• Journal of Broadcast Engineering
• /
• v.18 no.3
• /
• pp.463-473
• /
• 2013

In this paper, we propose a super-resolution (SR) image reconstruction algorithm using multi-view images. We acquire 25 images from multi-view cameras, which consist of a $5{\times}5$ array of cameras, and then reconstruct an SR image of the center image using a low resolution (LR) input image and the other 24 LR reference images. First, we estimate disparity maps from the input image to the 24 reference images, respectively. Then, we interpolate a SR image by employing the LR image and matching points in the reference images. Finally, we refine the SR image using an iterative regularization scheme. Experimental results demonstrate that the proposed algorithm provides higher quality SR images than conventional algorithms.

• Journal of KIISE:Software and Applications
• /
• v.37 no.4
• /
• pp.264-272
• /
• 2010

Example-based super resolution is a method to restore a high resolution image from low resolution images through training and retrieval of image patches. It is not only good in its performance but also available for a single frame low-resolution image. However, its time complexity is very high because it requires lots of comparisons to retrieve image patches in restoration process. In order to improve the restoration speed, an efficient patch retrieval algorithm is essential. In this paper, we applied various high-dimensional feature retrieval methods, available for the patch retrieval, to a practical example-based super resolution system and compared their speed. As well, we propose to apply the multi-phase candidate reduction approach to the patch retrieval process, which was successfully applied in character recognition fields but not used for the super resolution. In the experiments, LSH was the fastest among conventional methods. The multi-phase candidate reduction method, proposed in this paper, was even faster than LSH: For $1024{\times}1024$ images, it was 3.12 times faster than LSH.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.9
• /
• pp.21-32
• /
• 2022

Image labeling must be preceded in order to perform object detection, and this task is considered a significant burden in building a deep learning model. Tens of thousands of images need to be trained for building a deep learning model, and human labelers have many limitations in labeling these images manually. In order to overcome these difficulties, this study proposes a method to perform object detection without significant performance degradation, even though labeling some images rather than the entire image. Specifically, in this study, low-resolution oriental painting images are converted into high-quality images using a super-resolution algorithm, and the effect of SSIM and PSNR derived in this process on the mAP of object detection is analyzed. We expect that the results of this study can contribute significantly to constructing deep learning models such as image classification, object detection, and image segmentation that require efficient image labeling.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.46 no.6
• /
• pp.68-77
• /
• 2009

A single field deinterlacing method, namely interpolation algorithm derived from low resolution (ILR), is presented in this paper. Traditional deinterlacing methods usually employ edge-based interpolation technique within pixel-based estimation. However, edge-based methods are somehow sensitive to noise and intensity variation in the image. Moreover, the methods are not satisfied in deciding the exact edge direction which controls the performance of the interpolation. In order to reduce the sensitivity, the proposed algorithm investigates low-resolution characteristics of the pixel to be interpolated, and applies it to high-resolution image. Simulation results demonstrates that the proposed method gives not only a better objective performance in terms of PSNR results compare to conventional edge-based interpolation methods, but also better subjective image quality.

• Journal of the Korean Data and Information Science Society
• /
• v.14 no.3
• /
• pp.525-533
• /
• 2003

In this paper we introduce some existing preprocessing algorithm for character recognition and consider feature extraction method for the recognition of low resolution document. Image recognition of low resolution document including fax images can be frequently misclassified due to the blurring effect, slope effect, noise and so on. In order to overcome these difficulties in the character recognition we considered a mesh feature extraction and contour direction code feature. System for automatic character recognition were suggested.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.11
• /
• pp.62-71
• /
• 1996

In this paper we propose a method that improves PSNR at low bit rate and reduces computational complexity in fractal image coding based on discrete wavelet transform. The proposed method, which uses significant coefficient tree, improves PSNR of the reconstructed image and reduces computational comlexity of mapping domain block onto range block by matching only the significant coefficients of range block to coefficients of domain block. Also, the proposed method reduces error propagation form lower resolution subbands to higher resolution subbands by correcting error of lower resolution subbands. Some experimental results confirm that the proposed method reduces encoding and decoding time significantly and has fine reconstructed images having no blocking effect and clear edges at low bit rate.

