• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.561-562
• /
• 2023

본 논문에서는 YOLOv5m과 이미지 전처리 기법을 활용한 문 객체 감지 시스템을 제안한다. 이미지 전처리를 하지 않은 Original 이미지 그리고 이미지 전처리를 한 CLAHE 이미지, Bilateral 이미지 세 가지를 사용해서 가장 좋은 기법을 비교한다. mAP 진행 그래프 및 이미지 출력을 통해 결과를 검증한다. 본 논문의 목표는 인공지능이 문을 감지하는 알고리즘을 구현하여 배송 로봇이 목적지의 문을 찾아내는 것이다

• Archives of design research
• /
• v.18 no.1 s.59
• /
• pp.27-38
• /
• 2005

The purpose of this study is to find the unique image characteristics of Korea and identify the Korea culture by comparing the image characteristics of traditional palaces of Korea, China and Japan. This study identifies the unique image characteristics of Korea through a comparative analysis the image of three nations. Literature research, free association measurement, survey were conducted to select study objects and descriptive words. 5 outside-pictures and 2 inside-pictures of the Changduckung, the Forbidden City, the Nijo castle representing each country's traditional palaces were chosen as study objects. A questionnaire was made with 47 descriptive words and 5-scale measurement. The results of analysis are as follows. The representative words to express the image characteristics of Korea, China and Japan traditional palaces are adornment, stability, openness, lines, unfamiliarity, feminine. This study researched the common and different factors between the image characteristics in traditional palaces of Korea, China and Japan. The common image is the image of lines. The distinctive characteristics of Korea are stability, curved lines and femininity. The characteristics of China are adornment and straight lines. And, the characteristics of Japan are simplicity, unfamiliarity, and openness. Based on the research, it was found out that stability and curved lines are the main characteristics of Korea. This study researches the characteristics of images of Korea, China and Japan. This study shows the common and different factors of image characteristics of in the neighboring three nations of East Asia. Due to the geographical and cultural effect over a long time, the Korea was evaluated as the middle cultural between China and Japan. However, this study found unique characteristics of Korea's image.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2008.05a
• /
• pp.1779-1783
• /
• 2008

영상해석을 통한 흐름해석의 방법인 Large-Scale Particle Image Velocimetry (LSPIV)는 실험실내의 소규모 흐름해석에 이용하던 Particle Image Velocimetry (PIV)를 자연하천이나 실험실에서 넓은 영역($4m^2{\sim}45,000m^2$)에 적용할 수 있도록 확장시킨 것으로 지난 10여년전부터 세계적으로 널리 이에 대한 연구가 진행되고 있다. PIV는 seeding, illumination, recording 그리고 image processing으로 구성된다. LSPIV(Large Scale PIV)는 PIV의 기본원리를 근거로 하여 기존의 PIV에 비하여 실험실 내에서의 수리모형실험이나 일반 하천에서의 유속측정과 같은 큰 규모의 흐름해석을 할 수 있도록 seeding, illumination에 대한 조정이 필요 하고, 촬영된 image에 대한 왜곡을 없애는 작업이 필요하다. LSPIV는 PIV의 네 가지 단계를 포함하여 seeding, illumination, recording, image transformation, image processing 및 post-processing의 여섯 단계로 구성되어진다 (Li, 2002). LSPIV의 적용시 각 단계마다 유속계산시 오차를 발생시키는 27가지의 요인들이 존재하고 있는바 (Kim, 2006), 본 연구에서는 이들 중 실내의 실험실에서 파악이 가능한 인자들에 대해 그들 각각의 인자들이 유속 측정에 미치는 오차의 정도를 파악하고자 하였다. 본 연구에서는 LSPIV의 적용시 이용되는 이미지의 개수와 이미지 촬영시 적용된 이미지의 해상도에 따른 오차의 발생 정도를 조사하였다. 이미지 촬영에 있어서 비디오카메라를 이용할 경우 촬영시간에 따라 많은 수의 이미지를 취득할 수 있은바 이미지의 수에 따른 유속계산오차를 파악하고자 하였다. 또한 디지털 카메라를 이용할 경우 여러 가지 이미지 해상도를 이용할 수 있으므로 적용한 이미지 해상도에 따른 유속계산에 미치는 오차의 크기를 파악하고자 하였다. 이미지의 갯수가 유속계산시 미치는 오차의 영향의 정도를 조사하기 위해서 초당 30 frame을 촬영할 수 있는 비디오카메라를 이용하여 91초 동안 촬영된 이미지로부터 매 5번째의 이미지를 추출하여 455개의 이미지를 준비하였고 이로부터 이미지수를 10, 50, 100, 200, 300, 400의 순서로 증가시키면서 이미지 개수로부터 나타나는 유속계산 오차를 조사한 결과 이미지의 개수가 50매 이상인 경우는 이로 인한 오차가 1% 이하로 감소함을 파악하였다. 촬영된 이미지의 해상도가 유속계산시 미치는 영향을 조사하기 위해 디지털카메라를 적용하여 세가지 이미지 해상도(640*480, 1280*960, 2048*1536 pixel)로 변화시키면서 유속측정 오차를 분석한 결과 저해상도의 이미지를 이용한 경우 고해상도 이미지를 이용한 경우와 비교하여 3% 가량의 차이를 나타내었다.

