• 한국생산제조학회지
• /
• 제21권1호
• /
• pp.76-84
• /
• 2012

This paper presents the measurement system of arrow's point of impact using laser scan camera and describes the image calibration method. The calibration process of distorted image is primarily divided into explicit and implicit method. Explicit method focuses on direct optical property using physical camera and its parameter adjustment functionality, while implicit method relies on a calibration plate which assumed relations between image pixels and target positions. To find the relations of image and target position in implicit method, we proposed the performance criteria based polynomial theorem model that overcome some limitations of conventional image calibration model such as over-fitting problem. The proposed method can be verified with 2D position of arrow that were taken by SICK Ranger-D50 laser scan camera.

• 전기전자학회논문지
• /
• 제26권4호
• /
• pp.639-646
• /
• 2022

딥러닝의 발전으로 딥러닝 모델들이 이미지 인식, 음성 인식 등 여러 분야에서 활발하게 사용 중이다. 하지만 이 딥러닝을 효과적으로 사용하기 위해서는 대형 데이터 세트가 필요하지만 이를 구축하기에는 많은 시간과 노력 그리고 비용이 필요하다. 본 논문에서는 웹 크롤링이라는 이미지 수집 방법을 통해서 이미지를 수집하고 데이터 전처리 과정을 거쳐 이미지 분류 모델에 사용할 수 있게 데이터 세트를 구축한다. 더 나아가 전이학습을 이미지 분류 모델에 접목해 카테고리값을 넣어 자동으로 이미지를 분류할 수 있는 경량화된 모델과 적은 훈련 시간 및 높은 정확도를 얻을 수 있는 이미지 분류 모델을 제안한다.

• 한국정보통신학회:학술대회논문집
• /
• 한국해양정보통신학회 2001년도 춘계종합학술대회
• /
• pp.633-637
• /
• 2001

본 논문은 모델기반 다중이동물체 추적을 위한 모델생성 알고리즘을 제안하였다. 제안한 알고리즘은 배경영상에 이동물체가 초기 진입했을 때의 초기모델생성 단계와 이동물체 추적 단계에서의 모델 갱신 단계로 구분하였다. 초기모델생성 단계에서는 차영상과 클러스터링 기법을 이용하여 분할된 분할영상과 현재프레임 영상에 대한 윤곽선 영상과의 로직 AND 연산을 수행하여 초기모델을 생성하였다. 모델갱신 단계에서는 하우스돌프 거리(Hausdorff Distance)와 2D-Logarithmic 탐색 알고리즘을 이용하여 추적중인 이동물체의 형태변화에 적응할 수 있도록 매 프레임 마다 새로운 모델을 갱신하였다. 실험은 도로에서 주행하는 자동차를 대상으로 도_의 실험을 수행하였다. 그 결과 도로에서 주행하는 자동차의 진입방향과 추적 대상 수가 불규칙한 경우에도 모델생성이 98% 이상 이루어짐을 알 수 있었다.

• 동력기계공학회지
• /
• 제15권4호
• /
• pp.25-30
• /
• 2011

This study is made to provide with a method for correcting the geometric distortion of the digital radiography image by analytical approach based upon the inverse square law and Beer's law. This study is aimed to find out and improve a mathematic model of nonlinear type. Variations in the alignment of the X-ray source, the object, and imaging plate affect digital radiography images. A model which is expressed in parameter values; e.g, angle, position, absorption coefficient, length, width and pixel account of radiography source, is developed so as to match the sample image. For the best correction of the digital image that is the most similar to the model image, a correction technique based upon tangent is developed; then applied to the digital radiography images of steel tubes. As a result, the image correction is confirmed to be made successfully.

