• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.12
• /
• pp.2946-2952
• /
• 2014

Among the Example based Super Resolution(SR) techniques, Neighbor embedding(NE) has been inspired by manifold learning method, particularly locally linear embedding. However, the poor generalization of NE decreases the performance of such algorithm. The sizes of local training sets are always too small to improve the performance of NE. We propose the Learning Sparse-Neighbor Image Representation baesd on SVR having an excellent generalization ability to solve this problem. Given a low resolution image, we first use bicubic interpolation to synthesize its high resolution version. We extract the patches from this synthesized image and determine whether each patch corresponds to regions with high or low spatial frequencies. After the weight of each patch is obtained by our method, we used to learn separate SVR models. Finally, we update the pixel values using the previously learned SVRs. Through experimental results, we quantitatively and qualitatively confirm the improved results of the proposed algorithm when comparing with conventional interpolation methods and NE.

• Nuclear Engineering and Technology
• /
• v.52 no.1
• /
• pp.164-169
• /
• 2020

Use of SiPM has been considered as an alternative to PMT, because of its compact size, low-operating voltage, non-sensitive to electromagnetic, low costs and so on. The main limitation for the use of SiPM is due to its small sensitive area compared to PMT that limits the light collection, and therefore the sensor energy resolution. In this article we studied the effect of increasing the number of SiPM by connecting them in parallel to increase the active detection area. This allowed us to compare the different energy resolution measurements. ¹³⁷Cs has been selected as reference to study the energy resolution for 662 keV gamma-rays. Another investigation was to compare the minimum detectable gamma energy under various SiPM configurations. It has been found that the use of 4 SiPM arrays can greatly improve the energy resolution up to 4% than only one SiPM array, meanwhile use of more than 2 SiPM arrays does not increase the energy resolution significantly. Thus we can conclude that for a large area of cylindrical scintillator (3 × 3 inches), the use of SiPMs are limited to a certain number or certai active area depending on the commercial SiPMs, and its cost should be less than traditional PMT for the cost-effective and compact size considerations. It is well known that the gain of SiPM varies with temperature. In this article, we also calibrated gain to guarantee the same position of photoelectric peak in response of different temperatures.

• Journal of Broadcast Engineering
• /
• v.24 no.4
• /
• pp.633-642
• /
• 2019

Recently, the convolution neural network (CNN) model at a single image super-resolution (SISR) have been very successful. The residual learning method can improve training stability and network performance in CNN. In this paper, we propose a SISR using recursive residual network architecture by introducing dense skip connections for learning nonlinear mapping from low-resolution input image to high-resolution target image. The proposed SISR method adopts a method of the recursive residual learning to mitigate the difficulty of the deep network training and remove unnecessary modules for easier to optimize in CNN layers because of the concise and compact recursive network via dense skip connection method. The proposed method not only alleviates the vanishing-gradient problem of a very deep network, but also get the outstanding performance with low complexity of neural network, which allows the neural network to perform training, thereby exhibiting improved performance of SISR method.

• Korean Journal of Remote Sensing
• /
• v.38 no.5_1
• /
• pp.559-570
• /
• 2022

When a satellite image has low spatial resolution, it is difficult to detect small objects. In this research, we aim to check the effect of super resolution on object detection. Super resolution is a software method that increases the resolution of an image. Unpaired super resolution network is used to improve Sentinel-2's spatial resolution from 10 m to 3.2 m. Faster-RCNN, RetinaNet, FCOS, and S²ANet were used to detect vessels in the Sentinel-2 images. We experimented the change in vessel detection performance when super resolution is applied. As a result, the Average Precision (AP) improved by at least 12.3% and up to 33.3% in the ship detection models trained with the super-resolution image. False positive and false negative cases also decreased. This implies that super resolution can be an important pre-processing step in object detection, and it is expected to greatly contribute to improving the accuracy of other image-based deep learning technologies along with object detection.

• Journal of Korean Society for Atmospheric Environment
• /
• v.24 no.6
• /
• pp.649-661
• /
• 2008

The impact of the detailed surface data on regional meteorological fields in complex coastal area is studied using RAMS. Resolutions of topography and land use data are very important to numerical modeling, because high resolution data can reflect correct terrain height and detail characteristics of the surface. Especially, in complex coastal region such as Gwangyang area, southern area in Korean Peninsula, high resolution topography and land use data are indispensable for accurate modeling results. This study investigated the effect of resolutions of terrain data using SRTM with 3 second resolution topography and KLU with 1 second resolution land use data. Case HR was the experiment using high resolution data, whereas Case LR used low resolution data. In Case HR, computed surface temperature was higher than Case LR along the coastline and wind speed was $1{\sim}2m/s$ weaker than Case LR. Time series of temperature and wind speed indicated great agreement with the observation data. Moreover, Case HR indicated outstanding results on statistical analysis such as regression, root mean square error, index of agreement.

