(3) 自然光源 Planck’s law

全文

(1)Summary On E/O system engineering.

(2) E/O system. 資料壓縮. 儲存與傳輸.

(3) 自然光源 Planck’s law.

(4) Lambertian surface.

(5) Optics / Lighting Optics: 光的物理學 Electo-optics / Opto-electonics Photonics Lighting: 陽光的替身 (熱,螢,磷,激…) 共振腔. 激質.

(6) 光源(自然 /人造)的特性波長 / 光譜 --- 顏色空間 / 形狀 --- intensity control 時間 / 變化 --- function of time, (continuous, pulsed) . 型式：表面處理；色彩；控制 LED ??. 新世代光源革命. 品質規格：任務；興趣；亮度；舒適.

(7) 照明亮度環. 境. Lux = lm/m2.

(8) 照明場地及品質要求 . Ambient:. 均勻性全景(類陽光). 辦公室, 校園 . Aesthetic:. 裝飾性(藝術). 賣場, 展覽 . General :. 全方位(舒適與安全). 臥室, 房門 . Task lighting :. 無陰影. 閱讀, 烹飪, 化妝, 醫事手術 . Accent lighting : 展現特殊物件. 強調性.

(9) Ambient Illumination for optimal visual acuity.

(10) Light Source. Efficiency. (lumens/watt). Incandescent bulb Fluorescent lamp white LEDs 目前目標. Lifetime. (hours). 16. 1000. 85. 10,000. 25-80 sup to 200. 20,000 100,000.

(11) White LED 品質要求主要訴求: commercial and residential use . . . White LEDs is still tricky from design standpoint of view. Affordable white LED light that measures between 3000K and 4000K. A high-efficiency 120 - lumens/W output is a near-term goal..

(12) Surgical lamp . . much cooler, more energy efficient, and longer lived than current halogen surgical lamps. Available from October 2003. ter t i m e / 12 0 l m.

(13) 光電技術定義淺釋以光學原理與創新為基礎, 配合成熟的電子(機械)技術, 所組成的專業工程領域 Electro – Optics Opto – electronics Photonics. 精密與創新機械是工業之母!!.

(14)

(15) 當紅光電產品. 數位影像器.

(16) 數位晶片之攝影特性 . . 馬賽克映像, 容易失真. 較少麻點現象. 成像顏色不受曝光條件影響; 軟片則無此優點. 直接留影, 不需負片過程. 長時間曝光的技術很困難. 無重複曝光機制, 但容易儲存. (備影像重合處理之用)..

(17) Image Formation Real space. Fourier space. Obj. Spectrum. ＊ PSF. Image. x OTF. Amplitude.

(18) Two points resolution PSF behavior: 0.8 overlape ; ¼ λRayleigh criteria. A rule conveniently to define resolution.. . . Depends nothing more than size & shapes of aperture + wavelength of light. The radius of the Airy disk: 1.22λF/#. The measurement can never be S/N free..

(19) 數位晶片之特點 . . 即時顯影; 快速獲得結果. 儲存; 複製; 建檔(索引); 傳輸等執行方便. 即拍、即看、即傳(甚至修飾) ---具十足的娛樂效果. 多拍未必多消費---有‘量販、豐收’的快感. 展示多元化 ---提升教學、廣告等活潑性. 影像處理 ---優化、分類、研判..

(20) 技術趨勢 . . . High speed (ASA 1,600) film is poor. Color film may fade in time Digital stored on CDs or hard drives can take much less space Transmission: You have to send the original image everywhere 21st century digital capture is more linear than log as film is. This means that digital cameras often have better shadow detail.

(21) Image compression . Images take a lot of space . . a 512×512, 256 greyscale image consumes 256 kB. Compression techniques are used Examples: . GIF JPEG.

(22) 影像處理 . Why should an image be processed prior to analysis? . . It suffers from noise It fails to highlight the particular feature in which we are interested. In image processing, we remove noise & unnecessary features while highlighting the required features . Filtering.

(23) 影像處理產業.

(24) Pill-camera.

(25) 醫療檢查.

(26) Electro Luminescent (EL)顯示. . . 優勢: Flat panel; Bright, light emitting mono displays 缺點 :150V+ operating voltages; Color panels difficult to make; Development largely overtaken by PDPs.