• Journal of Global Scholars of Marketing Science
• /
• v.15 no.2
• /
• pp.49-69
• /
• 2005

This study is to investigate causal relationship among store image, private brand image and purchase intentions through comparing national distributors with local distributors in Pusan city. More specifically, the objective of this study can be explained as follows; First, what store image have an effect on private brand image. Second, individual store image and their private brand image have different effects. Last, store image and private brand image influence purchase intention of private brands. The results of this study suggests a positive relationship between consumers' perceptions of individual private brands and their associated store image dimensions and purchase intentions toward private brands. A key implication of this research suggests it is desirable for retailers to build up a positive consumer's perceptions of private brand by improving store image. Especially, it is suggested that through the development of specific features of store images, retailers are to manage private brands.

• Journal of the HCI Society of Korea
• /
• v.14 no.1
• /
• pp.5-13
• /
• 2019

SNS has been advanced from first to third generation by changing its service format in various ways. Nowadays, image-based SNS such as instagram and pinterest where users communicate via images has become popular as third generation service. Due to the fact that users communicate especially through images, image-based SNS utilizes images in different ways compared to other SNS. This study derived various characteristics of images in image-based SNS, and observed how users perceive each of them differently. Also, relationship between the characteristics and user satisfaction on image-based SNS is analyzed. The characteristics include 6 items; 'implicity', 'recordability', 'expressing identity', 'indirect experience', 'temporary amusement', and 'stimulating desire.' As a result of comparing user perception regarding those 6 characteristics, recordability and indirect experience were highly recognized than other characteristics. Also, according to users' age, motivation of using image-based SNS, and number of followers they have, users perceived each characteristic in different level. Finally, relationship between the characteristics and user satisfaction was analyzed, and the result showed that indirect experience had positive influence to user satisfaction. Regarding indirect experience, it was highly perceived by users and also had positive influence to their satisfaction, which means it is the most typical characteristic of image-based SNS.

• Journal of KIISE
• /
• v.42 no.6
• /
• pp.699-706
• /
• 2015

In recent years, as near-duplicate image has been increasing explosively by the spread of Internet and image-editing technology that allows easy access to image contents, related research has been done briskly. However, BoF (Bag-of-Feature), the most frequently used method for near-duplicate image detection, can cause problems that distinguish the same features from different features or the different features from same features in the quantization process of approximating a high-level local features to low-level. Therefore, a post-verification method for BoF is required to overcome the limitation of vector quantization. In this paper, we proposed and analyzed the performance of a post-verification method for BoF, which converts SIFT (Scale Invariant Feature Transform) descriptors into 128 bits binary codes and compares binary distance regarding of a short ranked list by BoF using the codes. Through an experiment using 1500 original images, it was shown that the near-duplicate detection accuracy was improved by approximately 4% over the previous BoF method.

• Journal of the Korean Society for Library and Information Science
• /
• v.45 no.2
• /
• pp.97-118
• /
• 2011

The study examines the interactive effect of spoken words and imagery not synchronized in audio/image surrogates for video gisting. To do that, we conducted an experiment with 64 participants, under the assumption that participants would better understand the content of videos when viewing audio/image surrogates rather than audio or image surrogates. The results of the experiment showed that overall audio/image surrogates were better than audio or image surrogates for video gisting, although the unsynchronized multimedia surrogates made it difficult for some participants to pay attention to both audio and image when the content they present is very different.

• Journal of Digital Convergence
• /
• v.10 no.2
• /
• pp.237-242
• /
• 2012

This paper proposed image matching technique that find out the truth of the wild ginseng through smart phone when the common person discovered a plant like a wild ginseng. Also sharing a location and information by SNS, we can improve the probability of wild ginseng discovery. Image matching technique using OpenCV porting in android finds out the truth of the wild ginseng with comparing existing it. Thus we are able to compare and analysis them in our application program. For more verification, we added marking function of wild ginseng position for information sharing between users.

• Journal of the HCI Society of Korea
• /
• v.3 no.1
• /
• pp.49-56
• /
• 2008

Most recent mobile devices can store a massive amount of images. However, the typical user interface of mobile devices, such as a small-size 2D display and discrete-input buttons, make the browsing and manipulation of images cumbersome and time-consuming. As an alternative, we adopt motion-based interaction along with a 3D layout of images, expecting such an intuitive and natural interaction may facilitate the tasks. We designed and implemented a motion-based interaction scheme for image browsing using an ultra mobile PC, and evaluated and compared its usability to that of the traditional button-based interaction. The effects of data layouts (tiled and fisheye cylindrical layouts) were also investigated to see whether they can enhance the effectiveness of the motion based interaction.

• 한국HCI학회:학술대회논문집
• /
• 2007.02a
• /
• pp.440-446
• /
• 2007

현재 대부분의 모바일 장치들엔 디지털 카메라가 설치되어 있으며 거대한 양의 이미지 데이터들을 저장할 수 있다. 이렇게 많은 양의 이미지가 저장되어 있을 경우 장치 속의 이미지들에 대한 브라우징을 하기 어려워지며, 그에 걸리는 시간도 증가하게 된다. 특히 모바일 장치의 경우 화면의 크기가 작으며, 일반 컴퓨터와 비교하여 부자연스럽고, 불편한 인터페이스를 갖고 있어 어려움을 더욱 증가 시킨다. 우리는 이를 해결하기 위해 3차원 가시화 방법과 모션 센싱을 이용한 인터페이스를 제안하고, 제안된 가시화 방법과 인터페이스의 조합을 통해 모바일 장치에서의 효과적인 이미지 브라우징 방법을 모색해 보았다.