수평 Bridgman(HB)법에 의한 GaAs 단결정 성장 및 그 특성 평가

탁영봉 慶尙大學校生産技術硏究所 1992 生産技術硏究所論文集 Vol.8 No.-

Undoped and Si-doped(n-type) GaAs single crystal were grown by the Horizontal bridgman(HB) method. The grown ingots are D-shape, 2 inch diameter and more than 20㎝ long, which are applicable for mass-production. The indirect growth method facilitated seeding, growing, stoichiometry control and especially wetting control in the case of Si-doping, which greatly improved the yield of single crystal from 25% of the direct growth method to 83%. And also quartz cloth was used to minimize the wetting reaction. The axial EPD distribution was measured along the longitudinal direction of the ingot(GSC 90) and showed the good results of 1-5x10³㎝² through the total length. The transverse EPD distribution was mapped onto a wafer. The dopant distribution was measured along the axial direction. The shape of solid-liquid interface was inferred from the C-V and Hal measurement, and directly obtained by photoetching a wafer and an ingot.

Silver/Sodium 이온교환된 도파로의 폭 측정

탁영봉 경상대학교 생산기술연구소 2003 工學硏究院論文集 Vol.19 No.-

In this paper, the width of silver/sodium ion-exchanged waveguides were determined by using diffraction patterns. It is necessary for efficient coupling with an optical fiber to measure the waveguide width accurately, which also makes possible the use of an integrated optical coupler. The reflected and the transmitted diffraction patterns of diffused strip waveguides fabricated by silver/sodium ion-exchanged were measured. For the reflected diffraction pattern, the back surface of slide glass was made rough by rubbing it or bombarded in a sand-blast cabin using aluminium oxide to eliminate the unwanted reflection from it. The diffusion mask was produced by Cr film photo mask with the width of 120pm. The waveguide widths were determined by the reflected and the transmitted diffraction patterns to be 123.7㎛ and 117.4㎛, respectively. The slight difference between them might be caused by the conditions of exposure and development for the two samples, the inaccuracy of the observed distance and point, and the deviation of the incident angle especially for the reflected diffraction pattern. However, the measured value is considered to be very accurate, approching the mask width of 120㎛.

컴퓨터비젼에 의한 자동차용 Head Lamp Cover 형상인식

탁영봉,김대원 경상대학교 생산기술연구소 2001 工學硏究院論文集 Vol.1 No.-

The computer vision technique was used to recognize the shape of automobile head lamp cover. To acquire a clear input image, two lightings were utilized to make the total reflection from the bottom of the jig on which transparent lamp covers were laid. The threshold value of the input image was obtained by the optimal adaptive thresholding of the Otsu method. The contour of a sample was extracted from the binarized image after being pre-processed, and then it's centroid was found out using the coordinate of each point. The interior of the contour was colored, and it's area was calculated by counting the pixels of the area. The length between the centroid and the contour was traced by rotating from 0˚ to 360˚. Using this polar-sampling graph, the feature values to recognize the type of a sample were extracted. comparing the feature values of a sample with those stored in teaching mode recognizes as an identical object if the error is within the predetermined tolerance of 5%. The feature data was obtained by teaching a sample 5 times each for 20 different samples. In on-line test, the recognition rate of 95% was obtained, but it will still be improved if we can reproducibly acquire clear input image by fixing the location of the jig.

