• 제목/요약/키워드: Image Representation

검색결과 785건 처리시간 0.023초

효율적인 이미지 검색 시스템을 위한 자기 감독 딥해싱 모델의 비교 분석 (Comparative Analysis of Self-supervised Deephashing Models for Efficient Image Retrieval System)

  • 김수인;전영진;이상범;김원겸
    • 정보처리학회논문지:소프트웨어 및 데이터공학
    • /
    • 제12권12호
    • /
    • pp.519-524
    • /
    • 2023
  • 해싱 기반 이미지 검색에서는 조작된 이미지의 해시코드가 원본 이미지와 달라 동일한 이미지 검색이 어렵다. 본 논문은 이미지의 질감, 모양, 색상 등 특징 정보로부터 지각적 해시코드를 생성하는 자기 감독 기반 딥해싱 모델을 제안하고 평가한다. 비교 모델은 오토인코더 기반 변분 추론 모델들이며, 인코더는 완전 연결 계층, 합성곱 신경망과 트랜스포머 모듈 등으로 설계된다. 제안된 모델은 기하학적 패턴을 추출하고 이미지 내 위치 관계를 활용하는 SimAM 모듈을 포함하는 변형 추론 모델이다. SimAM은 뉴런과 주변 뉴런의 활성화 값을 이용한 에너지 함수를 통해 객체 또는 로컬 영역이 강조된 잠재 벡터를 학습할 수 있다. 제안 방법은 표현 학습 모델로 고차원 입력 이미지의 저차원 잠재 벡터를 생성할 수 있으며, 잠재 벡터는 구분 가능한 해시코드로 이진화 된다. CIFAR-10, ImageNet, NUS-WIDE 등 공개 데이터셋의 실험 결과로부터 제안 모델은 비교 모델보다 우수하며, 지도학습 기반 딥해싱 모델과 동등한 성능이 분석되었다.

시점을 달리한 여러 장의 투영 영상을 이용한 반투명 재질의 실시간 렌더링 (Representation of Translucent Objects using Multiple Projection Images for Real-time Rendering)

  • 이재영;김강연;유재덕;이관행
    • 한국HCI학회:학술대회논문집
    • /
    • 한국HCI학회 2006년도 학술대회 1부
    • /
    • pp.878-883
    • /
    • 2006
  • 반투명 물체(Translucent Object)는 불투명한 물체와는 달리 물체 내부에서 산란이 일어난다. 반투명 물체의 한 표면(Surface)을 렌더링하기 위해서는 그 표면의 정규 벡터뿐만 아니라 그 표면의 주변 기하 정보가 필요하다. 그러나 그래픽 하드웨어 구조는 반투명 물체의 실시간 렌더링의 구현에 많은 제약을 준다. 3D 기하 정보 대신에 라디언스 맵(Radiance map)과 깊이 맵(Depth map)과 같은 투영 영상(Projected Image)을 기반으로 하는 영상 공간 접근 방법(Image Space Approach)을 사용함으로써 GPU 상에서 반투명 재질을 실시간으로 표현할 수 있다. 본 논문에서는 영상 공간 접근 방법(Image Space Approach)의 연장선에서 시점을 달리한 여러 장의 투영 영상을 이용함으로써 기존의 한 장의 투영 영상만을 이용한 방법이 가지고 있는 가시성 한계점을 해결한다. 또한 복수 투영 영상의 이용에 따른 계산량 증가에 의해서 손실된 프레임 속도(Frame Rate)에 대해 분석한다.

  • PDF

Effective Fragile Watermarking for Image Authentication with High-quality Recovery Capability

  • Qin, Chuan;Chang, Chin-Chen;Hsu, Tai-Jung
    • KSII Transactions on Internet and Information Systems (TIIS)
    • /
    • 제7권11호
    • /
    • pp.2941-2956
    • /
    • 2013
  • In this paper, we propose an effective fragile image watermarking scheme for tampering detection and content recovery. Cover image is divided into a series of non-overlapping blocks and a block mapping relationship is constructed by the secret key. Several DCT coefficients with direct current and lower frequencies of the MSBs for each block are used to generate the reference bits, and different coefficients are assigned with different bit numbers for representation according to their importance. To enhance recovery performance, authentication bits are generated by the MSBs and the reference bits, respectively. After LSB substitution hiding, the embedded watermark bits in each block consist of the information of itself and its mapping blocks. On the receiver side, all blocks with tampered MSBs can be detected and recovered using the valid extracted reference bits. Experimental results demonstrate the effectiveness of the proposed scheme.

비젼에 의한 감성인식 (Emotion Recognition by Vision System)

  • 이상윤;오재흥;주영훈;심귀보
    • 한국지능시스템학회:학술대회논문집
    • /
    • 한국퍼지및지능시스템학회 2001년도 추계학술대회 학술발표 논문집
    • /
    • pp.203-207
    • /
    • 2001
  • In this Paper, we propose the neural network based emotion recognition method for intelligently recognizing the human's emotion using CCD color image. To do this, we first acquire the color image from the CCD camera, and then propose the method for recognizing the expression to be represented the structural correlation of man's feature Points(eyebrows, eye, nose, mouse) It is central technology that the Process of extract, separate and recognize correct data in the image. for representation is expressed by structural corelation of human's feature Points In the Proposed method, human's emotion is divided into four emotion (surprise, anger, happiness, sadness). Had separated complexion area using color-difference of color space by method that have separated background and human's face toughly to change such as external illumination in this paper. For this, we propose an algorithm to extract four feature Points from the face image acquired by the color CCD camera and find normalization face picture and some feature vectors from those. And then we apply back-prapagation algorithm to the secondary feature vector. Finally, we show the Practical application possibility of the proposed method.

