• Title/Summary/Keyword: 이미지 변형

Search Result 428, Processing Time 0.023 seconds

Name card region detection scheme for name card recognition application based on android platform (안드로이드 플랫폼 기반 명함 인식 어플리케이션을 위한 명함 영역 검출 기법)

  • Lee, JeYul;Lee, KyuWon
    • Proceedings of the Korea Information Processing Society Conference
    • /
    • 2014.04a
    • /
    • pp.844-847
    • /
    • 2014
  • 본 논문에서는 다양한 형태의 어플리케이션 중 스마트폰에 탑재된 카메라를 이용하여 명함을 인식할 때 발생하는 문제점을 해결하기 위한 기법을 제시하고자 한다. 스마트폰의 카메라를 이용하여 이미지를 얻을 경우 카메라의 각도에 따라 객체의 모양이 변형된다. 명함인식에서 이러한 이미지 왜곡문제는 인식률에 많은 영향을 미친다. 본 논문에서는 카메라의 각도에 따른 이미지의 왜곡 문제를 해결하기 위해 캐니 에지를 이용하여 객체를 검출하고, 검출된 객체의 간단한 정보를 이용하여 명함 영역을 검출하고, 영역 내에 명함을 검출하는 기법을 제안한다.

Digital Watermarking Techniques Robust to Distortion Attacks (왜곡 공격에 강인한 디지털 워터마킹 기법)

  • Su-Kyoung Kim;Yu-ran Jeon;Jung-Hwa Ryu;Il-Gu Lee
    • Proceedings of the Korea Information Processing Society Conference
    • /
    • 2024.05a
    • /
    • pp.345-346
    • /
    • 2024
  • 디지털 기술과 정보통신 기술이 발전하면서 디지털 콘텐츠의 불법복제 및 유통으로 인한 저작권 침해 피해가 증가하고 있다. 저작권 침해 문제를 예방하기 위해 다양한 디지털 워터마킹 기술이 제안되었지만, 디지털 이미지 워터마킹은 이미지에 기하학적 변형을 가하면 삽입된 워터마크가 훼손되어 탐지가 어렵다는 문제가 있다. 본 연구에서는 왜곡 공격에 강인한 상관관계 측정 기반 워터마킹 기법을 제안한다. 제안한 방식은 교차 상관 기법을 이용해 이미지와 워터마크의 상관관계를 계산하고 임계값과 비교하여 공간 영역에서의 비가시성 워터마크의 존재 여부를 검증할 수 있는 디지털 워터마킹 방법이다. 실험 결과에 따르면 표준편차 120의 가우시안 노이즈 공격을 가해도 원본 워터마크와 0.1 이상의 상관관계를 보이며, 종래의 방식보다 높은 탐지 성능을 나타냈다.

  • PDF

Showing Morphological Evolution of the Strain Response Envelope of Clay with Fourier Descriptor Analysis (퓨리에 기술자를 이용한 점성토의 변형률 응답 곡선의 형상 변이 분석)

  • Kim, Taesik;Jung, Young-Hoon
    • Journal of the Korean GEO-environmental Society
    • /
    • v.18 no.3
    • /
    • pp.25-30
    • /
    • 2017
  • This paper introduces a novel method to quantify the morphological evolution of the strain response envelope. The strain response envelope is defined as an image in strain increment space corresponding to the unit stress input in stress space. Based on the shape of strain response envelopes, the deformation characteristics of soils can be described using the framework of elastic-plastic theory. Fourier descriptor analysis was used to investigate the morphological characteristics of strain response envelopes. The numerical results show that when the stress input remains in the initial yield surface the Fourier descriptors remain constant. Once the stress input crosses the initial yield surface, every descriptors deals in this study change. Numerical and experimental results of this study show that clear yielding response is only found in natural block samples. Among the Fourier descriptors, the descriptor called as asymmetry is the best for detecting the yield and is minimally sensitive to the number of input stress paths.

