Measuring intensity in watts rather than lumens

Similar documents
Radiometric and Photometric Measurements with TAOS PhotoSensors

Optical basics for machine vision systems. Lars Fermum Chief instructor STEMMER IMAGING GmbH

The Importance of Wavelengths on Optical Designs

COLOUR INSPECTION, INFRARED AND UV

IMAGE SENSOR SOLUTIONS. KAC-96-1/5" Lens Kit. KODAK KAC-96-1/5" Lens Kit. for use with the KODAK CMOS Image Sensors. November 2004 Revision 2

LECTURE III: COLOR IN IMAGE & VIDEO DR. OUIEM BCHIR

APPLICATIONS FOR TELECENTRIC LIGHTING

Lecture 2 Digital Image Fundamentals. Lin ZHANG, PhD School of Software Engineering Tongji University Fall 2016

Properties of LED considering museum lighting

CS559: Computer Graphics. Lecture 2: Image Formation in Eyes and Cameras Li Zhang Spring 2008

Preventive Conservation and Energy conservation. Units of light, Perception of colour, Energy used by lighting.

CRISATEL High Resolution Multispectral System

Lighting Terminologies Introduction

Digital Image Processing

Applications for cameras with CMOS-, CCD- and InGaAssensors. Jürgen Bretschneider AVT, 2014

BROADCAST ENGINEERING 5/05 WHITE PAPER TUTORIAL. HEADLINE: HDTV Lens Design: Management of Light Transmission

CS6670: Computer Vision

NFMS THEORY LIGHT AND COLOR MEASUREMENTS AND THE CCD-BASED GONIOPHOTOMETER. Presented by: January, 2015 S E E T H E D I F F E R E N C E

Photometry for Traffic Engineers...

Imaging Optics Fundamentals

Vision Lighting Seminar

LIGHT AND LIGHTING FUNDAMENTALS. Prepared by Engr. John Paul Timola

Chapter 2: Digital Image Fundamentals. Digital image processing is based on. Mathematical and probabilistic models Human intuition and analysis

Speed and Image Brightness uniformity of telecentric lenses

Visual Perception of Images

CPSC 4040/6040 Computer Graphics Images. Joshua Levine

Radiometry vs. Photometry. Radiometric and photometric units

Properties of LED considering museum lighting

Some Aspects of Light Pollution in the Near Infrared

The secrets of higher sensitivity CCTV cameras

e2v Launches New Onyx 1.3M for Premium Performance in Low Light Conditions

General Imaging System

OLYMPUS Digital Cameras for Materials Science Applications: Get the Best out of Your Microscope

Applied Machine Vision

Optimizing throughput with Machine Vision Lighting. Whitepaper

Advanced Camera and Image Sensor Technology. Steve Kinney Imaging Professional Camera Link Chairman

Camera Overview. Digital Microscope Cameras for Material Science: Clear Images, Precise Analysis. Digital Cameras for Microscopy

Ideal for display mura (nonuniformity) evaluation and inspection on smartphones and tablet PCs.

Technical Notes. Integrating Sphere Measurement Part II: Calibration. Introduction. Calibration

SCHOTT Lighting and Imaging. Contents. overview general lighting situations. illumination with fiber optics. direct LED illumination

INTRODUCTION TO CCD IMAGING

Make Machine Vision Lighting Work for You

Photometry for Traffic Engineers...

ME 6406 MACHINE VISION. Georgia Institute of Technology

Detector-Filter Combination Series

Applications of Optics

Camera Overview. Digital Microscope Cameras for Material Science: Clear Images, Precise Analysis. Digital Cameras for Microscopy

Sometimes the axis of the I-U-dependence are shown in reverse order. In this case the graph shows the stabilized current and measured voltage.

Fein. High Sensitivity Microscope Camera with Advanced Software 3DCxM20-20 Megapixels

From the start the main activity of our company was the development and production of infrared illuminators.

Bettina Selig. Centre for Image Analysis. Swedish University of Agricultural Sciences Uppsala University

Digital Image Processing COSC 6380/4393

Digital Camera Technologies for Scientific Bio-Imaging. Part 2: Sampling and Signal

The future of the broadloom inspection

Colorado School of Mines. Computer Vision. Professor William Hoff Dept of Electrical Engineering &Computer Science.

Digital camera. Sensor. Memory card. Circuit board

The Condor 1 Foveon. Benefits Less artifacts More color detail Sharper around the edges Light weight solution

How to Optimize the Sharpness of Your Photographic Prints: Part I - Your Eye and its Ability to Resolve Fine Detail

LEDs for Flash Applications Application Note

The Standard for over 40 Years

Image Formation and Capture

Vision. The eye. Image formation. Eye defects & corrective lenses. Visual acuity. Colour vision. Lecture 3.5

Camera Overview. Olympus Digital Cameras for Materials Science Applications: For Clear and Precise Image Analysis. Digital Cameras for Microscopy

Camera Requirements For Precision Agriculture

Capturing Light in man and machine. Some figures from Steve Seitz, Steve Palmer, Paul Debevec, and Gonzalez et al.