  • PDF

삼중 주기적 최소곡면을 이용한 조직공학을 위한 생체모사 스캐폴드의 컴퓨터응용 설계 및 제작 (Computer-aided Design and Fabrication of Bio-mimetic Scaffold for Tissue Engineering Using the Triply Periodic Minimal Surface)

  • 유동진
    • 한국정밀공학회지
    • /
    • 제28권7호
    • /
    • pp.834-850
    • /
    • 2011
  • In this paper, a novel tissue engineering scaffold design method based on triply periodic minimal surface (TPMS) is proposed. After generating the hexahedral elements for a 3D anatomical shape using the distance field algorithm, the unit cell libraries composed of triply periodic minimal surfaces are mapped into the subdivided hexahedral elements using the shape function widely used in the finite element method. In addition, a heterogeneous implicit solid representation method is introduced to design a 3D (Three-dimensional) bio-mimetic scaffold for tissue engineering from a sequence of computed tomography (CT) medical image data. CT image of a human spine bone is used as the case study for designing a 3D bio-mimetic scaffold model from CT image data.

2진 영상의 고속 세선화 장치 구현에 관한 연구 (A Study on Fast Thinning Unit Implementation of Binary Image)

  • 허윤석;이재춘;곽윤식;이대영
    • 대한전자공학회논문지
    • /
    • 제27권5호
    • /
    • pp.775-783
    • /
    • 1990
  • In this paper we implemented the fast thinning unit by modifying the pipeline architecture which was proposed by Stanley R. Sternberg. The unit is useful in preprocessing such as image representation and pattern recognition etc. This unit is composed of interface part, local memory part, address generation part, thinning processing part and control part. In thinning processing part, we shortened the thinning part which performed by means of look up table using window mapping table. Thus we improved the weakness of SAP, in which the number of delay pipeline and window pipeline are equal to image column size. Two independent memorys using tri-state buffer enable the two direction flow of address generated by address generation part. This unit avoids the complexity of architecture and has flexibility of image size by means of simple modification of logic bits.

  • PDF

A Review on Image Feature Detection and Description

  • Truong, Mai Thanh Nhat;Kim, Sanghoon
    • 한국정보처리학회:학술대회논문집
    • /
    • 한국정보처리학회 2016년도 추계학술발표대회
    • /
    • pp.677-680
    • /
    • 2016
  • In computer vision and image processing, feature detection and description are essential parts of many applications which require a representation for objects of interest. Applications like object recognition or motion tracking will not produce high accuracy results without good features. Due to its importance, research on image feature has attracted a significant attention and several techniques have been introduced. This paper provides a review on well-known image feature detection and description techniques. Moreover, two experiments are conducted for the purpose of evaluating the performance of mentioned techniques.

Image Feature Representation Using Code Vectors for Retrieval

  • ;조혜;박종안;박승진;양원일
    • 한국ITS학회 논문지
    • /
    • 제8권3호
    • /
    • pp.122-130
    • /
    • 2009
  • The paper presents an algorithm which uses code vectors to represent comer geometry information for searching the similar images from a database. The comers have been extracted by finding the intersections of the detected lines found using Hough transform. Taking the comer as the center coordinate, the angles of the intersecting lines are determined and are represented using code vectors. A code book has been used to code each comer geometry information and indexes to the code book are generated. For similarity measurement, the histogram of the code book indexes is used. This result in a significant small size feature matrix compared to the algorithms using color features. Experimental results show that use of code vectors is computationally efficient in similarity measurement and the comers being noise invariant produce good results in noisy environments.

  • PDF

Multiple Description Coding Using Directional Discrete Cosine Transform

  • Lama, Ramesh Kumar;Kwon, Goo-Rak
    • Journal of information and communication convergence engineering
    • /
    • 제11권4호
    • /
    • pp.293-297
    • /
    • 2013
  • Delivery of high quality video over a wide area network with large number of users poses great challenges for the video communication system. To ensure video quality, multiple descriptions have recently attracted various attention as a way of encoding and visual information delivery over wireless network. We propose a new efficient multiple description coding (MDC) technique. Quincunx lattice sub-sampling is used for generating multiple descriptions of an image. In this paper, we propose the application of a directional discrete cosine transform (DCT) to a sub-sampled quincunx lattice to create an MDC representation. On the decoder side, the image is decoded from the received side information. If all the descriptions arrive successfully, the image is reconstructed by combining the descriptions. However, if only one side description is received, decoding is executed using an interpolation process. The experimental results show that such the directional DCT can achieve a better coding gain as well as energy packing efficiency than the conventional DCT with re-alignment.

퍼지 클러스터링과 스트링 매칭을 통합한 형상 인식법 (Pattern Recognition Method Using Fuzzy Clustering and String Matching)

  • 남원우;이상조
    • 대한기계학회논문집
    • /
    • 제17권11호
    • /
    • pp.2711-2722
    • /
    • 1993
  • Most of the current 2-D object recognition systems are model-based. In such systems, the representation of each of a known set of objects are precompiled and stored in a database of models. Later, they are used to recognize the image of an object in each instance. In this thesis, the approach method for the 2-D object recognition is treating an object boundary as a string of structral units and utilizing string matching to analyze the scenes. To reduce string matching time, models are rebuilt by means of fuzzy c-means clustering algorithm. In this experiments, the image of objects were taken at initial position of a robot from the CCD camera, and the models are consturcted by the proposed algorithm. After that the image of an unknown object is taken by the camera at a random position, and then the unknown object is identified by a comparison between the unknown object and models. Finally, the amount of translation and rotation of object from the initial position is computed.