• Journal of the Korea Society of Computer and Information
• /
• v.25 no.11
• /
• pp.97-104
• /
• 2020

In this paper, we propose a random number generator PP(PingPong256) and a shuffle technique to improve the problem that the encrypted image is damaged due to a lot of noise by the channel coding of wireless communication recommended in the special environment of space. The PP can constantly generate random numbers by entering an initial value of 512 bits. Random numbers can be encrypted through images and exclusive logical computations. Random numbers can be encrypted through images and exclusive logical computations. The shuffle technique randomly rearranges the image pixel positions while synchronizing the image pixel position and the random number array position and moving the random number arrangement in ascending order. Therefore, the use of PP and shuffle techniques in channel coding allows all pixels to be finely distributed and transmit high-quality images even in poor transmission environments.

• Journal of Korea Multimedia Society
• /
• v.3 no.1
• /
• pp.34-40
• /
• 2000

'[n this paper, we propose an efficient multi-view image coding algorithm which finds the optimal texture from a restricted number of multi-view image. The X-Y plane of the normalized object space is divided into the triangular patches. The depth of each node is determined by appling a block based disparity compensation method. Thereafter the texture of each patch is extracted by appling an affine transformation based disparity compensation method to the multi-view images. We reduced the number of images needed to determine the texture compared to traditional methods which use all the multi-view image in the texture extraction. The experimental results show that the SNR of images encoded by the proposed algorithm is better than that of images encoded by the traditional method by the approximately 0.2dB for the test sets of multi -view image called dragon, santa, city and kid. Image data recovered after encoding by the proposed method show a better visual results than after using traditional method.

• Journal of Broadcast Engineering
• /
• v.22 no.6
• /
• pp.713-723
• /
• 2017

Since the panoramic image can overcome the limitation of the viewing angle of the camera and have a wide field of view, it has been studied effectively in the fields of computer vision and stereo camera. In order to generate a panoramic image, stitching images taken by a plurality of general cameras instead of using a wide-angle camera, which is distorted, is widely used because it can reduce image distortion. The image stitching technique creates descriptors of feature points extracted from multiple images, compares the similarities of feature points, and links them together into one image. Each feature point has several hundreds of dimensions of information, and data processing time increases as more images are stitched. In particular, when a panorama is generated on the basis of an image photographed by a plurality of unspecified cameras with respect to an object, the extraction processing time of the overlapping feature points for similar images becomes longer. In this paper, we propose a preprocessing process to efficiently process stitching based on an image obtained from a number of unspecified cameras for one object or environment. In this way, the data processing time can be reduced by pre-grouping images based on camera sensor information and reducing the number of images to be stitched at one time. Later, stitching is done hierarchically to create one large panorama. Through the grouping preprocessing proposed in this paper, we confirmed that the stitching time for a large number of images is greatly reduced by experimental results.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.7
• /
• pp.84-91
• /
• 2017

This study was performed using Somatom Definition Flash (Siemens, Enlarge, Germany) and GE 64-MDCT (Discovery 750 HD, GE HEALTHCARE, Milwaukee, USA.) using high-density medical materials that (are indispensable to?) computed tomography. We analyzed quantitatively the rate of change of the CT number of the CT reconstruction images by means of the IMAR and MAR algorithms using the phantom images acquired after scanning and previously captured raw data images. As a result, it was shown that the IMAR and MAR algorithms provided if ferent phantom images in the case of all medical high-density materials (p <0.05). The black streak artifacts were analyzed using the MAR and IMAR algorithms to determine if they corresponded to stainless steel materials (p>0.05). Also, it was found that the application of the IMAR algorithm affects the attenuation deviation, because there is a change in the image CT number compared to that before. The results suggest that, in the future, after the implant procedure, it would be useful to observe the surgical site and surrounding tissues during follow-up CT scans.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.13 no.4
• /
• pp.845-850
• /
• 2018

In this paper, we propose a method to estimate the number of biofouling attached to the surface of marine structures. This method estimates the number of biofouling by calculating the region maxima using images taken in underwater. To do this, we analyze the correlation between the region maxima and the number of biofouling. The analysis showed that there is a significant correlation between the number of region maxima and the number of biofouling. By using the results of this analysis, the experiments were conducted on images taken in the underwater. Experimental results show that the higher the region maxima of the image, is greater than the number of biofouling in the image. The proposed method can be used as an important technology in computer vision for underwater images.

• Smart Structures and Systems
• /
• v.34 no.2
• /
• pp.129-143
• /
• 2024

In recent years, with the vigorous development of visual algorithms, a large amount of research has been conducted on blade surface defect detection methods represented by deep learning. Detection methods based on deep learning models must rely on a large and rich dataset. However, the geographical location and working environment of wind turbines makes it difficult to effectively capture images of blade surface defects, which inevitably hinders visual detection. In response to the challenge of collecting a dataset for surface defects that are difficult to obtain, a multi-class blade surface defect generation method based on the StyleGAN3 (Style Generative Adversarial Networks 3) deep learning model and PBGMs (Physics-Based Graphics Models) method has been proposed. Firstly, a small number of real blade surface defect datasets are trained using the adversarial neural network of the StyleGAN3 deep learning model to generate a large number of high-resolution blade surface defect images. Secondly, the generated images are processed through Matting and Resize operations to create defect foreground images. The blade background images produced using PBGM technology are randomly fused, resulting in a diverse and high-resolution blade surface defect dataset with multiple types of backgrounds. Finally, experimental validation has proven that the adoption of this method can generate images with defect characteristics and high resolution, achieving a proportion of over 98.5%. Additionally, utilizing the EISeg annotation method significantly reduces the annotation time to just 1/7 of the time required for traditional methods. These generated images and annotated data of blade surface defects provide robust support for the detection of blade surface defects.