Synthesis of Realistic Facial Expression using a Nonlinear Model for Skin Color Change (비선형 피부색 변화 모델을 이용한 실감적인 표정 합성)

  • Lee Jeong-Ho;Park Hyun;Moon Young-Shik
    • Proceedings of the Korean Information Science Society Conference
    • /
    • 2006.06a
    • /
    • pp.121-123
    • /
    • 2006
  • 얼굴의 표정은 얼굴의 구성요소 같은 기하학적 정보와 조명이나 주름 같은 세부적인 정보들로 표현된다. 얼굴 표정은 기하학적 변형만으로는 실감적인 표정을 생성하기 힘들기 때문에 기하학적 변형과 더불어 텍스쳐 같은 세부적인 정보도 함께 변형해야만 실감적인 표현을 할 수 있다. 표정비율이미지 (Expression Ratio Image)같은 얼굴 텍스처의 세부적인 정보를 변형하기 위한 기존 방법들은 조명에 따른 피부색의 변화를 정확히 표현할 수 없는 단점이 있다. 따라서 본 논문에서는 이러한 문제를 해결하기 위해 서로 다른 조명 조건에서도 실감적인 표정 텍스처 정보를 적용할 수 있는 비선형 피부색 모델 기반의 표정 합성 방법을 제안한다. 제안된 방법은 동적 외양 모델을 이용한 자동적인 얼굴 특징 추출과 와핑을 통한 표정 변형 단계, 비선형 피부색 변화 모델을 이용한 표정 생성 단계, Euclidean Distance Transform (EDT)에 의해 계산된 혼합 비율을 사용한 원본 얼굴 영상과 생성된 표정의 합성 등 총 3 단계로 구성된다. 실험결과는 제안된 방법이 다양한 조명조건에서도 자연스럽고 실감적인 표정을 표현한다는 것을 보인다.

  • PDF

A Study on the Digital Image Creating Through Emotional Dimension (감성어휘을 통한 디지털이미지 크리에이팅에 관한 연구)

  • Park Sang-Jin;Kwak Hoon-Sung;Seo Mi-Ra
    • The Journal of the Korea Contents Association
    • /
    • v.6 no.4
    • /
    • pp.89-97
    • /
    • 2006
  • Such changes that the digital paradigm brought propels a change of the value system itself in the existing society so that a way of thinking that differentiates oneself from others is getting much stronger. As public interests are higher in the image production on the internet, such authoring environments as development of related technologies, expansibility of information, and systematization of available information continue to change. Therefore, a variety of unique image presentations that fit to the changed environments are necessary and an expression method is indispensable for such image production. This study focuses on digital image creating in which a synectics method is applied. It suggests digital image application methods and examples that contain not just simple image distortion or variation but intrinsic significance. It is believed that it will be of good use to produce a variety of unique and innate digital images in the end.

  • PDF

Design and Fabrication of Split Hopkinson Pressure Bar for Dynamic Mechanical Properties of Self-reinforced Polypropylene Composite (폴리프로필렌 자기 보강 복합재의 동적 물성 구축을 위한 Split Hopkinson Pressure Bar의 설계 및 제작)

  • Kang, So-Young;Kim, Do-Hyoung;Kim, Dong-Hyun;Kim, Hak-Sung
    • Composites Research
    • /
    • v.31 no.5
    • /
    • pp.221-226
    • /
    • 2018
  • The Split Hopkinson Pressure Bar(SHPB) has been the most widely used apparatus to characterize dynamic mechanical behavior of materials at high strain rates between $100s^{-1}$ and $10,000s^{-1}$. The SHPB test is based on the wave propagation theory which was developed to give the stress, strain and strain rate in the specimen using the strains measured in the incident and transmission bars. In this study, the SHPB was directly designed and fabricated for the dynamic mechanical properties of fiber reinforced plastic (FRP) composites. In addition, this apparatus was verified for the validity by comparing the strain data obtained through the high speed camera and Digital Image Correlation(DIC) during the high strain rate compression test of the self-reinforced polypropylene composite (SRPP) specimen.

