Vol.2 No.4 2010

14/68

Research paper : Development of a real-time all-in-focus microscope (K. Ohba)−244−Synthesiology - English edition Vol.2 No.4 (2010) translucent object, while there can be an object with several points in focus in the vertical direction, this method allows selection of one point only and unconditional observation of other areas become impossible.As another method for reducing the optical scaling effect, the technology for volume renderingTerm 4 directly from the images with multiple focal distances, as shown in Fig. 23, was considered. This has been patented[9].In the future, it is expected that there will be demand for a hyper-microscope that can produce image slices from any direction by volume rendering all the depth images, without having to consider the hardware limitations.AcknowledgementsThis study was based on the joint research by the Mechanical Engineering Laboratory, Agency of Industrial Science and Technology (current AIST), Ministry of International Trade and Industry; Delft High Tech Corporation (current DHT Corporation); Kawatetsu Techno Research Corporation (current JFE Techno Research Corporation); and Denso Corporation. The joint research for practical application was conducted by AIST and Photron Ltd. Part of the varifocal lens used in this study was manufactured by Denso Corporation, as a subcontract of the Micromachine Center that was subcontracted by the New Energy and Industrial Technology Development Organization (NEDO), as part of the “R&D for Micromachine Technology” based on the Industrial Science and Technology Frontier Program, Agency of Industrial Science and Technology, Ministry of International Trade and Industry.References[1][2][3][4][5][6][7][8][9]K. Ohba, J.C.P. Ortega, K. Tanie, G. Rin, R. Dangi, Y. Takei, T. Kaneko and N. Kawahara: Implementation of real-time micro VR camera, Denki Gakkai Ronbunshi E (IEEJ Transactions on Sensors and Micromachines), 120-E (6), 264-271 (2000) (in Japanese).K. Ohba, J.C.P. Ortega, K. Tanie, G. Rin, R. Dangi, Y. Takei, T. Kaneko and N. Kawahara: Micro-observation technique for tele-micro-operation, Advanced Robotics, 15 (8), 781-789 (2001).K. Ohba: Frame-rate all-in-focus microscopic system, Nihon Robotto Gakkaishi (Journal of Robotics Society of Japan), 21 (1), 43-44 (2003) (in Japanese).Patent 3737483, Real-time all-in-focus microscope cameraM. Ishihara and T. Furusawa: Saikin no hikari hyomen keijo keisoku gijutsu, O plus E, 20 (11), 1251-1258 (1998).S.K. Nayer and Y. Nakagawa: Shape from focus, IEEE Trans. on PAMI, 16 (8), 824-831 (1994).K. Kodama, T. Onishi, K. Aisawa and M. Hattori: Arbitrarily focused image acquisition by using multiple differently focused images, the Journal of the Institute of Image Information and Television Engineers, 51 (12), 2072-2081 (1997).T. Kaneko, N. Taya, N. Kawahara, N. Akita and T. Hattori: Kahen shoten lenzu o mochiita cho shoten shindo shikaku kiko, Denki Gakkai Maikuro Mashin Kenkyukai (1997).Patent 3627020, Three dimensional transmissive microscope system and image display methodFig. 23 Overview of volume rendering method.Whole viewVertical section imageHorizontal section imageOrdinary microscope image at certain focal distanceVolume rendered 3D microscope imageTerminologyLVDS (low voltage differential signaling) interface: standard of electric signal that enables high-speed function using twist pair cables. It has been used in high-speed cameras to handle high volume data. It is used in PCs as interface with LC displays.FPGA (field programmable gate array) processing: a type of programmable gate array where the users can write their original logic circuit. The gate array is arranged in two-dimensional lattice form. It is suitable for parallel processing operation.SIMD (single instruction multiple data) processing: process in an arithmetic device where simultaneous processing for several data is done with one command.Volume rendering: When expressing a 3D object as a 2D image, rather than the method where the depth is expressed by adding shadow on the object surface, in this method, the 3D object is made to look three dimensional by adding transmittance and color information inside the object.Term 1. Term 2.Term 3.Term 4.

※このページを正しく表示するにはFlashPlayer9以上が必要です