• Journal of the Microelectronics and Packaging Society
• /
• v.27 no.2
• /
• pp.1-10
• /
• 2020

The technology and market size of image sensors continue to develop thanks to the release of image sensors that exceed 100 million pixels in 2019 and expansion of black box camera markets for vehicles in addition to existing mobile applications. We review the technology flow of image sensors that have been constantly evolving for 40 years since Hitachi launched a 200,000-pixel image sensor in 1979. Although CCD has made inroads into image sensor market for a while based on good picture quality, CMOS image sensor (CIS) with active pixels has made inroads into the market as semiconductor technology continues to develop, since the electrons generated by the incident light are converted to the electric signals in the pixel, and the power consumption is low. CIS image sensors with superior characteristics such as high resolution, high sensitivity, low power consumption, low noise and vivid color continue to be released as the new technologies are incorporated. At present, new types of structures such as Backside Illumination and Isolation Cell have been adopted, with better sensitivity and high S/N ratio. In the future, new photoconductive materials are expected to be adopted as a light absorption part in place of the pn junction.

• Current Optics and Photonics
• /
• v.2 no.2
• /
• pp.153-164
• /
• 2018

In semiconductor manufacturing, critical dimensions indicate the features of patterns formed by the semiconductor process. The purpose of measuring critical dimensions is to confirm whether patterns are made as intended. The deposition process for an organic light emitting diode (OLED) forms a luminous organic layer on the thin-film transistor electrode. The position of this organic layer greatly affects the luminescent performance of an OLED. Thus, a system for measuring the position of the organic layer from outside of the vacuum chamber in real-time is desired for monitoring the deposition process. Typically, imaging from large stand-off distances results in low spatial resolution because of diffraction blur, and it is difficult to attain an adequate industrial-level measurement. The proposed method offers a new superresolution single-image using a conversion formula between two different optical systems obtained by a deep learning technique. This formula converts an image measured at long distance and with low-resolution optics into one image as if it were measured with high-resolution optics. The performance of this method is evaluated with various samples in terms of spatial resolution and measurement performance.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.26 no.3
• /
• pp.234-245
• /
• 2023

The Arena Fragmentation Test(AFT) is designed to analyze warhead performance by measuring fragmentation data. In order to evaluate the results of the AFT, a set of AFT images are captured by high-speed cameras. To detect objects in the AFT image set, ResNet-50 based Faster R-CNN is used as a detection model. However, because of the low resolution of the AFT image set, a detection model has shown low performance. To enhance the performance of the detection model, Super-resolution(SR) methods are used to increase the AFT image set resolution. To this end, The Bicubic method and three SR models: ZSSR, EDSR, and SwinIR are used. The use of SR images results in an increase in the performance of the detection model. While the increase in the number of pixels representing a fragment flame in the AFT images improves the Recall performance of the detection model, the number of pixels representing noise also increases, leading to a slight decreases in Precision performance. Consequently, the F1 score is increased by up to 9 %, demonstrating the effectiveness of SR in enhancing the performance of the detection model.

• Proceedings of the KSRS Conference
• /
• 2002.10a
• /
• pp.792-797
• /
• 2002

In the following paper two algorithms, suitable for the analysis of panchromatic data as provided by KOMPSAT-1 will be presented. One is a texture analysis which will be used to create a settlement mask based on the variations of gray values. The other is a fusion algorithm which allows the combination of high resolution panchromatic data with medium resolution multispectral data. The procedure developed for this purpose uses the spatial information present in the high resolution image to spatially enhance the low resolution image, while keeping the distortion of the multispectral information to a minimum. This makes it possible to use the fusion results for standard multispecatral classification routines. The procedures presented here can be automated to large extent, making them suitable for a standard processing routine of satellite data.