(27) Flip-Dot 顯示 . . 優勢: Only draws power during a transition; Retains display pattern when off; Very wide viewing angle; Low cost for large segments; Can have long life (100khrs+) 缺點 High power required to change pixel state; Difficult to make segments smaller than 8mm diameter; Limited to mono color; Requires some ambient light to be visible.

(28) Incandescent 顯示 . . 優勢: Mature technology, very available; Lowest cost for moderate to high light output; Standard sizes easy to obtain 缺點: Short life (1khrs to 10khrs); socket adds cost; sually inefficient compared with LED; Narrow operating voltage range; High peak current when starting; Usually run hot; Usually fragile.

(29) Light Emitting Diode . . 優勢: Lowest cost red or green emissive indicator; Available in very small sizes; Very bright versions available (higher cost); Red and green types work from 3V supply 缺點: LED is point source, so light shaping required to make segment shapes; White and blue LEDs expensive, need >3.6V supply; Can have narrow viewing angle; Color and efficiency vary with temperature and current; Care required to achieve 50 khrs+ life.

(30) Organic (Polymer) LEDs . . 優勢: Moderate cost for small (<4“) color panels; 較 LCD觀示角寬; element 反應快; Emissive, unlike LCD color panels; RGB and mono displays; Can be built on a flexible substrate 缺點: 6V to 16V operating voltages; Differential aging effects limit life; Power consumption high for matrix panels >128 x 64.

(31) Glow Discharge (Plasma) Indicators . . 優勢: Wide operating temperature range; No phosphor (gas choice sets discharge color); Long (40khrs+) life; AC or DC operation (indicators); Unregulated DC operation (digits) 缺點: Obsolete but for specialized applications; Not suitable for small, detailed display; 40V to 150V operating voltages.

(32) Plasma Display Panel . . 優勢: Very bright, thin, color display; Available in large (30"+) sizes; Wider viewing angle than TFT color LCD 缺點: Phosphors limit life (5khrs to 20khrs); 150V+ operating voltages; Moderate power consumption; Entire display must be exercised to avoid phosphor differential aging effects.

(33) Vacuum Fluorescent Display . . 優勢: Wide operating temperature range; Long (40khrs+) life; Wide viewing angle; Very bright, attractive, typically green display; Very easy custom segment shapes, sizes; Different colored segments easy; 12V grid/anode voltage versions available 缺點: Filament supply (±8% typ. tolerance) required ; 10V to 60V grid/anode operating voltages; RGB displays available, but expensive; Phosphors other than green limit display life.

(34) 結語 . . . 關鍵; miniaturization of patterning, and the development of large-sized substrates creating RGB layers by deposition that needs high-accuracy, microscopic shadow masks. organic materials are very expensive, and the improvement of usability of organic materials is also a hurdle to be cleared for cost reduction..

(35) 軟體功能 . 透鏡設計程式的結構描光優化. . 數學計算（影像處理）. Fourier transform; filtering; deconvolution. . 量測與模擬實驗.

(36) 程式的功能限制 1. Specification: little help can be expected here, and none is available.(經驗,專利查詢) 2. Gaussian design: 基礎性構思 (No program exist yet). 3. Thin lens, primary aberrations, thickening: Some simple programs listed in books are of help. 4. Finite ray tracing: The commercially available software becomes useful now. 5. Optimization: A lot of help. 6. Design evaluation: Depend on theory applied..

(37) performed functions: . . . Finite and paraxial ray tracing Tilted, decentered, cylindrical, and toric surfaces,&General aspherics Geometrical spot diagrams Glass catalog Third-order aberrations and chromatic aberrations Elementary optimization.

(38) Defining the PSF Functional dependence of f in x (position vector) : f = f （x） For linear distortion system, the blurred information: b(x) =. ∞. (ξ) h (x,ξ) dσ (ξ). ∫∫ f. −∞. General distortion function: h = h (x,ξ) is defined as point spread function (PSF)..

(39) The background. f (ξ). h (x,ξ). b(x).

(40) PSF & OTF for. an ideal imaging case. EFL = 24.99;. F/# = 10.

(41) The drops of PSF and OTF through W-C (typically with α=10) Normalized SR.

(42) PSF and OTF through W-C in case of α= 20 Can we measure This?.