광 분할 광섬유 결합기의 제작에 관한 연구

탁영봉 慶尙大學校生産技術硏究所 1993 生産技術硏究所論文集 Vol.9 No.-

Fiber optic data bus system의 분기소자로 사용될 광 분할 광섬유 결합기를 제작하었다. 이 결합기의 삽입손실에 결정적 요인이 되는 결합기 양단간의 정렬 상태를 개선하기 위하여, 또한 결합기의 제작시 절단과 연마 도중 발생될 수 있는 표면손상에 의한 손실을 최대한 줄이기 위하여 여러가지 형태의 결합기를 제작하였으며 그들의 삽입손실을 측정하였다. 삽입손실 측정에는 LED(λ= 842mm)를 광원으로 하였고 PIN photodiode를 사용하였으며, 측정에 큰 영향을 미치는 cladding modes를 제거하기 위해 2Km의 긴 launching fiber를 사용하였다. 제작된 소자들 중 몇몇은 -0.3 dB 이하의 작은 삽입손실을 보였으며, 그 중에는 -0.05 dB의 무시할만한 삽입손실을 갖는 결합기도 제작하였다. 작은 삽입손실을 갖는 모델의 소자를 선택하여 임의의 각 도로 절단 연마된 광섬유 단면 위에 e-beam에 의해 다충의 dielectric mirror를 증착시켰으며, 이는 광통신 시스템에서 WDM 소자로서 다중의 광신호를 결합 또는 분리하는데 유용하게 사용될 수 있다. He-Ne laser를 사용하여 45˚로 절단 연마된 광섬유 단면으로부터 far-field reflected radiation pattern을 측정하였으며, 이는 분기소자의 반사광에 대한 집광효율을 최적화 시키기 위하여 Photodetector나 분기 광섬유의 크기의 위치를 결정하는데 유용할 것이다.

Silver/Sodium 이온교환에 의한 집적광학 결합기의 제조

탁영봉 경상대학교 생산기술연구소 2002 工學硏究院論文集 Vol.2 No.-

본 논문에서는 Silver/Sodium 이온교환에 의해 채널도파로 소자들을 제작하였다. 광섬유와 채널도파로 간의 원활한 결합을 위하여 in-diffusion과 back-diffusion 사이에 어떤 비율을 가지고 확산시간을 변화시켜 가면서 채널도파로의 단면형상을 변화시켰다. 최적의 확산변수와 도파로 폭을 찾기위하여 S-형상을 가진 도파로를 제작하여 삽입손실을 측정하였다. 25㎛폭의 mask를 사용하여 9∼10의 시간비율(T_i/T_b)로 14∼15 시간동안 확산하여 제조한 도파로는 약 -4dB의 최소 삽입손실을 보였다. 이들 변수를 근거로 Y-junction과 분배회로와 같은 branching element들을 제조하였으며 그 측정 결과는 각각 -4.5dB와 -7.12dB 이었다. 이 branching element들은 집적광학 결합기로서 실제적으로 응용될 수 있다.

타이어 접지형상 자동측정 시스템 개발

김영탁,정태영,탁영봉 慶尙大學校生産技術硏究所 1995 生産技術硏究所論文集 Vol.11 No.-

The contribution describes an automatic measuring system for analysing tire contact shapes and air-pressure distribution upon weight load using the computer vision. Instead of analysing the safety degree of automobiles by manually calculation the tire contact area. This system is to analyse it automatically using a computer by the image processing technique. This system is divided into three major process : i) to binarize an input image for discriminating object from background, and then to remove noise from it. ii) to extract the image contour, applying the chain code to each seperated region and then labelling them, based on the process of dilation and erosion of the image. iii) to calculate the real contact area and to set corresponding color levels upon air pressure. Representing the gray level as pseudo-color have enabled us to visualize the air-pressure distribution upon weight load. We have developed an automatic system of calculating real contact area and its ratio against total tire contact area using the image processing technique. This has reduced the measuring time and error to half of those the manual method.

