• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2000년도 ITC-CSCC -2
• /
• pp.835-838
• /
• 2000

Popular methods for extracting a text region in video images are in general based on analysis of a whole image such as merge and split method, and comparison of two frames. Thus, they take long computing time due to the use of a whole image. Therefore, this paper suggests the faster method of extracting a text region without processing a whole image. The proposed method uses line sampling methods, FFT and neural networks in order to extract texts in real time. In general, text areas are found in the higher frequency domain, thus, can be characterized using FFT The candidate text areas can be thus found by applying the higher frequency characteristics to neural network. Therefore, the final text area is extracted by verifying the candidate areas. Experimental results show a perfect candidate extraction rate and about 92% text extraction rate. The strength of the proposed algorithm is its simplicity, real-time processing by not processing the entire image, and fast skipping of the images that do not contain a text.

• 대한전기학회논문지
• /
• 제37권11호
• /
• pp.804-815
• /
• 1988

Real-time image processing algorithms to detect position and orientation of rectangular type electronic components are developed. The position detection algorithm is implemented with the use of projection method which is insensitive to noise. Also dynamic thresholding method of projection is employed in order to distinguish between the boundary of a component and any marking on the component. The orientation is determined by Hough transform of boundary candidates of a component, which is obtained a priori by a simple edge detection method. For real-time processing of both position and orientation for a component which is not aligned well, parallel processing method of image data is proposed and tested in real-time.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2000년도 제15차 학술회의논문집
• /
• pp.171-171
• /
• 2000

We have developed m innovative surface inspection system for automated quality control for steel products in POSCO. We had ever installed the various kinds of surface inspection systems, such as a linear CCD and a laser typed surface inspection systems at cold rolled strips production lines. But, these systems cannot fulfill the sufficient detection and classification rate, and real time processing performance. In order to increase detection and classification rate, we have used the Dark, Bright and Transition Field illumination and area type CCD camera, and fur the real time image processing, parallel computing has been used. In this paper, we introduced the automatic surface inspection system and real time image processing technique using the Object Detection, Defect Detection, Classification algorithms and its performance obtained at the production line.

• 대한원격탐사학회지
• /
• 제29권2호
• /
• pp.183-195
• /
• 2013

영상 기반의 정밀한 내비게이션이나 증강현실을 구현하기 위해서 고정밀 영상 지오레퍼런싱의 실시간 수행이 필수적이다. 일반적으로 고정밀 영상 지오레퍼런싱은 일괄 번들 조정 알고리즘을 적용하여 성취될 수 있으나 처리시간으로 인해 후처리로만 가능하였다. 최근에 제안된 연속 번들 조정 알고리즘은 이와 유사한 정확도의 결과를 고속으로 산출하여 실시간 처리의 가능성을 열었다. 그러나 처리시간이 영상의 개수에 따라 점진적으로 증가하기 때문에 영상이 아주 많은 경우에 실시간 수행이 보장되지 못하는 한계가 있었다. 이러한 연속 번들 조정 알고리즘을 보완하여 본 연구는 실시간 처리를 위해 영상의 개수와 무관하게 처리시간을 항상 일정 범위로 한정시킬 수 있는 알고리즘을 제안한다. 제안된 알고리즘은 기존의 영상들 중에서 새롭게 취득된 영상과 통계적으로 관련성이 높은 영상들만 고려하기 때문에 처리시간을 일정 범위로 한정시키면서도 비교적 정확한 결과를 산출할 수 있다. 제안된 알고리즘을 1Hz로 취득된 영상에 적용한 결과, 평균적으로 영상을 취득할 때마다 0.02 초 이내의 처리시간을 소요하면서 기존의 일괄 번들 조정 알고리즘의 결과와 비교하여 지상점 좌표를 기준으로 ${\pm}5$ cm 이내의 정확도로 지오레퍼런싱을 수행할 수 있었다. 신뢰성 높은 고속의 영상 매칭 알고리즘과 결합된다면 스마트폰 또는 UAV 등으로 동영상을 취득하면서 함께 탑재된 위치/자세 센서의 성능을 보완하여 고정밀의 실시간 지오레퍼런싱이 가능할 것으로 판단된다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.118.5-118
• /
• 2001

This paper describes an image processing method for inspecting the surface outlook of high speed moving plates. Noise free image and a new real time processing methods are required to inspect the surface outlook of the high speed moving plates in real time. It is difficult to get a noise free image due to a signal noise, a light noise and background image in typical industrial factory. Thus, pre-processing techniques should be required to get a good image and produce so many time steps to proceed the image data. The objective of this research is to get image on the surface of the moving plates with a speed of 1m/sec and to detect some defaults on the surface image. So, the pre-processing techniques ...

