Line Imaging Spectrometer Sampler (LISS)
Contents
1. NearinfraredGPSDigital XUS and if a units is disabled the identifying letter is replaced with an underscore letter A filename where both the visual and the near infrared linespectrometer are enabled could be for ex ample Winter Wheat N trial 26 VN __180N mat When the measured data has been saved the trial counter value is increased one unit When enabled LISS subtracts the interpolated dark frame from the measured data right after the download of the image from each linespectrometer If operator for some reason doesn t want to use the standard dark frames if for example using another CCD temperature it is possible to generate a new dark frame estimation When activated LISS reads the se quence of dark exposure it should use for the estima tion from the setup definition file The operator should be aware of the dark frame es timation is a very time consuming process After the dark frame estimation the operator can save the result LISS will suggest a filename starting with Darkframe followed by the date and time for the dark frame estimation This darkframe will be used until a new is estimated or LISS is restarted or the setup pro cedure is initiated Note The path and filename of the darkframes LISS should use are set manually in the setup definition file When everything is setup and initialised and ready for measurements the operator starts the measuring proce dure by activating the Take Image button by one mouse2. The latter string creates an array with 2 times 21 exposures from 10 ms to 1 01 second in steps of 50ms and then backwards It is important to have at least 10 different exposures which covers the whole range of ex pected exposures during the sampling session The darkframe procedure is initiated from LISS by selecting Darkframe in the menu scrollbar and hereafter selecting Take Darkframes Remember to perform this procedure long before the planned measuring session because this is a very time consuming task When LISS has finished the recording of dark frames and estimated the pa rameters for the dark frame estimation it is possible to inspect the result visually before saving to the hard drive After the inspection switch back to LISS and save the dark frame data file This is done by selecting Darkframe in the menu scrollbar and hereafter selecting Save Darkframes Set the path and accept the preset darkframe filename and press OK 3 5 4 The structure of the dark frame data file The dark frame data file is a Matlab structure holding all data of importance for the actual dark frame estimation see section 3 7 page 27 for more info about Matlab structures Riso R 1189 EN 26 exposures cpu_datetime cpu_datetimestamp VIS_intercept VIS_slopes VIS_temperature_setpoint VIS_HEAD_OFFSET NIR_intercept NIR_slopes NIR_temperature_setpoint NIR_HEAD_OFFSET name Figure 12 Diagram showing the content of the structure
3. oe SSet the default values of the switch buttons in the control panel All are TRUE FALSE values control_info struct autosave 0O autoexposure 1 enabledarkframes 1 oe oe Set the default parameters for the temperature regulation enable TRUE FALSE setpoint The desired CCD temperature See Universal CPU Command Structure from SBIG for more details oe oe oe temperature_regulation struct enable 1 setpoint 10 406 samp_rate 10 tp gain 1000 yee a Garni 200540 reset_brownout 0 a SSet the default parameters for CCD camera mounted with a spec trograph for the visual VIS and near infrared NIR light re spectively exposure_time initial exposure amp line_start and line_len defines the spatial sampling area with start location of the line plus number of lines further pixel_start and pixel_len defines the spectral sampling in terval with start pixel plus number of pixels further sSee Universal CPU Command Structure from SBIG for more details VIS_take_image_conditions struct exposure_time 10 Vine start Opie line_len 242 pixel start 0 ca pixel_len 375 enable_decs 1l de_restore 0 gt lt tabg state Opecs Ris R 1189 EN abg_period 6000 Wdest_butter Lya rauto dark Opasg readout_mode 1 open_shutter 1