How does prism technology help to achieve superior color image quality?

Photometric Colorimetry

Image Formation. Dr. Gerhard Roth. COMP 4102A Winter 2015 Version 3

High Intensity LumiBright LE LED Light Engine

Kit for building your own THz Time-Domain Spectrometer

OPTICAL SYSTEMS OBJECTIVES

EC-433 Digital Image Processing

SIM University Color, Brightness, Contrast, Smear Reduction and Latency. Stuart Nicholson Program Architect, VE.

Introduction. Lighting

PRODUCT OVERVIEW FOR THE. Corona 350 II FLIR SYSTEMS POLYTECH AB

Guide to SPEX Optical Spectrometer

BTS256-PAR. Product tags: VIS, Spectral Data, PAR, Waterproof, General lighting.

Where Image Quality Begins

The Art of Light Measurement. Avantes BV Apeldoorn, The Netherlands

PRODUCTION DATA SHEET

BTS256-E WiFi - mobile light meter for photopic and scotopic illuminance, EVE factor, luminous color, color rendering index and luminous spectrum.

Introduction to Computer Vision

How interference filters can outperform colored glass filters in automated vision applications

Choosing the Best Optical Filter for your Application. 1

Imaging Photometer and Colorimeter

BMC s heritage deformable mirror technology that uses hysteresis free electrostatic

Signal-to-Noise Ratio (SNR) discussion

Using Machine Vision Cameras for Solar Imaging. Dr Stuart Green

LED Lighting Flux and Color Measurement System (LFC)

Camera Requirements For Precision Agriculture

Light-Emitting Diodes

LED Lighting - Error Consideration for Illuminance Measurement

Work environment. Retina anatomy. A human eyeball is like a simple camera! The way of vision signal. Directional sensitivity. Lighting.

WHITE PAPER. Guide to CCD-Based Imaging Colorimeters

CSE 527: Introduction to Computer Vision

Tunable wideband infrared detector array for global space awareness

Optoelectronics Data Book

Image acquisition. In both cases, the digital sensing element is one of the following: Line array Area array. Single sensor

Overview. Pinhole camera model Projective geometry Vanishing points and lines Projection matrix Cameras with Lenses Color Digital image

It should also be noted that with modern cameras users can choose for either

Transcription:

Specialist Article Appeared in: Markt & Technik Issue: 43 / 2013 Measuring intensity in watts rather than lumens Authors: David Schreiber, Developer Lighting and Claudius Piske, Development Engineer Hardware Lighting

MEASURING INTENSITY IN WATTS RATHER THAN LUMENS What are the right wavelengths for the LED lighting on any given machine vision system? It is not hard to find a suitable camera system along with the matching optical components for most machine vision applications. In many cases, the correct size and geometry are also known or already specified. However one question often remains: which wavelength is most suitable? This article shows how appropriate catalogue entries can make it simple to use and compare the non-trivial lighting technology units of measure. Now that LEDs are used almost exclusively as the light source in machine vision applications, determining which wavelength is the most suitable has taken on major importance. LEDs have a narrow-band spectral intensity distribution and how they act on the detector can vary greatly depending on the color. When viewed with the human eye for example, a red LED must produce around three and a half times as much light output to appear to be as bright as a green LED. Brightness perception is highest for green light, as is shown by the sensitivity curve for daylight vision 1 V(λ) in Fig. 1. The maximum of 683 lm/w is reached at a wavelength of 555 nm. Fig. 1: The differences between the spectral sensitivity of the human eye and a selection of standard image machine vision are shown on the graph. The machine vision system "eye" The detector in a machine vision system, particularly the sensor, is comparable to the eye. The CCS and CMOS sensors on these systems have spectrally distributed sensitivity which overlaps with the eye's sensitivity to some extent. The only difference is that the photo effect on the silicon chip has a wider bandwidth than the photo receptors in our retina. The brightness of LED components and lighting in the visual (visible) range is normally expressed in lumens (lm) which are photometric units. Lumens are used as the unit of measure because the main application of LEDs is "general use for illumination purposes" in which case the light is Technical article: Measuring intensity in watts rather than lumens 1

