• 대한전자공학회논문지
• /
• 제20권6호
• /
• pp.75-80
• /
• 1983

스므싱 필터를 사용하여 spot noise를 갖는 협상을 스므싱할 때에, 잡음은 거의 제거되지만 영상의 blurring을 초래한다. 본 논문에서는 이러한 영상 blurring을 동시에 개선하기 위하여 메디안 필터를 이용한 한 알고리즘을 제안하였다. 본 논문에서 제시한 방법은 잡음협상과 그 잡음영상을 메디안 필터링한 결과와의 차를 이용하였으며, 잡음제거와 동시에 blurring의 감소를 효과적으로 이룩할 수 있음을 보여 주었다. 본 논문에서는 또한 제안한 알고리즘을 여러가지 기존 스므싱 방법과 비교하여 그 특징을 제시하였다.

• 한국컴퓨터정보학회논문지
• /
• 제23권1호
• /
• pp.33-39
• /
• 2018

In this paper, We studied the possibility of SoC(System On Chip) design using infrared image processing IP(Intellectual Property). And, we studied NUC(Non Uniformity Correction), BPR(Bad Pixel Recovery), and CEM(Contrast Enhancement) processing, the infrared image processing algorithm implemented by IP. We showed the logic and timing diagram implemented through the hardware block designed based on each algorithm. Each algorithm was coded as RTL(Register Transfer Level) using Verilog HDL(Hardware Description Language), ALTERA QUARTUS synthesis, and programed in FPGA(Field Programmable Gated Array). In addition, we have verified that the image data is processed at each algorithm without any problems by integrating the infrared image processing algorithm. Particularly, using the directly manufactured electronic board, Processor, SRAM, and FLASH are interconnected and tested and the verification result is presented so that the SoC type can be realized later. The infrared image processing IP proposed and verified in this study is expected to be of high value in the future SoC semiconductor fabrication. In addition, we have laid the basis for future application in the camera SoC industry.

• 한국정밀공학회지
• /
• 제12권10호
• /
• pp.78-88
• /
• 1995

For producing iron or steel products with good quality, the concentration of the material components should be analyzed quickly with high relability using XRF(Fluorescent X-Ray Spectrometer). Since the analysis results are much dependent upon the surface con- dition, the samples have to be prepared to have good test condition. This study presents an image processing system for inspecting the surface condition of the iron test sample. In order to use thd computer vision system, we need to develop a lighting device and image processing algorithm. For the adequate lighting device of inspection system, the indirect lighting device is contrived to cut the external light and provide uniform, stable and cold light. The image processing algorithm is aimed to reduce inspection time and to get similar analyzing results to those of the experienced operators. At first, the image processing algorithm checks whether the surface of the iron sample is ground well or not. Then, the defects; hole or dig are conted and surface condition is evaluated. In addition, the algorithm gives the reliability of the analyzing results in order to help operator's decision.

• 한국산학기술학회논문지
• /
• 제4권4호
• /
• pp.434-440
• /
• 2003

Jacquin에 의해 프랙탈 이론을 이용한 영상 부호화 기법이 소개된 이래로, Fisher와 Beaumont 등에 의하여 낮은 비트율에서도 우수한 화질을 제공하는 프랙탈 영상 압축 기법들이 다수 제안되었다. 그러나 기존에 고안된 기법들이 갖고 있는 하나의 문제점은 복호화가 반복 처리를 통해 구현되며 그 복잡도가 각각의 영상에 따라 상이하다는 것이다. 본 논문에서는 복호 시간을 단축시키기 위해 반복 변환이 필요 없는 프랙탈 영상 복호 알고리즘을 제안하고자 한다. 제안된 방법은 복호기에서 사용할 초기 영상과 동일한 코드북 영상을 부호기에 보유하고 있는 상태에서 부호화 과정에서는 이 코드북 영상과 부호화하려는 영상의 유사성을 찾아 프랙탈 계수를 구한다. 이후, 복호화 과정에서는 수신된 프랙탈 계수와 기설정된 초기 영상을 이용하여 반복 변환 없이 한 번에 영상을 복호함으로써 복호 시간을 현저하게 단축시킬 수 있었다.

