• Smart Media Journal
• /
• v.11 no.7
• /
• pp.61-69
• /
• 2022

In the learned machine learning, the performance of machine learning degrades at the same time as drift occurs in terms of learning models and learning data over time. As a solution to this problem, I would like to propose the concept and evaluation method of ML drift to determine the re-learning period of machine learning. An XAI test and an XAI test of an apple image were performed according to strawberry and clarity. In the case of strawberries, the change in the XAI analysis of ML models according to the clarity value was insignificant, and in the case of XAI of apple image, apples normally classified objects and heat map areas, but in the case of apple flowers and buds, the results were insignificant compared to strawberries and apples. This is expected to be caused by the lack of learning images of apple flowers and buds, and more apple flowers and buds will be studied and tested in the future.

• Journal of Ocean Engineering and Technology
• /
• v.36 no.1
• /
• pp.32-40
• /
• 2022

Underwater optical images face various limitations that degrade the image quality compared with optical images taken in our atmosphere. Attenuation according to the wavelength of light and reflection by very small floating objects cause low contrast, blurry clarity, and color degradation in underwater images. We constructed an image data of the Korean sea and enhanced it by learning the characteristics of underwater images using the deep learning techniques of CycleGAN (cycle-consistent adversarial network), UGAN (underwater GAN), FUnIE-GAN (fast underwater image enhancement GAN). In addition, the underwater optical image was enhanced using the image processing technique of Image Fusion. For a quantitative performance comparison, UIQM (underwater image quality measure), which evaluates the performance of the enhancement in terms of colorfulness, sharpness, and contrast, and UCIQE (underwater color image quality evaluation), which evaluates the performance in terms of chroma, luminance, and saturation were calculated. For 100 underwater images taken in Korean seas, the average UIQMs of CycleGAN, UGAN, and FUnIE-GAN were 3.91, 3.42, and 2.66, respectively, and the average UCIQEs were measured to be 29.9, 26.77, and 22.88, respectively. The average UIQM and UCIQE of Image Fusion were 3.63 and 23.59, respectively. CycleGAN and UGAN qualitatively and quantitatively improved the image quality in various underwater environments, and FUnIE-GAN had performance differences depending on the underwater environment. Image Fusion showed good performance in terms of color correction and sharpness enhancement. It is expected that this method can be used for monitoring underwater works and the autonomous operation of unmanned vehicles by improving the visibility of underwater situations more accurately.

• Archives of design research
• /
• v.17 no.1
• /
• pp.329-338
• /
• 2004

This study did derive a characteristic of a cover design through considering a concept, a function and components of a cover design based on a literature study and through analyzing cases of the best-seller cover designs for the literature section from 1999 to 2002. In terms of a reader's preference for a cover design according to an age and sex of a reader, it tried to suggest a condition and a method of a book-cover design through a questionnaire survey and the image evaluation. In a survey of the preference for a cover design, there was the difference of interest degree in a kind of a book, a place to buy a book and a cover design according to an age and sex. In addition, it was shown that a woman prefers the stylish typography, and both a man and a woman prefer the cool color. However, except these two, the reason why there is no difference in preference for visual elements are that a cover design is an integrative image comprising several elements. Accordingly, through the Semantic Differential Method, it carried out the evaluation as to a cover design as an integrative image. As a result of that, an abstract image was shown to be difficult to convey a meaning because of its fall in clarity. Further, a cover design mainly comprising the stylish typography gained the high evaluation in unique and clarity. It is thought to be attributable to the easiness to achieve a goal of seeing and reading in a cover design in which an image and the typography are balanced, rather than the excessive visual. Even though the visual is excellent, there was difference from what having been excellent in conveying a meaning, which is a functional aspect. This study tried to examine, limiting it only to a visual element. It will be hereafter required to proceed with a study on the influence on a reader's preference according to material, bookbinding and a form of book which are elements comprising a cover design.

• Journal of information and communication convergence engineering
• /
• v.7 no.4
• /
• pp.442-448
• /
• 2009

This paper proposes a fingerprint image processing algorithm to accurately extract minutiae in the process of fingerprint recognition. We improved the matching accuracy of low quality fingerprint images by using effective ridge vector and ridge probability. The proposed algorithm improves the clarity of ridge structures and reduces undesired noise. We collected thumb print images from 10 individuals 5 separate times each, in total using 50 thumbprints. We registered one of the five thumbprint images from each individual to match the registered one with the other four thumbprint images, and alternated the registered thumbprint image. We matched thumbprints 20 times for each individual. In total, we conducted 200 matches for the thumbprints from the 10 individuals. We improved the verification accuracy and reliability compared to conventional methods.