A Study on Modified Median Filter for Impulse Noise Removal (임펄스 잡음 제거를 위한 변형된 메디안 필터에 관한 연구)

  • Lee, Kyung-Hyo;Kim, Nam-Ho
    • Journal of the Korea Institute of Information and Communication Engineering
    • /
    • v.13 no.2
    • /
    • pp.376-381
    • /
    • 2009
  • The image data compression, recognition, restoration, etc. are parts of the digital image processing technology. In the process by various devices, noises would be made. Because the noise could damage the image, we use the image filter to preserve the original image from the noise. The image filter used in digital image process basically has a two-dimensional structure. There an two methods of creating the filter - One is reiterating one dimension and the other is using an indivisible two-dimension image filter. The image filter is being widely used along with one-dimension filter according to each noise, and various median filters are being used to remove the impulse noise. In this paper, I suggested a powerful modified median filter, and compared with conventional filters for objective verification.

A Study on School Uniform Reforming Behavior in Relation to Body Image:In the Case of High School Girls (신체이미지와 교복변형행동 연구: 여고생을 중심으로)

  • Wee, Eun-Hah;Kwon, Yu-Min
    • Journal of the Korean Home Economics Association
    • /
    • v.47 no.10
    • /
    • pp.73-85
    • /
    • 2009
  • The purpose of this study was to examine whether uniform reforming behavior in high school girls is related to body image distortion, and to obtain objective data to help in guiding students in the wearing of uniforms at school. 661 freshmen and junior girls at six high schools in Kwangju city responded to questionnaires. The data was analyzed by one-way ANOVA and Pearson’s product-moment correlations using the SPSS 12.0 program. According to their body satisfaction level, three groups (haggard distortion, normal, overweight distortion) were identified. The data showed that a few high school girls were unable to recognize their body types correctly. It also indicated that school uniform reforming behaviors in high school girls were not related to their body image or body satisfaction level. So, it seems that students need guidance in the building up of an objective body image and wearing uniforms to fit their body types.

Detectability of Pore Defect in Wind Turbine Blade Composites Using Image Correlation Technique (이미지 상관 기법을 이용한 풍력 발전 블레이드용 복합재료의 기공 결함 검출능)

  • Kim, Jong Il;Huh, Yong Hak;Lee, Gun Chang
    • Transactions of the Korean Society of Mechanical Engineers A
    • /
    • v.37 no.10
    • /
    • pp.1201-1206
    • /
    • 2013
  • Defects that occur during the manufacturing process or operation of a wind turbine blade have a great influence on its life and safety. Typically, defects such as delamination, pore, wrinkle and matrix crack are found in a blade. In this study, the detectability of the pores, a type of defect that frequently occur during manufacturing, was examined from the full field strain distribution determined with the image correlation technique. Pore defects were artificially introduced in four-ply laminated GFRP composites with $0^{\circ}/{\pm}45^{\circ}$ fiber direction. The artificial pores were introduced in consideration of their size and location. Three different-sized pores with diameter of 1, 2 and 3 mm were located on the top and bottom surface and embedded. By applying static loads of 0-200 MPa, the strain distributions over the specimen with the pore defects were determined using image correlation technique. It was found the pores with diameter exceeding 2 mm can be detected in diameter.

Performance Evaluation of VTON (Virtual-Try-On) Algorithms using a Pair of Cloth and Human Image (이미지를 사용한 가상의상착용 알고리즘들의 성능 분석)

  • Tuan, Thai Thanh;Minar, Matiur Rahman;Ah, Heejune
    • Journal of Korea Society of Industrial Information Systems
    • /
    • v.24 no.6
    • /
    • pp.25-34
    • /
    • 2019
  • VTON (Virtual try-on) is a key technology that can activate the online commerce of fashion items. However, the early 3D graphics-based methods require the 3D information of the clothing or the human body, which is difficult to secure realistically. In order to overcome this problem, Image-based deep-learning algorithms such as VITON (Virtual image try-on) and CP-VTON (Characteristic preserving-virtual try-on) has been published, but only a sampled results on performance is presented. In order to examine the strength and weakness for their commercialization, the performance analysis is needed according to the complexity of the clothes, the object posture and body shape, and the degree of occlusion of the clothes. In this paper, IoU and SSIM were evaluated for the performance of transformation and synthesis stages, together with non-DL SCM based method. As a result, CP-VTON shows the best performance, but its performance varies significantly according to posture and complexity of clothes. The reasons for this were attributed to the limitations of secondary geometric deformation and the limitations of the synthesis technology through GAN.