• 한국소성가공학회:학술대회논문집
• /
• 한국소성가공학회 2007년도 추계학술대회 논문집
• /
• pp.100-103
• /
• 2007

A microlens array has been required to improve light conversion efficiency in image sensors. A microlens array can be usually fabricated by photoresist reflow, hot-embossing, micro injection molding, and UV-imprinting. Among these processes, a UV-imprinting, which is operated at room temperature with relatively low applied pressure, can be a desirable process to integrate microlens array on image sensors, because this process provides the components with low thermal expansion, enhanced stability, and low birefringence, furthermore, it is more suitable for mass production of high quality microlens array. In this study, to analyze the optical properties of the wafer scale microlens array integrated image sensor, another wafer scale simulated image sensor chip array was designed and fabricated. An aspherical square microlens was designed and integrated on a simulated image sensor chip array using a UV-imprinting process. Finally, the optical performances were measured and analyzed.

• 한국멀티미디어학회논문지
• /
• 제22권9호
• /
• pp.1000-1010
• /
• 2019

Image semantic segmentation and dehazing are key tasks in the computer vision. In recent years, researches in both tasks have achieved substantial improvements in performance with the development of Convolutional Neural Network (CNN). However, most of the previous works for semantic segmentation assume the images are captured in clear weather and show degraded performance under hazy images with low contrast and faded color. Meanwhile, dehazing aims to recover clear image given observed hazy image, which is an ill-posed problem and can be alleviated with additional information about the image. In this work, we propose a deep multi-task network for simultaneous semantic segmentation and dehazing. The proposed network takes single haze image as input and predicts dense semantic segmentation map and clear image. The visual information getting refined during the dehazing process can help the recognition task of semantic segmentation. On the other hand, semantic features obtained during the semantic segmentation process can provide cues for color priors for objects, which can help dehazing process. Experimental results demonstrate the effectiveness of the proposed multi-task approach, showing improved performance compared to the separate networks.

• 한국정밀공학회지
• /
• 제12권4호
• /
• pp.67-77
• /
• 1995

In modern industry the requirement of automation of manufacturing process increases so that unmanned system has been popular as an ultimate goal of modern manufacturing process. In unmanned manufacturing process the tool fracture is a very serious problem because it results in the damage of workpieces and can stop the operation of whole manufa- turing system. In this study, image processing technique is used to detect the fracture of insert tip of face milling using multi-point-tool. In order to acquire the image information of fracture shape of rotation insert tip. We set up the optical system using a light beam chopper. In this system we can reduce the image degradation generated from stopped image of rotating insert tip using image restoration technique. We calculated the mean square error to diagnose the condition of tool fracture, and determind the criteria of tool fracture using experimental and staticstical method. From the results of this study we've developed non- contact detection technique of tool fracture using image processing method and proposed the fracture direction of automation and unmanned system considering the optimal time of tool change milling.

• 한국과학교육학회지
• /
• 제42권1호
• /
• pp.51-60
• /
• 2022

본 연구는 과학고등학교 입학전형에 인재상을 고려하는 개선 방향을 탐색하는 것을 목적으로 하였다. 이를 위해 먼저 과학고등학교 인재상과 역량을 '공공성'과 '효용성', '수행'과 '성과' 유형으로 구분하였는데, 대부분 '공공성-수행'이 우세하였다. 다음으로 3개 과학고등학교의 입학전형에서 인재상 반영 정도와 개선 방향에 대한 설문조사를 과학고등학교 교사, 교육 전문가, 중학교 교사 총 28명을 대상으로 실시하였다. 설문의 결과는 다음과 같았다. 첫째, 1단계 전형에서는 '공공성' 평가가 매우 제한적이었고, '성과' 중심으로 선발하는 경향이 높았다. 둘째, 2단계 면접 평가에서는 '효용성-성과' 중심으로 선발하는 경향이 매우 강했다. 셋째, 1단계 전형에 '역량 중심의 자기소개서 작성, 학교생활기록부의 타 교과 성적 및 비교과 반영, 전문가의 면담'을 도입하여 '공공성-수행' 유형의 인재상에 부합하는 학생선발로 개선해야 한다. 넷째, 2단계 면접 평가에서 '공공성-수행' 유형의 인재상을 반영하기 위해서는 발산적 사고가 가능한 융합적이고 개방형 문항의 공동 출제가 이루어져야 한다. 결론적으로 과학고등학교 입학전형에 인재상을 고려하는 개선 방향은 역량 중심의 1단계 서류평가와 개방형 문항으로 구성된 2단계 면접 평가로 변화되어야 한다는 것이다.