• Transactions on Control, Automation and Systems Engineering
• /
• 제2권1호
• /
• pp.47-55
• /
• 2000

In this paper, a vision-based roadway detection algorithm for an automated vehicle control system, based on roadway sign information on roads, is proposed. First, in order to detect roadway signs, the color scene image is enhanced under hue-invariance. Fuzzy logic is employed to simplify the enhanced color image into a binary image and the binary image is morphologically filtered. Then, an effective algorithm of locating signs based on binary rank order transform (BROT) is utilized to extract signs from the image. This algorithm performs better than those previously presented. Finally, the inner shapes of roadway signs with curving roadway direction information are recognized by neural networks. Experimental results show that the new detection algorithm is simple and robust, and performs well on real sign detection. The results also show that the neural networks used can exactly recognize the inner shapes of signs even for very noisy shapes.

• IEIE Transactions on Smart Processing and Computing
• /
• 제2권5호
• /
• pp.317-321
• /
• 2013

This paper proposes a detection algorithm for view reversal in a stereoscopic video using a disparity map and motion vector field. We obtain the disparity map of a stereo image was obtained using a specific stereo matching algorithm and classify the image into the foreground and background. Next, the motion vector field of the image on a block basis was produced using a full search algorithm. Finally, the stereo image was considered to be reversed when the foreground moved toward the background and the covered region was in the foreground. The proposed algorithm achieved a good detection rate when the background was covered sufficiently by its moving foreground.

• 한국멀티미디어학회논문지
• /
• 제7권12호
• /
• pp.1737-1744
• /
• 2004

In the compression methods widely used today, the image compression by VQ is the most popular and shows a good data compression ratio. Almost all the methods by VQ use the LBG algorithm that reads the entire image several times and moves code vectors into optimal position in each step. This complexity of algorithm requires considerable amount of time to execute. To overcome this time consuming constraint, we propose an enhanced self-organizing neural network for color images. VQ is an image coding technique that shows high data compression ratio. In this study, we improved the competitive learning method by employing three methods for the generation of codebook. The results demonstrated that compression ratio by the proposed method was improved to a greater degree compared to the SOM in neural networks.

• 대한기계학회논문집A
• /
• 제31권3호
• /
• pp.395-402
• /
• 2007

This paper considers about the tilt measurement of lens module with multiple lenses. The tilt between lenses in lens module and barrel or between image sensor and barrel can be measured precisely with the proposed algorithm. The magnitude and direction of the tilt vector of lens and image sensor can be measured from the best focal surface. The selecting and setting of image sensor, test chart, image sensor centering to lens module, axis align, focus measure method are also explained to get highly precise tilt results. The proposed algorithm is verified with the lens module inspection system we developed, and the experimental results show that the tilt measure proposed in this paper is robust and precise. With the proposed tilt measurement algorithm, the tilt of an image sensor and any other lens which intermediates light can be measured.

• Journal of Information Processing Systems
• /
• 제5권1호
• /
• pp.1-4
• /
• 2009

Level set methods are the numerical techniques for tracking interfaces and shapes. They have been successfully used in image segmentation. A new variational level set evolving algorithm without re-initialization is presented in this paper. It consists of an internal energy term that penalizes deviations of the level set function from a signed distance function, and an external energy term that drives the motion of the zero level set toward the desired image feature. This algorithm can be easily implemented using a simple finite difference scheme. Meanwhile, not only can the initial contour can be shown anywhere in the image, but the interior contours can also be automatically detected.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2006년도 춘계학술대회 논문집
• /
• pp.483-484
• /
• 2006

This paper introduces an improved surface matching algorithm that can be used to reconstruct the surface topography of an object that is scanned from multiple overlapping regions by an AFM. The essence of the image matching technique is stitching two neighboring images intentionally overlapped with each other. To enhance the computational efficiency, this paper introduces a pyramid matching algorithm which makes use of reduced images for primary images. The results show that the proposed image pyramid matching algorithm is useful fer enhancing the computational efficiency.