• Journal of Korea Multimedia Society
• /
• v.22 no.11
• /
• pp.1242-1250
• /
• 2019

Super resolution technique aims to convert a low-resolution image with coarse details to a corresponding high-resolution image with refined details. In the past decades, the performance is greatly improved due to progress of deep learning models. However, universal solution for various objects is a still challenging issue. We observe that learning super resolution with a general dataset has poor performance on faces. In this paper, we propose a super resolution fusion scheme that works well for both general- and face datasets to achieve more universal solution. In addition, object-specific feature extractor is employed for better reconstruction performance. In our experiments, we compare our fusion image and super-resolved images from one- of the state-of-the-art deep learning models trained with DIV2K and FFHQ datasets. Quantitative and qualitative evaluates show that our fusion scheme successfully works well for both datasets. We expect our fusion scheme to be effective on other objects with poor performance and this will lead to universal solutions.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.31 no.12C
• /
• pp.1194-1200
• /
• 2006

In this paper, we propose an adaptive MAP (Maximum A Posteriori) high-resolution image reconstruction algorithm using local statistics. In order to preserve the edge information of an original high-resolution image, a visibility function defined by local statistics of the low-resolution image is incorporated into MAP estimation process, so that the local smoothness is adaptively controlled. The weighted non-quadratic convex functional is defined to obtain the optimal solution that is as close as possible to the original high-resolution image. An iterative algorithm is utilized for obtaining the solution, and the smoothing parameter is updated at each iteration step from the partially reconstructed high-resolution image is required. Experimental results demonstrate the capability of the proposed algorithm.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.18 no.3
• /
• pp.263-269
• /
• 2000

As the accuracy of digital photogrammetry is restricted by the resolution of image to be used, it is axiomatic that the resolution of image should be improved. As for the method to constitute hardware with CCD sensor that capacity was expanded or the method to acquire the image of high resolution by deciding the quantity of sub-pixel in advance through moving sensor, the price is expensive. This study tries to enhance the resolution of low resolution image by acquiring the image with the digital camera that the price is cheap and deciding shifts and rotations through matching multiple digital image by means of least square method. As the result of study, the resolution of digital image was improved greatly. So, not only the digital photogrammetry which has the competitive power of price economically is possible in the future but also the application is expected widely.

• Smart Structures and Systems
• /
• v.16 no.1
• /
• pp.47-65
• /
• 2015

A novel finite element (FE) model updating method based on multi-resolution analysis (MRA) is proposed. The true stiffness of the FE model is considered as the superposition of two pieces of stiffness information of different resolutions: the pre-defined stiffness information and updating stiffness information. While the resolution of former is solely decided by the meshing density of the FE model, the resolution of latter is decided by the limited information obtained from the experiment. The latter resolution is considerably lower than the former. Second generation wavelet is adopted to describe the updating stiffness information in the framework of MRA. This updating stiffness in MRA is realized at low level of resolution, therefore, needs less number of updating parameters. The efficiency of the optimization process is thus enhanced. The proposed method is suitable for the identification of multiple irregular cracks and performs well in capturing the global features of the structural damage. After the global features are identified, a refinement process proposed in the paper can be carried out to improve the performance of the MRA of the updating information. The effectiveness of the method is verified by numerical simulations of a box girder and the experiment of a three-span continues pre-stressed concrete bridge. It is shown that the proposed method corresponds well to the global features of the structural damage and is stable against the perturbation of modal parameters and small variations of the damage.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.2C
• /
• pp.145-152
• /
• 2009

Ultra-wideband communication systems for impulse radio have merits that are possible for either high resolution ranging system or radio determination. Conventionally, in order to accomplish these functions, the rapid analog to digital convertor (ADC) is necessary to apply radio determination system operating in time domain. However, considering that low rate - wireless personal area network (LR-WPAN) aims to low-cost hardware implementation, the expensive ADC converting GHz sampling per second is not appropriate. So, this paper introduces the high resolution ranging system operating in frequency domain with using low sampling rate ADC, and a new non-coherent ranging scheme utilizing analog Frequency Modulation (FM) mode for the frequency domain transformation. To verify the superiority of the proposed ranging algorithm working in frequency domain, the suggested IEEE 802.15.4a TG channel model is used to exploit affirmative features of the proposed algorithm with conducting the simulation results.