• Journal of The Korean Astronomical Society
• /
• v.47 no.2
• /
• pp.43-47
• /
• 2014

Many advanced ground-based solar telescopes improve the spatial resolution of observation images using an adaptive optics (AO) system. As any AO correction remains only partial, it is necessary to use post-processing image reconstruction techniques such as speckle masking or shift-and-add (SAA) to reconstruct a high-spatial-resolution image from atmospherically degraded solar images. In the New Vacuum Solar Telescope (NVST), the spatial resolution in solar images is improved by frame selection and SAA. In order to overcome the burden of massive speckle data processing, we investigate the possibility of using the speckle reconstruction program in a real-time application at the telescope site. The code has been written in the C programming language and optimized for parallel processing in a multi-processor environment. We analyze the scalability of the code to identify possible bottlenecks, and we conclude that the presented code is capable of being run in real-time reconstruction applications at NVST and future large aperture solar telescopes if care is taken that the multi-processor environment has low latencies between the computation nodes.

• Journal of the Korean Society for Precision Engineering
• /
• v.20 no.7
• /
• pp.120-127
• /
• 2003

A novel three degrees of freedom sensing method utilizing binary grid pattern image and vision camera is presented. The binary grid pattern image is designed by Pseudo-Random Binary Arrays and referenced to encode in-plane position of a moving stage of the contact-free planar actuator. First, the yaw motion of the stage is detected using fast image processing and then the other planar positions, x and y, are decoded with a sequence of images. This method can be applied to the system that needs feedback of in-plane position, with advantages of a good accuracy and high resolution comparable with the encoder, a relatively compact structure, no friction, and a low cost. In this paper, all the procedures of the above sensing mechanism are described in detail, including simulation and experiment results.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.9 no.2 s.25
• /
• pp.111-117
• /
• 2006

This paper describes the design principles and methods of electronic processor for thermal imager with 320$\times$240 staring array infrared detector. For the detector's nonuniformity correction and excellent image quality, we have designed the multi-point correction method using the defocusing technique of the optics. And to enhance the image of low contrast and improve the detection capability, the new technique of histogram processing has been designed. Through these image processing techniques, we have developed the high quality thermal imager and acquired a satisfactory thermal image. The result of MRTD(Minimum Resolvable Temperature Difference) is $0.1^{\circ}C$ at 4cycles/mard.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.49 no.12
• /
• pp.131-141
• /
• 2012

This paper proposes a method of analyzing unrecognizable numbers of license plate images, which are degraded by various factors such as low resolution, low light level, geometric distortion, and periodic noise, to name a few. With existing vehicle license plate recognition methods, it is difficult to recognize license plate if images are not recognizable in the pre-process of removing degradation factors. Although images of license plate have not been improved to be recognizable in the pre-process, the proposed method makes it possible to recognize numbers of license by distorting pre-saved reference images of license plate numbers same as sample plates, and by assuming likelihood ratio using statistical methods. The proposed method also makes it possible to identify suspect vehicle license plate under unstable light conditions and with low resolution images that are unrecognizable by the naked eye. This method has been used in real criminal investigation to recognize numbers of license plate of criminal vehicle, and has proved to be useful as criminal evidence through experiments under various conditions.

• Journal of Radiation Industry
• /
• v.11 no.3
• /
• pp.131-137
• /
• 2017

To compare the radiation dose and image noise of low dose computed tomography (CT) and high resolution CT using the fixed tube current technique and automatic tube current modulation (CARE Dose 4D). Chest CT and human anthropomorphic phantom were used the RPL (radiophotoluminescence) dosimeters. For image evaluation, standard deviation of mean CT attenuation coefficient and CT attenuation coefficient was measured using ROI analysis function. The effective dose was calculated using CTDIvol and DLP. CARE Dose 4D was reduced by 74.7% and HRCT by 64.4% compared to the fixed tube current technique in low dose CT of chest phantom. In CTDIvol and DLP, the dose of CARE Dose 4D was reduced by fixed tube current technique. For effective dose, CARE Dose 4D was reduced by 47% and HRCT by 46.9% compared to the fixed tube current method, and the dose of CARE Dose 4D was significantly different (p<.05). Noise in the image was higher than that in the fixed tube current technique. Noise difference in the image of CARE Dose 4D in low dose CT was significant (p<.05). The low radiation dose and the noise difference of the CARE Dose 4D were compared with the fixed tube current technique in low dose CT and HRCT using chest phantom. The radiation doses using CARE Dose 4D were in accordance with the national and international dose standards. CARE Dose 4D should be applied to low dose CT and HRCT for clinical examination.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.16 no.1
• /
• pp.41-49
• /
• 2023