• 한국멀티미디어학회논문지
• /
• 제12권6호
• /
• pp.815-823
• /
• 2009

Most of existing watermarking schemes insert and extract a watermark, focusing on the visual conservation of an original image. However, existing watermarking schemes could be difficult for a watermark detection in case of various distortion caused by display-capture devices. Therefore, we propose a new display-capture based mobile watermarking scheme. The proposed watermarking scheme is a new concept for embedding a watermark, which uses the generated image instead of a given original image. For effective watermark decoding, we also present a method for detecting the background image whose error bit can not be corrected because of various heavy distortion and for avoiding it from the decoding process. For this scheme, we adopt distortion coefficients of camera calibration when we separate a background image from a captured image. For finding available correction bits of ECC through the decoding process, we capture 30,000 images and then calculate the separation ratio of a background image and the average error bits per an image. As experimental result, the separation ratio of a background image is about 96.5% in 30,000 captured image. And the false alarm ratio shows about $5.18{\times}10^{-4}$ in the separated background image. And also we can confirm the availability of real-time processing because the mean execution time is about 82ms per an image for capturing and decoding.

• 한국멀티미디어학회논문지
• /
• 제17권11호
• /
• pp.1270-1278
• /
• 2014

The modern infrared (IR) imaging systems use mainly one or more wavelength bands among short wavelength IR (SWIR), middle wavelength IR (MWIR), and long wavelength IR (LWIR) bands. In the process of IR image synthesis and modeling, IR image wavelength-band conversion which transforms arbitrary band image to other band one is required. The wavelength-band conversion procedure includes a temperature estimation process of an object surface. However, in this procedure, an approximated Planck's radiation equation causes errors in estimated temperature. In this paper, we propose an estimation temperature error attenuation method in IR image band conversion procedure. The estimated temperature is corrected with a slope information of radiance according to it. The corrected temperature is used for generation of the other band IR image. The verification of proposed method is demonstrated through the simulation.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2006년도 하계종합학술대회
• /
• pp.969-970
• /
• 2006

This paper develops a recognition system of assembly parts using a hardware independent image module. Using a shared memory, the image module consists of the image acquiring process and the image processing process. We preprocess an acquisition image from the module, approximate the image edges to an ellipse, and then recognize an assembly part by matching the ellipse to a model base one.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권8호
• /
• pp.4255-4269
• /
• 2019

This paper proposes an image generation method using a Multi Discriminator Generative Adversarial Net (MDGAN) as a next generation 2D game sprite creation technique. The proposed GAN is an Autoencoder-based model that receives three areas of information-color, shape, and animation, and combines them into new images. This model consists of two encoders that extract color and shape from each image, and a decoder that takes all the values of each encoder and generates an animated image. We also suggest an image processing technique during the learning process to remove the noise of the generated images. The resulting images show that 2D sprites in games can be generated by independently learning the three image attributes of shape, color, and animation. The proposed system can increase the productivity of massive 2D image modification work during the game development process. The experimental results demonstrate that our MDGAN can be used for 2D image sprite generation and modification work with little manual cost.

• 대한기계학회논문집A
• /
• 제27권5호
• /
• pp.697-704
• /
• 2003

In this paper, the visual tracking system for arbitrary shaped moving object is proposed. The established tracking system can be divided into model based method that needs previous model for target object and image based method that uses image feature. In the model based method, the reliable tracking is possible, but simplification of the shape is necessary and the application is restricted to definite target mod el. On the other hand, in the image based method, the process speed can be increased, but the shape information is lost and the tracking system is sensitive to image noise. The proposed tracking system is composed of the extraction process that recognizes the existence of moving object and tracking process that extracts dynamic characteristics and shape information of the target objects. Specially, active contour model is used to effectively track the object that is undergoing shape change. In initializatio n process of the contour model, the semi-automatic operation can be avoided and the convergence speed of the contour can be increased by the proposed effective initialization method. Also, for the efficient solution of the correspondence problem in multiple objects tracking, the variation function that uses the variation of position structure in image frame and snake energy level is proposed. In order to verify the validity and effectiveness of the proposed tracking system, real time tracking experiment for multiple moving objects is implemented.

• 한국군사과학기술학회지
• /
• 제9권3호
• /
• pp.5-11
• /
• 2006

This paper is launched to get an exact bore-sighting value in the process of assembling for a dual-barrel weapon. Image processing method with a CCD(Charged Coupled Device) camera is chosen for the error reduction of bore-sighting. The results of this method by using a CCD camera and the testing target method are described in this paper. After the performance of the dual-barrel weapon and that of the fire-control system in the dual barrel weapon system is confirmed, the bore-sighting which is a part of the process in the system assembly is accomplished. In this process, the position of the barrel is identified by using the testing target method that is an existing bore-sighting method. Then, the fixing line of the fire-control system is checked by a day-optical part. The precision of the bore-sighting is required within several mils, however the manual method using the naked eyes makes it worse. Therefore, a CCD camera is installed in the eyepieces. Next, we can get an image of the sighting and the center coordinate values of the laser-pointer from each barrel by image processing method. A required bore-sighting value is calculated from the eccentricity of the center coordinate. Finally, It can be applied to adopt this result in the assembling process of the dual-barrel weapon.