• Journal of Broadcast Engineering
• /
• v.28 no.2
• /
• pp.178-184
• /
• 2023

Finding texts in general scene images and recognizing their contents is a very important task that can be used as a basis for robot vision, visual assistance, and so on. However, for the low-resolution text images, the degradations, such as noise or blur included in text images, are more noticeable, which leads to severe performance degradation of text recognition accuracy. In this paper, we propose a new Korean text image super-resolution based on a Transformer-based model, which generally shows higher performance than convolutional neural networks. In the experiments, we show that text recognition accuracy for Korean text images can be improved when our proposed text image super-resolution method is used. We also propose a new Korean text image dataset for training our model, which contains massive HR-LR Korean text image pairs.

• Proceedings of the KSRS Conference
• /
• 2007.10a
• /
• pp.85-89
• /
• 2007

This paper presents the method to provide a fast service for user in image manipulation such as zooming and panning of huge size high resolution satellite image (e.g. Giga bytes per scene). The proposed technique is based on the hierarchical structure that has 3D-Tiling in horizontal and vertical direction to provide the image service more effectively than 2D-Tiling technique in the past does. The essence of the proposed technique is to create tiles that have optimum level of horizontal as well as vertical direction on the basis of current displaying area which changes as user manipulates huge image. So this technique provides seamless service, and will be very powerful and useful for manipulation of images of huge size without data conversion.

• Korean Journal of Remote Sensing
• /
• v.23 no.5
• /
• pp.375-383
• /
• 2007

This paper presents the method to provide a fast service for user in image manipulation such as zooming and panning of huge size high resolution satellite image(e.g. Giga bytes per scene). The proposed technique is based on the hierarchical structure that has 3D-Tiling in horizontal and vertical direction to provide the image service more effectively than 2D-Tiling technique in the past does. The essence of the proposed technique is to create tiles of optimum level in real time on the basis of current displaying area, which change as user manipulates huge image. Consequently, this technique provides seamless service, and will be very powerful and useful for manipulation of images of huge size without data conversion.

• Korean Journal of Remote Sensing
• /
• v.25 no.3
• /
• pp.233-242
• /
• 2009

This paper presents two different types of supervised classifiers such as support vector machine (SVM) and spectral angle mapper (SAM). The Compact Airborne Spectrographic Imager (CASI) high resolution aerial image was classified with the above two classifier. The image was classified into eight land use /land cover classes. Accuracy assessment and Kappa statistics were estimated for SVM and SAM separately. The overall classification accuracy and Kappa statistics value of the SAM were 69.0% and 0.62 respectively, which were higher than those of SVM (62.5%, 0.54).

• ETRI Journal
• /
• v.33 no.4
• /
• pp.497-505
• /
• 2011

Image fusion is a technical method to integrate the spatial details of the high-resolution panchromatic (HRP) image and the spectral information of low-resolution multispectral (LRM) images to produce high-resolution multispectral images. The most important point in image fusion is enhancing the spatial details of the HRP image and simultaneously maintaining the spectral information of the LRM images. This implies that the physical characteristics of a satellite sensor should be considered in the fusion process. Also, to fuse massive satellite images, the fusion method should have low computation costs. In this paper, we propose a fast and efficient satellite image fusion method. The proposed method uses the spectral response functions of a satellite sensor; thus, it rationally reflects the physical characteristics of the satellite sensor to the fused image. As a result, the proposed method provides high-quality fused images in terms of spectral and spatial evaluations. The experimental results of IKONOS images indicate that the proposed method outperforms the intensity-hue-saturation and wavelet-based methods.

• Proceedings of the Korean Society of Surveying, Geodesy, Photogrammetry, and Cartography Conference
• /
• 2004.04a
• /
• pp.383-388
• /
• 2004

The fusion of a different kind sensor is fusion of the obtained data by the respective independent technology. This is a important technology for the construction of 3D spatial information. particularly, information is variously realized by the fusion of LiDAR and mobile scanning system and digital map, fusion of LiDAR data and high resolution, LiDAR etc. This study is to generate union DEM and digital ortho image by the fusion of LiDAR data and high resolution image and monitor precisely topology, building, trees etc in urban areas using the union DEM and digital ortho image. using only the LiDAR data has some problems because it needs manual linearization and subjective reconstruction.

• Journal of the Korean Society for Precision Engineering
• /
• v.32 no.9
• /
• pp.833-840
• /
• 2015

Inkjet technology is one of the commonly used technologies in the printed-electronics field. The primary issue regarding the usage of inkjet technology as a printed-electronics tool is the printing resolution; therefore, to print complicated patterns of precise dimensions, we developed software that can convert a CAD file into a bitmap image. Moreover, as a bitmap image is not comprised of physical dimensions, its resolution needs to be defined in terms of pixel distance to print a rasterized bitmap image. In this study, we investigated the effects of pixel distance and dot size on printing resolution; based on the experiment results, we concluded that the printed dot size should be used to determine the optimal pixel distance. Lastly, we also propose inkjet printing procedures for bitmap images.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.8
• /
• pp.147-159
• /
• 1994

The performance of the improvement method of spatial resolution for satellite images based on the generalized inverse matrix is superior to the conventional methods. But, this method calculates the coefficient values for extracting the spatial information from the relation between a small pixel and large pixels. Accordingly it has the problem of remaining the blocky patterns at the result image. In this paper, a new generalized inverse matrix method is proposed which is different in the calculation method of coefficient values for extracting the spatial information. In this proposed metod, it calculates the coefficient values for extracting the spatial information from the relation between a small pixel and small pixels. Consequently it can improve the spatial resolution more efficiently without remaining the blocky patterns at the result image. The effectiveness of the proposed method is varified by simulation experiments with real TM image data.

• Journal of applied mathematics & informatics
• /
• v.41 no.3
• /
• pp.603-613
• /
• 2023

In this paper, the concept of geodesic centered tractography is explored for diffusion tensor imaging (DTI). In DTI, where geodesics has been tracked and the inverse of the fourth-order diffusion tensor is inured to determine the diversity. Specifically, we investigated geodesic tractography technique for High Angular Resolution Diffusion Imaging (HARDI). Riemannian geometry can be extended to a direction-dependent metric using Finsler geometry. Euler Lagrange geodesic calculations have been derived by Finsler geometry, which is expressed as HARDI in sixth order tensor.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.19 no.9
• /
• pp.943-949
• /
• 2009

Many studies for measuring vibration using image signal are suggested. These methods can measure vibration of multi-points simultaneously. However, it has the disadvantage that is very sensitive to an environment. If the measured environment is not good, image signals can be measured including much background noise. So, it is difficult to obtain accurate vibration from the measured image signals. Another problem is that camera imaging has a resolution limit. Because the resolution of the camera image is relatively much lower than that of a data acquisition system, accurate measuring vibration cannot be performed. In this paper, we proposed the enhanced technique for measuring vibration using camera signal. The key word of this paper is a curve fitting. The curve fitting can exactly detect the measurement line of interested object. So, we can measure the vibration in noisy environment. Also, it can overcome the resolution limit.