• 한국통신학회논문지
• /
• 제22권3호
• /
• pp.481-492
• /
• 1997

This paper presents an algorithm for recognizing surface mount device(SMD) IC pattern based on the error back propoagation(EBP) neural network and discrete cosine transform(DCT). In this approach, we chose such parameters as frequency, angle, translation and amplitude for the shape informantion of SMD IC, which are calculated from the coefficient matrix of DCT. These feature parameters are normalized and then used for the input vector of neural network which is capable of adapting the surroundings such as variation of illumination, arrangement of objects and translation. Learning of EBP neural network is carried out until maximum error of the output layer is less then 0.020 and consequently, after the learning of forty thousand times, the maximum error have got to this value. Experimental results show that the rate of recognition is 100% in case of the random pattern taken at a similar circumstance as well as normalized training pattern. It also show that proposed method is not only relatively relatively simple compare with the traditional space domain method in extracting the feature parameter but also able to re recognize the pattern's class, position, and existence.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제5권1호
• /
• pp.228-246
• /
• 2011

A 3D model based approach for a face representation and recognition algorithm has been investigated as a robust solution for pose and illumination variation. Since a generative 3D face model consists of a large number of vertices, a 3D model based face recognition system is generally inefficient in computation time and complexity. In this paper, we propose a novel 3D face representation algorithm based on a pixel to vertex map (PVM) to optimize the number of vertices. We explore shape and texture coefficient vectors of the 3D model by fitting it to an input face using inverse compositional image alignment (ICIA) to evaluate face recognition performance. Experimental results show that the proposed face representation and recognition algorithm is efficient in computation time while maintaining reasonable accuracy.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2012년도 한국컴퓨터종합학술대회논문집 Vol.39 No.1(C)
• /
• pp.418-420
• /
• 2012

스테레오 매칭의 지역적 방법은 구현의 용이성과 낮은 계산복잡도로 인하여 많은 연구가 진행되고 있다. 하지만 대부분의 지역적 방법들은 영상이 외부환경에 변형되었을 때의 경우를 고려하지 않고 있기 때문에, 외부환경에 의해 많이 변형된 영상에 대해서는 제대로 된 변이 정보를 추출해내지 못한다. 본 논문에서는 양쪽 영상에서 곡선접합을 이용하여 서로 대응되는 영역을 찾는 스테레오 매칭 기법을 제안한다. 제안하는 기법은 조명과 같은 외부요소에 강인한 특징을 가진다. 이 기법은 전 처리나 후처리 과정에서 부가적인 작업의 수행 없이 기법 자체만으로 외부요소에 대한 보상을 실행한다는 면에서 장점을 가진다. 비록 다양한 영상에서 변이를 추출하는 실험 결과, 거시적인 특성을 반영하는 곡선접합만으로도 조명에 의해 변형된 영상에 대해서 변이결과를 추출해내었다. 차후 미시적인 방법과의 결합을 통해, 변이정보의 추출의 정확도를 올릴 수 있을 것 이라고 기대된다.

• Journal of Multimedia Information System
• /
• 제8권4호
• /
• pp.211-220
• /
• 2021

The process of inspecting SMDs on the PCB boards improves the product quality, performance and reduces frequent issues in this field. However, undesirable scenarios such as assembly failure and device breakdown can occur sometime during the assembly process and result in costly losses and time-consuming. The detection of these components with a model based on deep learning may be effective to reduce some errors during the inspection in the manufacturing process. In this paper, YOLO models were used due to their high speed and good accuracy in classification and target detection. A SMD detection and classification method using YOLO networks based on robust data preprocessing and augmentation techniques to deal with various types of variation such as illumination and geometric changes is proposed. For 9 different components of data provided from a PCB manufacturer company, the experiment results show that YOLOv4 is better with fast detection and classification than YOLOv3.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권2호
• /
• pp.871-891
• /
• 2019

Visual tracking is a challenging task that needs learning an effective model to handle the changes of target appearance caused by factors such as pose variation, illumination change, occlusion and motion blur. In this paper, a novel tracking algorithm based on weighted local sub-space reconstruction error is presented. First, accounting for the appearance changes in the tracking process, a generative weight calculation method based on structural reconstruction error is proposed. Furthermore, a template update scheme of occlusion-aware is introduced, in which we reconstruct a new template instead of simply exploiting the best observation for template update. The effectiveness and feasibility of the proposed algorithm are verified by comparing it with some state-of-the-art algorithms quantitatively and qualitatively.

• 미국학
• /
• 제44권1호
• /
• pp.155-182
• /
• 2021

This study explores how American war films modify or reenact American exceptionalism depending on political and social situations in the United States. To this end, it analyzes the two war films, Saving Private Ryan and We Were Soldiers that were released before and after the 9/11 attacks as a critical juncture in the twenty first century, respectively. While the former conducts a partial modification on American exceptionalism in order to restore the national identity and moral authority of the United States lost ever since the Vietnam War, the latter demonstrates a complete reenactment of American exceptionalism in accordance with the foreign policy of the Bush administration and neoconservatism. It concludes with the illumination that the examination of American war films within the framework of American exceptionalism is efficacious in understanding from a broad perspective how American exceptionalism is utilized depending on political and social situations in the United States.