• 패션비즈니스
• /
• 제8권6호
• /
• pp.103-112
• /
• 2004

A Politician Appearance Acceptance Model (PAAM model) was formed and designed based on an analysis of how the electorate would accept a female politician. The PAAM model evaluated factors which influenced the voter's view of the female politician based on appearance. Causative factors were assessed that impacted acceptance based on appearance and analyzed whether voting was influenced by the appearance image; appearance image preferences for a female politician included the classic, dramatic, romantic and natural images. Through validations, the appearance image and competency had a causative factor that contributed to the acceptance of the politician image. The Classic Image demonstrated the strongest and most important image among the appearance images. As voters were more interested in the appearance image of a female politician, more emphasis and weight was on the appearance image during the voting selection process.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2021년도 춘계학술대회
• /
• pp.338-339
• /
• 2021

본 논문에서는 변환 학습을 기반으로 한 다중 클래스 이미지 장면 분류 방법을 제안한다. 이미지 분류를 위해 대형 이미지 데이터 세트 ImageNet에 대해 사전 학습 한 ResNet (ResNet) 모델을 사용하는 방법이다. CNN 모델의 이미지 분류 방법에 비해 분류 정확도 및 효율성을 크게 향상시킬 수 있다.

• The Journal of the Acoustical Society of Korea
• /
• 제17권3E호
• /
• pp.27-34
• /
• 1998

The validity of the image model is investigated both analytically and experimentally in a half space with an infinite single reflecting surface present. This paper exploits the Sommerfeld integral that represents the exact solution for the reflected field in the half space. The solution is shown to be obtained by direct numerical integration which yields more accurate and stable results. The predicted results from the image model are compared to those from the direct numerical integration of the Sommerfeld integral. It is also experimentally demonstrated that the image model gives acceptably accurate results. It is of significance that this paper reveals analytical and experimental validation of using the image model except near-grazing incidence.

• International Journal of Computer Science & Network Security
• /
• 제24권2호
• /
• pp.1-14
• /
• 2024

While text-to-image models have made remarkable progress in image synthesis, certain models, particularly generative diffusion models, have exhibited a noticeable bias to- wards generating images related to the culture of some developing countries. This paper introduces an empirical investigation aimed at mitigating the bias of image generative model. We achieve this by incorporating symbols representing Saudi culture into a stable diffusion model using the Dreambooth technique. CLIP score metric is used to assess the outcomes in this study. This paper also explores the impact of varying parameters for instance the quantity of training images and the learning rate. The findings reveal a substantial reduction in bias-related concerns and propose an innovative metric for evaluating cultural relevance.

• ETRI Journal
• /
• 제42권1호
• /
• pp.67-77
• /
• 2020

Image captioning has received significant interest in recent years, and notable results have been achieved. Most previous approaches have focused on generating visual descriptions from images, whereas a few approaches have exploited visual descriptions for image classification. This study demonstrates that a good performance can be achieved for both description generation and image classification through an end-to-end joint learning approach with a loss function, which encourages each task to reach a consensus. When given images and visual descriptions, the proposed model learns a multimodal intermediate embedding, which can represent both the textual and visual characteristics of an object. The performance can be improved for both tasks by sharing the multimodal embedding. Through a novel loss function based on class activation mapping, which localizes the discriminative image region of a model, we achieve a higher score when the captioning and classification model reaches a consensus on the key parts of the object. Using the proposed model, we established a substantially improved performance for each task on the UCSD Birds and Oxford Flowers datasets.

• 유통과학연구
• /
• 제18권4호
• /
• pp.13-26
• /
• 2020

Purpose: In this study, we will explore how the attributes of the airline's advertising model affect the corporate image and brand loyalty and the medium effect of the corporate image. Research design, data and methodology: Data collection for empirical analysis of this study was conducted online for about seven months from Jan. 2 to July 12, 2019, and was confirmed as part 292 of the final effective sample and used for demonstration analysis. Results: The property of the advertising model shown to have a significant impact in corporate image and brand loyalty. The property of the advertising, reliability and professionalism shown to have an impact in the social responsibility, but attractiveness is its responsibility and brand loyalty. Corporate images have been shown to play a meaningful role in the impact of advertising models on brand loyalty. Conclusions: The attributes of the airline's advertising model are divided into four categories, and reliability has the most influence on the image of a company and the formation of brand loyalty. The impact of the attributes of the advertising model on the relationship between corporate image and brand loyalty was investigated through an empirical analysis, and several implications were derived.