광 분할 광섬유 결합기의 반사광 강도 분포

탁영봉,Lagasse, P. 慶尙大學校 工科大學 自動化및컴퓨터應用技術硏究所 1997 自動化 및 컴퓨터應用技術 Vol.4 No.1

본 논문에서는 광 분할 광섬유 결합기로부터 반사된 빛을 효율적으로 검출하기 위하여 반사광 강도에 대한 식을 유도하고 강도 분포를 계산하고자 한다. 이를 위하여 incoherent하게 조사된 다중 모드 광섬유에서는 반사광 강도가 굴절률의 자승에 비례한다고 가정하였다. 반사광 강도를 계산하기 위하여 약하게 도파되는 광섬유에서는 근사식을 사용하였으나 STEP-INDEX 광섬유에서는 Δ=0.005∼0.05의 굴절률 차이 범위에서 정확한 식을 사용하였다. 또한 CONE의 면적은 가우시안 구적의 수치적인 방법으로 구하였다. 본 연구에서 얻어진 결과는 반사광의 검출 효율을 높이기 위하여 광 분할기 위에 연결된 branch fiber나 photodetector의 크기 및 위치를 결정하는데 유용하게 사용될 수 있을 것이다.

인터넷상에서의 실시간 주문형 비디오 설계 및 구현

탁영봉,이종철 慶尙大學校 工科大學 自動化및컴퓨터應用技術硏究所 1998 自動化 및 컴퓨터應用技術 Vol.5 No.1

In implementing real time video on demand(VOD), the increase of user on internet causes a network traffic congestion. In this paper, we programmed a CGI able to login in VOD home for limiting the number of user in solving the problem, and also applied an adaptive multimedia synchronization technique for controlling video and audio data in a network. In addition, a real time multimedia player was designed and implemented in a personal computer operating at Window95/98/NT.

핵 연료봉번호 자동인식 시스템 개발

정태영,탁영봉 慶尙大學校 工科大學 自動化및컴퓨터應用技術硏究所 1998 自動化 및 컴퓨터應用技術 Vol.5 No.1

In this paper, we developed an automatic recognition system of nuclear fuel rod numbers by the computer vision using a neural network, which facilitated the automation of the production process and posterior management of fuel rods. Prior to this, workers on the line manually put fuel rod numbers on record and manafement. The recognition of fuel rod numbers is mainly divided into three processes. ⅰ) To build up a line-scanned entire image from an image captured by a CCD camera, and then to magnify and binarize the image. The regions of discrete numbers are then detected by projecting the binarized image. ⅱ) To extract features from the regions of detected discrete numbers. ⅲ) To input the extracted features into a neural network which have input nodes 33, hidden nodes 20, and output nodes 10, using the back propagation algorithm. This system made it possible to reduce the man power and to process the data faster than the conventional manual process by automatically recognizing fuel rod numbers and transmitting them to a workstation. The recognition rate was successfully tested to be 99% from almost every batch of 450 fuel rodes. The 1% incorrect recognition was mainly due to bad font appearance of fuel rod numbers.

음·양각의 타이어문자 인식을 위한 영상전처리

류한성,탁영봉,정태영 慶尙大學校 工科大學 自動化및컴퓨터應用技術硏究所 1996 自動化 및 컴퓨터應用技術 Vol.3 No.1

Alpha-numeric characters representing sizes, patterns and production dates etc. are engraved in dented and raised appearances in automobile tires. The existing methods for tire classification comprise the recognition of sizes, widths and tread patterns by human eyes or the reading of expensive bar-code labels for high temperature use. On the contrary, this work will realize the automatic classification by recognizing the engraved characters using the computer vision technique. Since the characters and backgrounds in tire images have similar grey-level values, it is not easy to recognize the characters. Since the one side of the character string is raised and the other dented, the captured images are changed on the angles of camera and illumination. Therefore it is requested to find out the input conditions for obtaining the optimum images. The input images are smoothed by the fuzzy MIN/MAX operators, and then the character string is searched by using the stochastic line histogram obtained. The existing operators are applied to the detected character string and the comparison among them determines the optimum edge detection operator, which enables the segmentation of the string into the discrete characters. The mask patterns is devised to connect missing pixels in the discrete characters. And then We will extract the features from the retriered discrete characters, and apply them to the input of a neural network for the recognition of the tire characters.