• Current Optics and Photonics
• /
• 제7권5호
• /
• pp.537-544
• /
• 2023

Differential phase contrast (DPC) microscopy, a central quantitative phase imaging (QPI) technique in cell biology, facilitates label-free, real-time monitoring of intrinsic optical phase variations in biological samples. The existing DPC imaging theory, while important for QPI, is grounded in paraxial diffraction theory. However, this theory lacks accuracy when applied to high numerical aperture (NA) systems that are vital for high-resolution cellular studies. To tackle this limitation, we have, for the first time, formulated a nonparaxial DPC imaging equation with a transmission cross-coefficient (TCC) for high NA DPC microscopy. Our theoretical framework incorporates the apodization of the high NA objective lens, nonparaxial light propagation, and the angular distribution of source intensity or detector sensitivity. Thus, our TCC model deviates significantly from traditional paraxial TCCs, influenced by both NA and the angular variation of illumination or detection. Our nonparaxial imaging theory could enhance phase retrieval accuracy in QPI based on high NA DPC imaging.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제8권10호
• /
• pp.3490-3507
• /
• 2014

License plate detection (LPD) is one of the most important aspects of an automatic license plate recognition system. Although there have been some successful license plate recognition (LPR) methods in past decades, it is still a challenging problem because of the diversity of plate formats and outdoor illumination conditions in image acquisition. Because the accurate detection of license plates under different conditions directly affects overall recognition system accuracy, different methods have been developed for LPD systems. In this paper, we propose a license plate detection method that is rapid and robust against variation, especially variations in illumination conditions. Taking the aspects of accuracy and speed into consideration, the proposed system consists of two stages. For each stage, Haar-like features are used to compute and select features from license plate images and a cascade classifier based on the concatenation of classifiers where each classifier is trained by an AdaBoost algorithm is used to classify parts of an image within a search window as either license plate or non-license plate. And it is followed by connected component analysis (CCA) for eliminating false positives. The two stages use different image preprocessing blocks: image preprocessing without adaptive thresholding for the first stage and image preprocessing with adaptive thresholding for the second stage. The method is faster and more accurate than most existing methods used in LPD. Experimental results demonstrate that the LPD rate is 98.38% and the average computational time is 54.64 ms.

• 한국인터넷방송통신학회논문지
• /
• 제13권2호
• /
• pp.171-179
• /
• 2013

영상처리에서 인식에 관련된 기술들은 환경에 아주 많은 영향을 받게 되는데 이러한 인식률을 결정짓는 요소 중인 파라미터는 환경에 적절한 값을 얼마나 잘 선택하느냐에 따라서 인식률의 큰 차이를 보인다. 본 논문은 눈동자 추적 알고리즘이 사람이나 실험 환경의 변화에 따라 인식률이 저하되는 현상을 보완하기 위한 성능 향상 및 환경에 적응하는 시스템의 구현에 대한 방법이다. 최적의 파라미터를 얻기 위해 전 처리에 사용되는 이진화 알고리즘의 문턱값을 학습이 필요한 시기를 적절히 판단해 강화학습을 이용하여 다시 학습시켜 인식률을 향상시키는 방법을 사용했다. 실험데이터를 수집하기 위해 입력 장치는 가격이 저렴하고 일반적인 웹 카메라를 사용 하였으며 얼굴 영역에 해당하는 많은 양의 이미지를 수집하여 강화학습의 적응력을 실험하였다. 이미지의 그룹을 다양하게 변화시켜 실험한 결과 강화학습을 사용한 경우 그렇지 않은 경우에 비해 작게는 3% 많게는 14%가량의 성능이 향상됨을 확인하였다. 이렇게 성능이 향상된 눈동자 추적 시스템은 휴먼 컴퓨터 인터랙션 분야에 효과적으로 활용될 수 있을 것이다.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제2권9호
• /
• pp.643-650
• /
• 2013

본 논문은 스마트폰 영상의 실시간 눈동자 검출에서 허프 원 변환 연산의 연산량 축소를 통한 속도 및 검출율 개선 방법을 제안한다. 눈동자를 검출하기 위해서는 입력 영상에서 얼굴과 눈을 검출하고, 눈 영역의 크기에 따라 눈동자의 크기가 변하는 것을 방지하기 위해 일정크기로 눈 영역을 정규화하며, 다양한 조명환경에서 눈동자가 검출이 가능하도록 히스토그램 평활화를 실시하고, 눈의 양쪽 끝점간의 거리를 구하여 영상에서의 실제 눈동자의 크기를 포함할 수 있는 최소한의 눈동자 크기 범위를 계산하여 허프 원 변환에 적용함으로써 연산량을 최소화 하였다. 제안한 방법을 밝은 조명과 어두운 조명에서 실험한 결과 기존 방법들과 비교하여 눈동자 검출 속도는 40% 이상, 검출율은 14% 이상 향상된 것을 보였다.