perceived and used by humans. In machine vision, visible light must be regarded as being only a portion of the electromagnetic spectrum and consequently light intensity should be expressed in radiometric units such as watts. Radiometric and photometric units are both listed in the Vision & Control product catalogue. This is done so that proper consideration can be given to the following parameters which have an effect on the resulting image brightness in the machine vision system: Lighting efficiency LED spectral intensity distribution Reflection/transmission of the object under evaluation Spectral transmission of the lens Spectral sensitivity of the detector It is necessary to consider the spectral and radiometric characteristics of the lighting from the eye safety standpoint as well. An evaluation is done in house by Vision & Control for all vicolux lighting systems based on DIN EN 62471. Application scenario In order to achieve high illumination intensity, selection of the "right wavelength" can be crucial, especially in high-speed inspection applications. Sufficient exposure must take place at the image sensor in a very short time. The following machine vision task will be used to illustrate the effects which changing the wavelength has on the system's image sensor. A colorless fluid is filled into vials made of colorless glass at a pharmaceutical company. A machine vision system checks the fill level and cap tilt. During a system retrofit, the machine vision unit is being optimized so that it can handle twice the cycle rate. The resolution and accuracy of the overall system must remain the same. Total-system lighting comparison Fig. 2 show a brightness comparison for the vicolux FDL60x90 transmitted light system at different wavelengths (blue, green, red and infrared) at the Sony ICX204AL image sensor under continuous light. The spectral sensitivity of the standard CCD sensor is very similar to that of the human eye. The lighting is operated at 6W at each wave length. The lens is an entocentric lens with a 25 mm focal length. The caps on the vials are about 50 mm high and about 15 mm in diameter. They are spaced at about 15 mm on a material transport system. An image scale of about 1:3 is needed to fully inspect the object. The vials are filled at a rate of about 600 units per minute, so the conveyor speed is roughly 0.3 m/s. In order not to exceed the allowable motion blur of around 30 µm (about 10 µm in the image), the maximum allowable sensor exposure time for the sensor is 100 µs. The vicolux FDL60x90 lighting system operating in flash mode is a good choice for such short exposure times. Flash and continuous modes are a standard feature of vicolux lighting systems. Fig. 2 (center) shows the results of comparative brightness testing on lighting systems operating in flash mode using the same sensor and lens (adjusted aperture setting). The LEDs on the infrared lighting system, especially at the peak wavelength of 850 nm, can be made to light up very Technical article: Measuring intensity in watts rather than lumens 2

"bright" because of the short flash times. The intensity of infrared lighting increases even faster compared to red lighting. Fig. 2:shows a comparison of transmitted light for the vicolux FDL60x90 in four different colors (blue, green, red and infrared) using the Sony ICX204AL image sensor under continuous light (top) and in 100 µs flash mode (center). The images at the bottom were taken using a MT9V034 CMOS sensor, also operating in 100 µs flash mode. (Lab photos with stationary objects) If the machine vision system has a CCD sensor, red light is advisable to produce the maximum grayscale level in the resulting image. If the system has a CMOS sensor, infrared lighting could be the better choice. Fig. 2 below shows a comparison of four illumination colors taken with the MT9V034 CMOS sensor (also 1/3 inch). In relative terms, the sensitivity of CMOS sensors is considerably higher in the near infrared range (e.g. at IR 850 nm) compared to CCD sensors. (The factors obtained are guideline values and can vary for different LED lots). Comparison using other LED types It is important to keep an open mind, however, when looking at the results of the example described above. Substantial efficiency gains, based for example on thin-film technology, have been made in recent years in InGaN chips which are used to fabricate white LEDS, especially power LEDs. InGaAlP chips for red LEDs used to be at the leading edge of technology but more efficient technologies are now available. vicolux -SFD30/9 and vicolux -42/12 series dome lights are one example. Using power LEDs as the basis of comparison, the results are markedly different from those obtained with the lighting discussed in the previous section. At a working distance of 5 mm, the illuminance of red lights in this product series is 58 klx. The figure is 98 klx for the white version. Expressed in the equivalent radiometric units, the numbers are 325 W/qm for red and 319 W/qm for white. This shows the enormous increase in efficiency compared to red lights. Technical article: Measuring intensity in watts rather than lumens 3

The comparison in Fig. 3 shows that in contrast to the first example, the white light version is brighter when a CCD sensor is used and blue light produces roughly the same results as red light. Fig. 3: Brightness comparisson for vicolux-sfd30/9 and vicolux-42/12 series lights (power LEDs) in four different colors, measured with different detectors. When making comparisons between CCD and CMOS sensors, it is important to keep in mind that amplification of the sensor signal in the camera can vary and that the figures presented above were calculated based on standardized quantum efficiency. Summary "Don't trust your eyes." The comparison of photometric data and subjective image and brightness perception can sometimes be very different from what the camera in a machine vision system "sees". It is therefore advisable to estimate brightness on the basis of radiometric units. Vision & Control provides both radiometric and photometric intensity data to make things easier for you. The CCD and CMOS comparison value which is also listed helps you select the most suitable illumination wavelength for standard sensors. For non-standard sensors, the relative sensor effect can also easily be determined with the aid of the illumination spectrum and quantum efficiency of the sensor. appeared in Markt&Technik 43/2013 Technical article: Measuring intensity in watts rather than lumens 4