(43) Questions How is the size and accuracy of the PSF we have measured??. ?? How much we could recovered?.

(44) Difficulties 1. 2.. nonlinearly blurred transform, Digital filtering. − i 2πmn mP DFT ( d ) = ∑ d ( ) exp( ) N N m =0 N −1. the exponentials used in this equation do not obey the Nyquist criterion. because They don't stop at N but continue up to N -1..

(45) whenever to truncate a Fseries. overshoot.

(46) 3. what is the # of R we may have ?.

(47) CDM Optics. “BEFORE”. Optical Engineering “BAG OF TRICKS”. “AFTER”. •Annular •Annulargeometry geometry •Path Folding •Path Folding •Digital •DigitalDiffractives Diffractives •Volume •VolumeHolography Holography •Aspheric •AsphericElements Elements Compound refractive lens. Primary Design Constraints:. Folded hybrid lens. Derived Constraints:. - 5mm overall track (1st surface to sensor). - Image field size h = 1000x pixel pitch δ. - 0.1 rad (5.7°) field of view. - Focal length f = h / tanθ ~ 10,000x δ. - 0.1 mrad resolution (1000x700 pixels) - Visible wavelengths - 35 mm effective input aperture. (e.g.: 3.6 μm pixels → f = 36 mm). - Effective aperture: Deff = 2(router2 – rinner2)1/2 (e.g., if router = 30 mm, rinner = 24.3 mm, Deff = 35 mm).

(48) Fabricated folded optic Diamond machined blank. Patterned Aluminum-coated lens. 5 mm thickness. Input aperture. Test pattern. Incandescent illumination. lens. Overcoated absorber Reflective asphere. δx CMOS Sensor. 2.5 m. stage. 2592x1944 CMOS sensor.

(49) . Optical architecture of the band-splitting solar cell developed by the DARPA project shows how a single dichroic mirror splits the spectrum between two junctions with different absorption bands. The design can be extended to four semiconductor converters..

(50) . Nipple” arrays that simulate the surface of antireflective moth eyes can be fabricated from a molding process that starts with the selfassembly of spherical silica particles in a polymer suspension. Atomicforce-microscope images show the shape of the arrays based upon different reactive-ion etch times for the molded replicas; specifically, 0 seconds (top left), 20 seconds (top right), and 45 seconds (bottom left). The depth profiles for the nipple arrays vary considerably (bottom right), and determine the reflectivity of the resultant motheye antireflection coating..

(51) . A “big” lens imaging system built for use with an electron-beam microscope at the University of California at San Diego shows a splitting architecture that splits the image into four quadrants (top). A single arm of the system shows the complex path that the photons take through multiple lens elements (bottom)..

(52) quantum computing matching the local Zeeman splitting. oscillatory field (electron spin resonance ).

(53)

(54) DATA RATES Pulse code modulation (PCM) is a time-sampled or multiplex data system that is appropriate for use with optical communication systems. The communication link bandwidth is based upon the criterion that only the presence or absence of a pulse is required rather than that pulse magnitude accuracy is necessary. Each sample is taken from the analog signal channel and "encoded" into a series of N binary digits, referred to as a "word.".

(55) Information Handling The number of discrete values, M, obtained for an N-bit system is given by M = 2N #/bit Using 12 bits, for example, 4096 discrete numbers, 0 through 4095, can be transmitted. The accuracy of such a system is given by Accuracy = ( 1 – 1/ M ) x 100% which for a 12-bit system is 99.976%. . . Encoders are capable of 16-bit accuracy..

(56) Data-stream format . . . . Extra pulses for synchronization included. Synchronization patterns generally comprise 2 to 5% of the data transmitted. The sampling rate required is of the first consideration. It has been empirically determined that the link bandwidth should be from one-half to threefourths of the bit rate frequency. The presence or absence of a pulse rather than pulse shape / voltage levels is necessary to be determined..

(57) Example In case of information bandwidth for each channel is 5 kHz, 10 data channels are to be multiplexed, and the accuracy required is 99%. Find the (PCM) link frequency response required. M can be solved from: M = 1 /( 1-accuracy/100 ) = 100 And, taking log. Of both sides of M = 2N, i.e. N = int(log M/log2) + 1 = 7 bits ∴ 8-bit will provide an accuracy better than required (=99.61%, as a check).

