002.05b
• /
• pp.982-988
• /
• 2002

• Korean Journal of Geomatics
• /
• v.1 no.1
• /
• pp.7-13
• /
• 2001

Geospatial Information Systems (GIS) have been employed to systematically manage and design land use in urban areas. This has increased the need for more accurate vector and raster data. In Korea, l/l,000-scale digital maps are used as vector data for the facility management in urban areas. This has increased the need for large scale orthoimages. Orthoimages generated from aerial imagery can provide accurate information, making possible the more effective city management. However, there is a large problem in using the orthoimages, i.e., currently available conventional orthoimages have not been generated based on Digital Elevation Model (DEM) that takes into account the building heights. So this causes the displacements of building image in large scale orthoimages. The present study is an attempt to generate the large scale orthoimages based on building DEM. The semiautomatic building extraction method can detect building outlines by mouse clicking on either building roofs or corners. Building DEM, based on the outline and calculated building height, was used to produce the large scale orthoimages with the corrected building occlusion.

• Korean Journal of Remote Sensing
• /
• v.36 no.6_1
• /
• pp.1393-1405
• /
• 2020

UAV (Unmanned Aerial Vehicle) can acquire high-resolution images due to low-altitude flight, and it can be photographed at any time. Therefore, the UAV images can be updated at any time in map production. Due to these advantages, studies on the possibility of producing large-scale digital maps using UAV images are actively being conducted. Precise digital maps can be used as base data for digital twins or smart cites. For producing a precise digital map, precise sensor modeling using GCPs (Ground Control Points) must be preceded. In this study, geometric models of UAV images were established through a precision sensor modeling algorithm developed in house. Then, a digital map by stereo plotting was produced to evaluate the possibility of large-scale digital map. For this study, images and GCPs were acquired for Ganseok-dong, Incheon and Yeouido, Seoul. As a result of precision sensor modeling accuracy analysis, high accuracy was confirmed within 3 pixels of the average error of the checkpoints and 4 pixels of the RMSE was confirmed for the two study regions. As a result of the mapping accuracy analysis, it satisfied the 1:1,000 mapping accuracy announced by the NGII (National Geographic information Institute). Therefore, the precision sensor modeling technology suggested the possibility of producing a 1:1,000 large-scale digital map by UAV images.

• Journal of Korean Medical Science
• /
• v.33 no.43
• /
• pp.239.1-239.12
• /
• 2018

Background: We described a novel multi-step retinal fundus image reading system for providing high-quality large data for machine learning algorithms, and assessed the grader variability in the large-scale dataset generated with this system. Methods: A 5-step retinal fundus image reading tool was developed that rates image quality, presence of abnormality, findings with location information, diagnoses, and clinical significance. Each image was evaluated by 3 different graders. Agreements among graders for each decision were evaluated. Results: The 234,242 readings of 79,458 images were collected from 55 licensed ophthalmologists during 6 months. The 34,364 images were graded as abnormal by at-least one rater. Of these, all three raters agreed in 46.6% in abnormality, while 69.9% of the images were rated as abnormal by two or more raters. Agreement rate of at-least two raters on a certain finding was 26.7%-65.2%, and complete agreement rate of all-three raters was 5.7%-43.3%. As for diagnoses, agreement of at-least two raters was 35.6%-65.6%, and complete agreement rate was 11.0%-40.0%. Agreement of findings and diagnoses were higher when restricted to images with prior complete agreement on abnormality. Retinal/glaucoma specialists showed higher agreements on findings and diagnoses of their corresponding subspecialties. Conclusion: This novel reading tool for retinal fundus images generated a large-scale dataset with high level of information, which can be utilized in future development of machine learning-based algorithms for automated identification of abnormal conditions and clinical decision supporting system. These results emphasize the importance of addressing grader variability in algorithm developments.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.14 no.3
• /
• pp.142-149
• /
• 2006

A PIV(Particle Image Velocimetry) was applied to measure in-cylinder velocity field according to inlet valve angle during intake stroke. Two engines, one is conventional DOHC 4 valve and the other is narrow valve angle, were used to compare real intake flow. The results show that the intake flow pattern of conventional engine is more complicated than that of narrow angle one in horizontal plane and the vertical component of in-cylinder flow is rapidly decayed at the end stage of intake. On the other hand, the flow pattern of narrow angle one is relatively well arranged in horizontal plane and the vertical velocity component remains so strongly that forms large-scale strong tumble. Two engines also form commonly three tumble; two are small and bellow the intake valve and one is large-scale. The center of large scale tumble moves to bottom of cylinder as the vertical velocity increases.

• Journal of Information Processing Systems
• /
• v.20 no.2
• /
• pp.149-158
• /
• 2024

Traditional manual identification of crop leaf diseases is challenging. Owing to the limitations in manpower and resources, it is challenging to explore crop diseases on a large scale. The emergence of artificial intelligence technologies, particularly the extensive application of deep learning technologies, is expected to overcome these challenges and greatly improve the accuracy and efficiency of crop disease identification. Crop leaf disease identification models have been designed and trained using large-scale training data, enabling them to predict different categories of diseases from unlabeled crop leaves. However, these models, which possess strong feature representation capabilities, require substantial training data, and there is often a shortage of such datasets in practical farming scenarios. To address this issue and improve the feature learning abilities of models, this study proposes a deep transfer learning adaptation strategy. The novel proposed method aims to transfer the weights and parameters from pre-trained models in similar large-scale training datasets, such as ImageNet. ImageNet pre-trained weights are adopted and fine-tuned with the features of crop leaf diseases to improve prediction ability. In this study, we collected 16,060 crop leaf disease images, spanning 12 categories, for training. The experimental results demonstrate that an impressive accuracy of 98% is achieved using the proposed method on the transferred ResNet-50 model, thereby confirming the effectiveness of our transfer learning approach.

• Proceedings of the KSRS Conference
• /
• v.2
• /
• pp.631-634
• /
• 2006

Three main important sources for establishing GIS are the orthomap in scale 1:5 000 with Ground Sampling Distance of 0,5m; DEM/DTM data with height error of ${\pm}$1,0m and topographic map in scale 1: 10 000. The new era with Very High Resolution Satellite (VHRS) images as IKONOS, QuickBird, EROS, OrbView and other ones having Ground Sampling Distance (GSD) even lower than 1m has been in potential for producing orthomap in large scale 1:5 000, to update existing maps, to compile general-purpose or thematic maps and for GIS. The accuracy of orthomap generated from VHRS image affects strongly on GIS reliability. Nevertheless, orthomap accuracy taken from VHRS image is at first dependent on chosen sensor geometrical models. This paper presents, at fist, theoretical basic of the Rational Polynomial Coefficient (RPC) model installed in the commercial ImageStation Systems, realized for orthorectifying VHRS images. The RPC model of VHRS image is a replacement camera mode that represents the indirect relation between terrain and its image acquired on the flight orbit. At the end of this paper the practical accuracies of IKONOS and QuickBird image orthorectified by RPC model on Canadian PCI Geomatica System have been presented. They are important indication for practical application of producing digital orthomaps.