A convolutional neural network (CNN) is a representative algorithm for implementing artificial neural networks. CNNs have improved on the issues of rapid increase in calculation amount and low object classification rates, which are associated with a conventional multi-layered fully-connected neural network (FNN). However, because of the rapid development of IT devices, the maximum resolution of images captured by current smartphone and tablet cameras has reached 108 million pixels (MP). Specifically, a traditional CNN algorithm requires a significant cost and time to learn and process simple, high-resolution images. Therefore, this study proposes an improved CNN algorithm for implementing an object classification learning model for simple, high-resolution images. The proposed method alters the adjacency matrix value of the pooling layer's max pooling operation for the CNN algorithm to reduce the high-resolution image learning model's creation time. This study implemented a learning model capable of processing 4, 8, and 12 MP high-resolution images for each altered matrix value. The performance evaluation result showed that the creation time of the learning model implemented with the proposed algorithm decreased by 36.26% for 12 MP images. Compared to the conventional model, the proposed learning model's object recognition accuracy and loss rate were less than 1%, which is within the acceptable error range. Practical verification is necessary through future studies by implementing a learning model with more varied image types and a larger amount of image data than those used in this study.

• Journal of information and communication convergence engineering
• /
• v.15 no.3
• /
• pp.193-198
• /
• 2017

In this study, we propose a three-dimensional (3D) visualization technique of occluded objects using integral imaging with a plenoptic camera. In previous studies, depth map estimation from elemental images was used to remove occlusion. However, the resolution of these depth maps is low. Thus, the occlusion removal accuracy is not efficient. Therefore, we use a plenoptic camera to obtain a high-resolution depth map. Hence, individual depth map for each elemental image can also be generated. Finally, we can regenerate a more accurate depth map for 3D objects with these separate depth maps, allowing us to remove the occlusion layers more efficiently. We perform optical experiments to prove our proposed technique. Moreover, we use MSE and PSNR as a performance metric to evaluate the quality of the reconstructed image. In conclusion, we enhance the visual quality of the reconstructed image after removing the occlusion layers using the plenoptic camera.

• Proceedings of the Korean Society of Surveying, Geodesy, Photogrammetry, and Cartography Conference
• /
• 2010.04a
• /
• pp.277-280
• /
• 2010

There is a need to convert the old low- or medium-resolution satellite image-based thematic mapping to the high-resolution satellite image-based mapping of GSD 1m grade or lower. There is also a need to generate middle- or large-scale thematic maps of 1:5,000 or lower. In this study, the DEM and orthoimage is generated with the KOMPSAT-2 stereo image of Yuseong-gu, Daejeon Metropolitan City. By utilizing the orthoimage, automatic extraction experiments of land cover information are generated for buildings, roads and urban areas, raw land(agricultural land), mountains and forests, hydrosphere, grassland, and shadow. The experiment results show that it is possible to classify, in detail, for natural features such as the hydrosphere, mountains and forests, grassland, shadow, and raw land. While artificial features such as roads, buildings, and urban areas can be easily classified with automatic extraction, there are difficulties on detailed classifications along the boundaries. Further research should be performed on the automation methods using the conventional thematic maps and all sorts of geo-spatial information and mapping techniques in order to classify thematic information in detail.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.57 no.6
• /
• pp.91-97
• /
• 2015

The Unmanned Aerial Vehicles (UAVs) have several advantages over conventional RS techniques. They can acquire high-resolution images quickly and repeatedly. And with a comparatively lower flight altitude i.e. 80~400 m, they can obtain good quality images even in cloudy weather. Therefore, they are ideal for acquiring spatial data in cases of small agricultural field with mixed crop, abundant in South Korea. This paper discuss the use of low cost UAV based remote sensing for classifying crops. The study area, Gochang is produced by several crops such as red pepper, radish, Chinese cabbage, rubus coreanus, welsh onion, bean in South Korea. This study acquired images using fixed wing UAV on September 23, 2014. An object-based technique is used for classification of crops. The results showed that scale 250, shape 0.1, color 0.9, compactness 0.5 and smoothness 0.5 were the optimum parameter values in image segmentation. As a result, the kappa coefficient was 0.82 and the overall accuracy of classification was 85.0 %. The result of the present study validate our attempts for crop classification using high resolution UAV image as well as established the possibility of using such remote sensing techniques widely to resolve the difficulty of remote sensing data acquisition in agricultural sector.

• ETRI Journal
• /
• v.27 no.4
• /
• pp.385-393
• /
• 2005

This paper presents an application of digital signal processing to data acquired by the radio imaging method (RIM) that was adopted to measure moisture distribution inside the human body. RIM was originally developed for the mining industry; we are applying the method to a biomedical measurement because of its simplicity, economy, and safety. When a two-dimensional image was constructed from the measured data, the method provided insufficient resolution because the wavelength of the measurement medium, a weak electromagnetic wave in a VHF band, was longer than human tissues. We built and measured a phantom, a model simulating the human body, consisting of two water tanks representing large internal organs. A digital equalizer was applied to the measured values as a weight function, and images were reconstructed that corresponded to the original shape of the two water tanks. As a result, a two-dimensional image containing two individual peaks corresponding to the original two small water tanks was constructed. The result suggests the method was applicable to biomedical measurement by the assistance of digital signal processing. This technique may be applicable to home-based medical care and other situations in which safety, simplicity, and economy are important.