(58) Data transform . . Since each channel is sampled once / frame. According to the sampling theory, 2 samples / cycle of the maximum data channel frequency are required; this results in a frame rate of ; 2 x 5 = 10 kHz. Two bits are used for word synchronization, so the word length is 10 bits. One word is used for frame synchronization, so each frame contains 11 words. Thus, the bit rate is given by :. The link frequency response is : 0.5 X bit rate = 550 kHz.

(59) digital image compression techniques . . Image compression can be lossy or lossless A good compression scheme is always composed of many compression methods..

(60) Basic concept of compression . . Amount of data ≠ Amount of information In effect, the objective is to reduce redundancy of the image data in order to be able to store or transmit data in an efficient form. How to compress visual information? Exploit the: . Spatial redundancy Temporal redundancy (for video) Psycho-visual redundancy.

(61) Data Information . Different quantities of data used to represent same information . . 同一語意父母的表達結構不同. Redundancy . if a representation contains data that is not necessary. . Compression ratio. N1 CR = N2. . Relative data redundancy. RD = 1−. 1 CR.

(62) Brief guide to image files (BMP, GIF, PNG, JPG) . . . . BMP (BitMaP) files are the Windows "standard" image file format. (16 bit--very bulky and unsophisticated). GIF (Graphics Interface Format): limited amount of colors (in general a maximum of 256 colors / image). PNG (The Portable Network Graphics): lossless format, a good format for saving images. JPG (Joint Photographic Group): designed for storage of photographs & in fact used in most digital cameras. It is lossy, but can choose how much detail you lose, great for sending photographs via email..

(63) PNG and GIF details compared.

(64) Build Optical Systems Precision and Accuracy Optical tools Assembly, Alignment & Testing Collimating and Focusing Beam-to-Beam Alignment .

(65) Precision vs. Accuracy. random. Precision not Accuracy. Accuracy not Precision. Precision and Accuracy.

(66) Dictionary definitions . . . Accurate means "capable of providing a correct reading or measurement." In physical science it means 'correct/standard'. A measurement is accurate if it correctly reflects the size of the thing being measured. Precise means "exact, as in performance, execution, or amount. "In physical science it means "repeatable, reliable/refinement, getting the same measurement each time." We can never make a perfect measurement. The best we can do is to come as close as possible within the limitations of the measuring instruments..

(67) Type of errors 1.. 2.. 3.. 4.. Accidental: uncontrollable, may reduced with ‘least squares’ Systematic: not grouped around the standard, correction table needed Short-term: short duration, repeated observations may helpful Constant: errors from all readings, use different procedures.

(68) Unit . 英制? 米制? 公制?. SI, the modern metric system of measurement SI base units : time, space The metre : Originally defined by a prototype object meant to represent 1⁄10 000 000 the distance between the poles and the Equator. Today: 1⁄299 792 458 of a light-second The second : Duration of 9,192,631,770 periods of the radiation transition between 2 hyperfine levels of the ground state of the caesium 133 atom at a temperature of 0° Kelvin. The fraction 1/31,556,925.9747 of the tropical year for 1900 January 0 at 12 hours ephemeris time Si-prefix .

(69) 設計規格與工程規格 . . . . 設計過程中, 程式均選用固定的數值, 所得結果便形成設計(design)規格。在正常製造過程中, 其製作精度必須有一定的寬溶度, 則形成所謂的工程(nominal)規格。前述兩種必須吻合, 才能完成產品, 也是公差訂定的基礎。系統公差總值由各元件公差分攤。部分元件公差的量可能由系統來吸收。.

(70) SURFACE QUALITY MIL-O-13830A Scratch - dig number; for example 60-40 以 mil 或 um 為單位 ∴ 數值越小精度越高 . 其他：staining and edge chips. Overall cosmetic inspection also includes defects within the material, such as bubbles and inclusions, including striae..

(71) System criteria . Wavefront Magnification Boresight Distortion Spot size/resolution chart User defined.

(72) decentration. Related ： EDGE TREATMENT (seamed /ground edge) Bevels PRISM ANGLE ACCURACY.

(73) 影響光學品質的關鍵機構因素. θ. d. a Despace. Decenter. Tilt.

(74) 光機工程產生的誤差. 組裝誤差類別: EDGE CHIPS; CEMENTED INTERFACES; BEVELS；.

