• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.4
• /
• pp.591-597
• /
• 2022

Recently, many studies have been done on the data augmentation technique as a way to efficiently build datasets. Among them, a representative data augmentation technique is a method of utilizing Generative Adversarial Network (GAN), which generates data similar to real data by competitively learning generators and discriminators. However, when learning GAN, there are cases where a broken pixel image occurs among similar data generated according to the environment and progress, which cannot be used as a dataset and causes an increase in learning time. In this paper, an algorithm was developed to select these damaged images by analyzing the histogram of image data generated during the GAN learning process, and as a result of comparing them with the images generated in the existing GAN, the ratio of the damaged images was reduced by 33.3 times(3,330%).

• Journal of KIISE:Software and Applications
• /
• v.37 no.10
• /
• pp.751-759
• /
• 2010

This paper is to propose a method to enhance the connectivity of the broken structures. The pixels to be connected are found by a decision rule which is based on the intensity and gradient information of the neighboring pixels. The intensity of the pixel, if it is decided to be filled, is increased so that the pixel together with its neighbors looks connected. The anisotropic diffusion follows to make the connected structural region look more natural. The same structure matrices have been used both to get gradient information and to decide the direction of diffusion to improve the computational speed. It has been shown by the experiments on the real ultrasound images that the broken structural regions can be connected by the proposed method.

• Proceedings of the IEEK Conference
• /
• 2002.06d
• /
• pp.373-376
• /
• 2002

This paper represents a survey of a variety thresholding techniques including both global and local thresholding. Several thresholding methods are examined in detail to evaluate their performance based on a given set of test images. We also attempt to evaluate the performance of several thresholding methods for construction field documents image recognition system using a broken line structures, broken symbols and text, blurring of lines, symbols and text, noise in homogeneous areas measure as a criterion functions.

• Fashion & Textile Research Journal
• /
• v.7 no.1
• /
• pp.56-62
• /
• 2005

Fancy Yarn has developed diverse textures in fabrics, reducing the time in yarn and fabric production or apparel making in order to develop creative goods. In this study aimed to propose the use of a 4D box system to make fancy yarn shapes with loops, knops and spirals and the like. The change in texture was analysed and simulated to produce a suitable fabric image by using the fancy yarns fabric. The results are as follows. The plain weave, 2/2 basket weave, 2/2 twill weave, 2/2 2 complete broken weaves, and 5 harness sateen weaves were woven and a fabric image formed. In the case of the loop and the knop yarns fabric image, compared to the twisted fabric image the surface was covered by loops or some parts became partially black. In the case of the spiral shape it showed pattern continuity in spiral shapes 1, 2 and 3. The more twisted spirals produced a diamond shaped pattern or a twill line and a herring bone shaped twill line. An evenly distributed black fabric image appeared in 5 harness sateen weave. For the loop shape the broken weave or 5 harness sateen weave was produced; basket weave and broken weave for the knop yarn 1 or knop yarn 2; and for the spiral shape a plain fabric or 5 harness sateen weave were produced much similar to the fabric image. The surface texture of the mapped image compared to the twisted fabric image produces fancy yarn fabric images covered with loops or irregular spots caused by the knop and the spiral. Therefore it is appropriate or suitable for the simulation of tweed or woolen wool fabrics. The fabric image which produced consistent and continuous lines is therefore more suitable for simulations of twill or herringbone fabric images.

• The Research Journal of the Costume Culture
• /
• v.11 no.2
• /
• pp.177-192
• /
• 2003

The purpose of this study is to define masculinity as a shilling concept, stemming from the process of power groups controlling the conscious and unconscious of human beings to diffuse their ideologies, and to examine how masculinity has been reflected in aesthetic images in men's bodies and fashion. As for the research methodology, literary research and demonstrative studies through the analysis of pictures and photos were undertaken. Emerging in the wake of the Industrial Revolution in the 19th century, the absolute powers such as the bourgeois elite, males and the Wet emphasized the importance of reason, and made aesthetic images such as authority, robustness and modernization in men's fashion. In the conversion to the Information Society, poweres horizintally spread out over society. By using non-reason, diverse power groups have broken down traditional masculinity and express diverse aesthetic images in men's fashion such as eroticism, multi-culturalism and liberation.