• International journal of advanced smart convergence
• /
• v.8 no.1
• /
• pp.24-34
• /
• 2019

In this research, a practical deep learning framework to differentiate the lesions and nodules in breast acquired with ultrasound imaging has been proposed. 7408 ultrasound breast images of 5151 patient cases were collected. All cases were biopsy proven and lesions were semi-automatically segmented. To compensate for the shift caused in the segmentation, the boundaries of each lesion were drawn using Fully Convolutional Networks(FCN) segmentation method based on the radiologist's specified point. The data set consists of 4254 benign and 3154 malignant lesions. In 7408 ultrasound breast images, the number of training images is 6579, and the number of test images is 829. The margin between the boundary of each lesion and the boundary of the image itself varied for training image augmentation. The training images were augmented by varying the margin between the boundary of each lesion and the boundary of the image itself. The images were processed through histogram equalization, image cropping, and margin augmentation. The networks trained on the data with augmentation and the data without augmentation all had AUC over 0.95. The network exhibited about 90% accuracy, 0.86 sensitivity and 0.95 specificity. Although the proposed framework still requires to point to the location of the target ROI with the help of radiologists, the result of the suggested framework showed promising results. It supports human radiologist to give successful performance and helps to create a fluent diagnostic workflow that meets the fundamental purpose of CADx.

• Proceedings of the Korean Association of Geographic Inforamtion Studies Conference
• /
• 2008.10a
• /
• pp.245-246
• /
• 2008

The goal of this study is to explore the possibility of KLT tracker for tracking the features between two images including rotation and shift. As a test site, Jangsu-Gun area of South Korea is selected and the images taken from UAV camera are used for analysis. The analysis was carried out using KLT tracker developed in a PC environment. The results of the experiment used two images with the large overlapping area are compared with the results of two images with the little overlapping area and rotation. Overall, the research indicates that the integrated features of littlerotation and motion images can significantly increase during the tracking process. But using KLT tracker for extracting and tracking features between images with large rotation and motion, the number of tracked features are decreased.

• Journal of Korean Academy of Oral and Maxillofacial Radiology
• /
• v.28 no.2
• /
• pp.321-327
• /
• 1998

The author examined bone changes from 1274 poly tomographic images of 182 temporomandibular joints which showed symptoms of temporomandibular disorder and the following results were obtained; 1. The number of temporomandibular joints which showed bone changes were 64 (35.2％) among 182 joints. 2. The age and sex distribution of 64 joints which had bone changes showed the prevalence of female (90.6％) and third decade (25.0％) followed by fourth (21.2％) and second decade (17.2％). 3. The 252 images which showed bone changes consisted of 56 images from lateral side (22.2％). 118 images from center (46.8％) and 78 images from medial side (30.9％). 4. The most frequently observed bone changes were flattening (22.7％) followed by sclerosis (19.3％) and cortical unsharpness (19.3％)

• Journal of Radiation Protection and Research
• /
• v.40 no.1
• /
• pp.17-24
• /
• 2015

To improve accuracy of dose calculation on kilovoltage cone beam computed tomography (kV CBCT) images, a custom-made phantom was fabricated to acquire an accurate CT number to electron density curve by full scatter of cone beam x-ray. To evaluate the dosimetric accuracy, 9 volumetric modulated arc therapy (VMAT) plans for head and neck (HN) cancer and 9 VMAT plans for lung cancer were generated with an anthropomorphic phantom. Both CT and CBCT images of the anthropomorphic phantom were acquired and dose-volumetric parameters on the CT images with CT density curve (CTCT), CBCT images with CT density curve ($CBCT_{CT}$) and CBCT images with CBCT density curve ($CBCT_{CBCT}$) were calculated for each VMAT plan. The differences between $CT_{CT}$ vs. $CBCT_{CT}$ were similar to those between $CT_{CT}$ vs. $CBCT_{CBCT}$ for HN VMAT plans. However, the differences between $CT_{CT}$ vs. $CBCT_{CT}$ were larger than those between $CT_{CT}$ vs. $CBCT_{CBCT}$ for lung VMAT plans. Especially, the differences in $D_{98%}$ and $D_{95%}$ of lung target volume were statistically significant (4.7% vs. 0.8% with p = 0.033 for $D_{98%}$ and 4.8% vs. 0.5% with p = 0.030 for $D_{95%}$). In order to calculate dose distributions accurately on the CBCT images, CBCT density curve generated with full scatter condition should be used especially for dose calculations in the region of large inhomogeneity.