(75) Thickness Axial sequential. Global reference.

(76) 變焦軌跡在移動過程中保持精密度. B/S. Focusing. Zooming.

(77) Opto-mechanics.

(78) Zoom 技術的巧思. Control rod 1 & 2 Cam cylinder ↑ moving cell 2. Moving cell 1. control rod 3.

(79) The. Cam.

(80) Optical Tools. . The need for producing work to closer tolerances quickly and economically makes optical instruments particularly important in plants, Providing a means of checking machine tool alignment deviations from expected tolerances.. . Advantages: easy to access; more practical to match a required tolerances; accuracy is beyond limitations of other measuring devices or methods; physical contact is not desired..

(81) Surface Flatness.

(82) Power and Irregularity. A. ½λ. A.

(83) Interferometer. with tilted fringes.

(84)

(85) Computer aided interferogram.

(86) Foci-meter.

(87) Optical bench.

(88) Uncertainty in Measurements. . Significant Figures percent error.

(89)

(90) 準線望遠鏡.

(91) 讀數. 數位顯示.

(92) 分劃讀取.

(93) 望遠鏡的應用.

(94) •雷射技術應用量測 interferometry 測距 Range finding 正線 alignment 指標 designation .

(95) Statistical tolerancing . . . 公差的妥善規劃能在一定的產品規範中, 有效地降低成本。某些因素未必會在設計階段納入。統計運算模擬, 能發揮一定的作用, 減少試行生產的循環次數, 節省支出。虛擬(virtual)生產程式, 有助於新興生產的快速規劃, 也是累積經驗數據的最佳方案。.

(96) 公差計算準則 2. I = U p' = 1− (. 2π. λ. φ dxdy (∫∫φdxdy) ∫∫ ){ − } 2. 2. 2. A. 2. A.

(97) 離焦容許差估算. 2 =λ /197. ∴ α = ± 0.245λ.

(98) 系統性能解讀. 以 MTF 為主.

(99) In Situ Alignment autofocusing. . Mainly because our eyes are easily fooled into believing an image on a ground glass screen is sharp when, in fact, it's slightly blurred.. Sure, we can tell it’s unsharp when we see an enlarged print or a slide － when it’s too late.. .

(100) 自動對焦方法 . 三角量測亮度對比時序取樣像差校正解像率辨識.

(101) Two classical approaches to focusing: 1). 2). . By actually analyzing the subject and bringing an image into focus on a focusing screen. By measuring the subject distance using a system completely apart from through -lens observations. Auto-focus systems are based on these same two focusing systems: 1) sharpness measurement: or 2) distance measurement..

(102) Classical approaches.

(103) Echo ranging. Camera sends out ultrasonic chirp signal which bounces back from the subject. Time between is measured by camera. Camera's control circuit then adjusts the lens to proper focus departed on time interval..

(104) 自動對焦模組.

(105) Sharpness comparison The prism elements of the vibrating grating split the image-forming rays from the left and right-hand halves (L) and (R) of the camera lens into separate section that are diverted to four photo-detectors (la, 2a, 1b and 2b). The vibration of the grating shifts the distribution between the cells for the left and for the right half..

(106) 相位與對比偵測.

(107) Critical Fiber Coupling Light propagation is highly directional Tolerances very tight: micron or sub-micron range Rework often required after fixing.

(108) Fiber alignment procedure and methods . Passive alignment of a fiber in a. groove etched in e.g. Silicon.

(109) Active alignment algorithms Hill-Climb algorithm (one axis at a time) . . Recursive triangulation algorithm • works with two axes. 8 7 5. • step size gets smaller 6. 4 3 2 1.

(110) 生產成本比較(傳統與自動化).

(111) 生產規劃過程支出分配.

(112) Tolerancing on Optical System.

(113) 規劃階段的重要性 . . 越早期偵知風險, 對計畫越有利。匯集專業人員, 組成評估團隊, 快速解決困難問題, 容易發揮效果。成員包括：市場及計畫主管—充分了解產品需求, 得以正確付託各分項專業適當的權責。專業人員需提供完整技術資訊, 各领域間之溝通管道務須暢通。任務執行盡可能並行處理。.

(114)

(3) 自然光源 Planck&rsquo;s law

(3) 自然光源 Planck’s law