• International Journal of Contents
• /
• v.2 no.3
• /
• pp.27-32
• /
• 2006

We present an effective algorithm for automatic tracing of retinal vessel structure and vascular landmark extraction of bifurcations and ending points. In this paper we deal with vascular patterns from RGB images for personal identification. Vessel tracing algorithms are of interest in a variety of biometric and medical application such as personal identification, biometrics, and ophthalmic disorders like vessel change detection. However eye surface vasculature tracing in RGB images has many problems which are subject to improper illumination, glare, fade-out, shadow and artifacts arising from reflection, refraction, and dispersion. The proposed algorithm on vascular tracing employs multi-stage processing of ten-layers as followings: Image Acquisition, Image Enhancement by gray scale retinal image enhancement, reducing background artifact and illuminations and removing interlacing minute characteristics of vessels, Vascular Structure Extraction by connecting broken vessels, extracting vascular structure using eight directional information, and extracting retinal vascular structure, and Vascular Landmark Extraction by extracting bifurcations and ending points. The results of automatic retinal vessel extraction using jive different thresholds applied 34 eye images are presented. The results of vasculature tracing algorithm shows that the suggested algorithm can obtain not only robust and accurate vessel tracing but also vascular landmarks according to thresholds.

• Proceedings of the Korean Information Science Society Conference
• /
• 2007.10a
• /
• pp.229-230
• /
• 2007

Lane detection is one of the major components of traffic intelligence. It is impossible to recognize lanes as human do in all kinds of special situations; however, we can try to solve special problems with special methods. In this paper we propose a simple method using color segmentation, the Probabilistic Hough Transform (PHT), and the Least-Square in real-time lane tracking. Vehicles in neighborhood can be eliminated with one simple threshold in segmentation. Meanwhile, broken shape lanes in different road conditions can be successfully detected using the combination of PHT and Least-Square method. Eventually, this method is tested with groups of static images downloaded from internet and video sequences shot randomly on some highways. Satisfactory results are received.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.14 no.3
• /
• pp.482-491
• /
• 2011

The purpose of this study is to design and implement a 2-level secret information transmission system which can be used for information hiding of images transmitted over various IT communication media. To increase the robustness of the hiding power, we combined the steganography method which inserts secret object into cover object to hide the very fact of information hiding itself, and the preprocessing stage to encrypt the secret object before the stego-insertion stage. As a result, even when the stego-image is broken by an attacker, the secret image is protected by encryption. We implemented the 2-level image insertion and extraction algorithm by using C++ programming language. Experiment shows that the PSNR values of stego-images of ours exceed 30.00db which is the threshold of human recognition. The methodology of this study can be applied broadly to the information hiding and protection of the military secret images.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2003.05a
• /
• pp.161-164
• /
• 2003

A study on the gray images morphing using morphology and spline interpolation is realized. The adapted method is that join the 2 broken line together with the dilation of morphology to the morphing image, remove the holes to be a soft edge at the image boundaries and extract the featuring points with the cubic spline interpolation. As a result of experiment the method elicits more smoothing image morphing in detail and speed up the interframe processing.

• International Journal of Highway Engineering
• /
• v.3 no.4 s.10
• /
• pp.137-146
• /
• 2001

Pavement distresses are one of the most important data for pavement management systems. Inspection machines and its related programs have been used for operating tools in PMS developed in advanced countries. In Korea imported machines and programs for the length price ale utilized to get information of pavement condition from the field This study is launched for developing the program which can detect cracks on asphalt pavement due to many drawbacks in current PMS operation such as improper maintenance work and long resting period when it was broken. The focus of this study is to define principles to analyze pavement surface with digital image processing techniques, to test property of pavement images and to suggest an algorithm that reduces noises at test. To test images, the camera attached on the Automatic Road Analyser(ARAN) was used. Through the FFT images, histogram and statistical values of pavement images, it was found that the images had many noises with high-frequency components against general images, and it was difficult to subdivide pavement images into background or crack. Through several testing with various filters for noise reduction a 3X3 median filter was suggested to reduce noises effectively.