• Current Optics and Photonics
• /
• 제2권5호
• /
• pp.453-459
• /
• 2018

We have developed a high-performance signal-processing and image-rendering heterogeneous computation system for optical coherence tomography (OCT) on mobile processor. In this paper, we reveal it by demonstrating real-time OCT image processing using a Snapdragon 800 mobile processor, with the introduction of a heterogeneous image visualization architecture (HIVA) to accelerate the signal-processing and image-visualization procedures. HIVA has been designed to maximize the computational performances of a mobile processor by using a native language compiler, which targets mobile processor, to directly access mobile-processor computing resources and the open computing language (OpenCL) for heterogeneous computation. The developed mobile image processing platform requires only 25 ms to produce an OCT image from $512{\times}1024$ OCT data. This is 617 times faster than the naïve approach without HIVA, which requires more than 15 s. The developed platform can produce 40 OCT images per second, to facilitate real-time mobile OCT image visualization. We believe this study would facilitate the development of portable diagnostic image visualization with medical imaging modality, which requires computationally expensive procedures, using a mobile processor.

• Applied Microscopy
• /
• 제46권1호
• /
• pp.6-13
• /
• 2016

Fluorescence lifetime imaging microscopy (FLIM) has been considered an effective technique to investigate chemical properties of the specimens, especially of biological samples. Despite of this advantageous trait, researchers in this field have had difficulties applying FLIM to their systems because acquiring an image using FLIM consumes too much time. Although analog mean-delay (AMD) method was introduced to enhance the imaging speed of commonly used FLIM based on time-correlated single photon counting (TCSPC), a real-time image reconstruction using AMD method has not been implemented due to its data processing obstacles. In this paper, we introduce a real-time image restoration of AMD-FLIM through fast parallel data processing by using Threading Building Blocks (TBB; Intel) and octa-core processor (i7-5960x; Intel). Frame rate of 3.8 frames per second was achieved in $1,024{\times}1,024$ resolution with over 4 million lifetime determinations per second and measurement error within 10%. This image acquisition speed is 184 times faster than that of single-channel TCSPC and 9.2 times faster than that of 8-channel TCSPC (state-of-art photon counting rate of 80 million counts per second) with the same lifetime accuracy of 10% and the same pixel resolution.

• 제어로봇시스템학회논문지
• /
• 제3권6호
• /
• pp.626-631
• /
• 1997

This paper represents the design of a real-time image preprocessing system. The preprocessing system performs hardware-wise mask operations and thresholding operations at the speed of camera output single rate. The preprocessing system consists of the preprocessing board and the main processing board. The preprocessing board includes preprocessing unit that includes a $5\times5$ mask processor and LUT, and can perform mask and threshold operations in real-time. To achieve high-resolution image input data($20485\timesn$), the preprocessing board has a linescan camera interface. The main processing board includes the image processor unit and main processor unit. The image processor unit is equipped with TI's TMS320C32 DSP and can perform image processing algorithms at high speed. The main processor unit controls the operation of total system. The proposed system is faster than the conventional CPU based system.

• 전기전자학회논문지
• /
• 제26권2호
• /
• pp.226-231
• /
• 2022

정보 엔트로피의 개념은 다양한 분야에서 폭넓게 응용되고 있다. 최근 영상처리 분야에서도 정보 엔트로피 개념을 응용한 기술들이 많이 개발되고 있다. 현대 산업에서 컴퓨터 비전 기술들의 중요성과 수요가 증가함에 따라, 영상처리 기술들이 현대 산업에 효율적으로 적용되기 위해서는 실시간 처리가 가능해야 한다. 영상의 엔트로피 값을 추출하는 것은 소프트웨어로는 계산량이 복잡해 실시간 처리가 어려우며 실시간 처리가 가능한 영상 엔트로피 필터의 하드웨어 구조는 제안된 적이 없다. 본 논문에서는 barrel shifter를 사용하여 실시간 처리가 가능한 히스토그램 기반 엔트로피 필터의 하드웨어 구조를 제안한다. 제안한 하드웨어는 Verilog HDL을 이용하여 설계하였고, Xilinx사의 xczu7ev-2ffvc1156을 Target device로 설정하여 FPGA 구현하였다. Xilinx Vivado 프로그램을 이용한 논리합성 결과 4K UHD의 고해상도 환경에서 최대 동작 주파수 750.751MHz를 가지며, 1초에 30장 이상의 영상을 처리하며 실시간 처리 기준을 만족함을 보인다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.35.4-35
• /
• 2001

Image detection system for road traffic data analysis is a real time detection system using image processing techniques to get the real-time traffic information which is used for traffic control and analysis. One of the most important functions in this system is to match the coordinates of real world and that of image on video camera. When there in no way to know the exact position of camera and it´s height from the object. If some points on the road of real world are known it is possible to calculate the coordinates of real world from image.