• Journal of Korean Academy of Oral and Maxillofacial Radiology
• /
• v.19 no.1
• /
• pp.25-29
• /
• 1989

The author has evaluated the panoramic image clarity of the midfacial anatomic structures in dry skull according to the skull position. The radiopaque markers were attached to the anatomic structures: infraorbial rim, upper and lower borders of zygomatic arch, pterygomaxillary fissure, lateral pterygoid plate, pyriform aperture of nasal cavity, lateral wall of maxilla, orbital floor, infraorbital foramen, and nasal floor. Position of the skull were divided into four groups. standard, 25mm forward, chin-down, chin-up position. The results were as follows: 1. The pyriform aperture of nasal cavity, lateral wall of the maxilla, orbital floor, infraorbital foramen and nasal floor did net cast any discernible image. 2. Nearly all images of midfacial structures were blurred in the chin-up position. 3. The forward position provided good visualization of the maxillary sinus. 4. The chin-down position provided good visualization of the zygomatic arch, pterygomaxillary fissue, and lateral pterygoid plate.

• Industrial Engineering and Management Systems
• /
• v.9 no.1
• /
• pp.20-27
• /
• 2010

The purpose of this study was to compare the consistency of external ear measurements obtained by electronic digital caliper, photocopier scanning and digital photographic methods. Photogrammetric measurements were made after image editing software was used to optimize the brightness, contrast, size and image clarity. The CorelDRAW dimension tool was used to create a dimension line that measured the vertical and horizontal length between any two landmarks. Results of repeated measures analysis of variance revealed no significant differences in length and width of the pinna between the three methods. The reliability of the three measurement techniques showed a high degree of consistency. Further study and efforts could be extended to measurement hands, foots and facial dimensions by present techniques.

• Journal of the Korea Fashion and Costume Design Association
• /
• v.14 no.2
• /
• pp.11-21
• /
• 2012

The purpose of this study is to understand of future image in fashion for 20C, to suggest a guideline's role in the development on fashion designs study. The methods of this study are used academic literatures as well as practical study. Futuristic image in fashion is to introduce about the future period, as fashion goes ahead times. Trends and formative characteristics of future image in 20C fashion are summarized as follows. First, futuristic images in fashion for 20C had emerged strongly futurism, space look, kinetic look, glitter look, techno look, cyber look. Second, perspectives on response for future society's changes are summarized as follows. Attitude of futuristic images of fashion, futurism, space look, kinetic look and glitter look, had been favorable about future society's changes until 1980. In 1990, techno look had been defensive and cyber look had been aggressive about future society's changes. Third, formative characteristics of future image in 20C fashion are studied shape, material, color, pattern. Shape was preferred linear and geometric form in the first half of the 20C, but body conscious type had appeared often in the second half of 20C. Material maintained the luster or the metal touch. Color was based image of colorless and preferred the gold, silver or color feels like metal. Pattern was preferred clarity with geometric type in the first half of the 20C. On the other hand in the second half of 20C pattern of future image was dominated by vague and abstract chaotic type.

• Computers and Concrete
• /
• v.19 no.2
• /
• pp.203-210
• /
• 2017

The purpose of this paper is to evaluate the settlement of lightweight coarse aggregate of self-compacting lightweight concrete (SCLC) after placement of concrete on its final position. To investigate this issue, sixteen samples of concrete mixes were made. The water to cementitious materials ratios of the mixes were 0.35 and 0.4. In addition to the workability tests of self-compacting concrete (SCC) such as slump flow, V-funnel and L-box tests, a laboratory experiment was made to examine the segregation of lightweight coarse aggregate in concrete. Because of the difficulties of this test, the image processing technique of MATLAB software was used to check the segregation above too. Moreover, the fuzzy logic technique of MATLAB software was utilized to improve the clarity of the borders between the coarse aggregate and the paste of the mixtures. At the end, the results of segregation tests and software analyses are given and the accuracy of the software analyses is evaluated. It is worth noting that the minimum and maximum differences between the results of laboratory tests and software analyses were 1.2% and 9.19% respectively. It means, the results of image processing technique looks exact enough for estimating the segregation of lightweight coarse aggregate in SCLC.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.10
• /
• pp.2553-2561
• /
• 1997

In this paper, a hardware architecture to calculate straight-line Hough transform algorithm for image recognition is suggested. This processor consists of the filtering module for gradient calculation and the HT calculation module, and the angle information are stored in memory table. For the suggested architecture, firstly, algorithm simulation is executed using C language to confirm the operation and to decide the precision of calculation, and secondly, architecture simulation is executed using VHDL language for the total blocks. According to C & VHDL simulation results, it is confirmed that the calculated data value is similarly obtained and the calculation defference is decreased as image clarity and bits increase.

• Journal of Information Processing Systems
• /
• v.18 no.3
• /
• pp.319-331
• /
• 2022

The capability to examine an X-ray image is so far a challenging task. In this work, we suggest a practical and novel algorithm based on image fusion to inspect the issues such as background noise, blurriness, or sharpness, which curbs the quality of dual-energy X-ray images. The current technology exercised for the examination of bags and baggage is "X-ray"; however, the results of the incumbent technology used show blurred and low contrast level images. This paper aims to improve the quality of X-ray images for a clearer vision of illegitimate or volatile substances. A dataset of 40 images was taken for the experiment, but for clarity, the results of only 13 images have been shown. The results were evaluated using MSE and PSNR metrics, where the average PSNR value of the proposed system compared to single X-ray images was increased by 19.3%, and the MSE value decreased by 17.3%. The results show that the proposed framework will help discern threats and the entire scanning process.