4. 159 VIS pixel en dja s NIR line start 0 4 lt NIR_line_len 242 INIR pixel Stare 8l pera NIR_pixel_len 1 oe SSet the exposures used for the interpolation of the dark frame equation dark frame A exposure time B used for estimating a dark frame at specific exposure The exposure array is common for both CCD cameras darkframe_exposures 1 5 10 40 10 100 SSet the path and filename to the darkframe to be used when meas uring with the darkframe subtraction enabled darkframe name d sampling darkframe_26 Apr 2001 11 50 44 mat oe SEnables GPS routines in LISS and sets the port number of the GPS unit enable TRUE FALSE Ris R 1189 EN 21 22 TimeGPS enable 1 TimeGPS gps_port COM4 ole oe Set the parameters for the auto sampling routine enable TRUE FALSE repeatcycles total number of measurements interval sets time in minutes between measurements ole ole oe autosample struct enable 0O repeatcycles 5 interval 2 oe ole Set the parameters for the Canon Digital IXUS enable TRUE FALSE zoom integer zoom number from 0 4 show_image defines which part of the image to be shown in the LISS preview screen oe ole oe digital_ixus struct enable 0 goom 4 6 show_image 1 1600 1 1200 Set the po
5. VIS _data temperature If a unit has been disabled in the setup definition file the field with the related device information is not included in the resultstruct 3 8 Shutting down LISS and equipment When shutting down the CCD cameras after a measuring campaign it is impor tant to close LISS first in order to send a shutdown command to the camera CPU This turns off the camera s cooling and otherwise prepares it for being powered down It s a good idea to wait 30 seconds or so after closing LISS be fore turning off the power because it gives the CCD cooler time to warm up Santa Barbara Instrument Group 1998 Ris R 1189 EN 4 References Bennedsen B S and P Rasmussen 2001 Plant spectra classification by neural net works In G Grenier and S Blackmore editors Agro Montpellier Montpellier France June 2001 151 156 Borregaard T 1997 Application of imaging spectrocsopy and multivariate methods in crop weed discrimination The Royal Veterinary and Agricultural University Department of Agricultural Science 1 174 Borregaard T H Nielsen L Norgaard and H Have 2000 Crop weed discrimination by line imaging spectroscopy Journal Of Agricultural Engineering Research 75 389 400 Guyot G F Baret and S Jacquemoud 1992 Imaging spectroscopy for vegetation studies In F Toselli and J Bodechtel editors Imaging Spectroscopy Fun damentals and Prospective Applications ECSC EEC EAEC Brussels and Luxenborg
6. 12 2 2 3 Grey Flow of actions when overruling the standard dark frame interpolations Uesvies cb cesb Vien Beedle e a es sded oa aeae a a tokit ei iia 12 2 2 4 Dark blue Flow of actions when auto sampling are enabled 0 5 12 2 2 5 Light blue Free inspection of the measurement performed 1 cc1cccee 13 3 LISS SOFTWARE USER S MANUAL sesssessesesesecceeeroeseoseesoesesececossosoeseseeeee 14 3 1 INSTALLING AND SETTING UP MATLAB TO HOST LISS cccccccceeeeeseeeeeees 14 Sided The LISS JOM CHS oeei iaeiei ds heated elbes gi ges bie os a e AA A hates does 14 3 1 2 Placing the LISS folder and setting up Matlab 15 3 1 3 Setting up the PC enabling Matlab to interface the Canon IXUS 0 15 3 2 THE LISS USER INTERFACE ris apnee t ar etie p EE EER 15 3 21 SAMPLING AAA oesie oe reaot an A E a E AEE EE E A 16 3 2 2 Time and GPRS 1nfO nee ei te a eich e e e E EE a 17 32 3 RGB Photos aao at EE tel thes E e a iain Haein eect San a aS 17 3 24 Control panel ereien ata e e a E a atten its 17 3 2 3 Camera 1 Gnd CAM era 2 20 atenists io atbech dea er a a aai 18 3 3 INITIALISATION OF THE LISS SOFTWARE cssssesessescseseneeseeeececsenesseseeceesenes 19 3 4 THE SETUP DEFINITION FILE ccccsssseseseeesecsenseseseeececsenscsesesececsensssessseeeesenes 20 3 5 USE OF DARK FRAMES nino ioina o E n E E E E S 23 3 5 1 How to reduce the dark frame effect to the quality 23 3 5 2 Procedure for using da
7. NIR_take_image_conditions struct exposure_time 10 line start Ores line_len 242 pikel Start Oy aaa pixel_len 375 enable odes 1 74 dc_restore 0O abg state 0 84 abg_period 6000 dest burter 1y ees Tauto dark 0ps readout_mode 1 open_shutter 1 oe SSet the default parameters for the auto exposure routine setpoint The maximal pixel value the auto exposure routine is aiming for maxlimit If the pixel value exceeds this value the exposure time is halved and auto exposure reruns S VIS_line_start and VIS_line_len define the spatial auto ex posure area with the first line location plus number of lines further for the spectrometer in the VIS range S VIS_pixel_start and VIS_pixel_len define the spectral auto exposure area with start pixel plus number of pixels further for the spectrometer in the VIS range amp NIR_line_start and NIR_line_len define the spatial auto ex posure area with the first line location plus number of lines further for the spectrometer in the NIR range amp NIR_pixel_start and NIR_pixel_len define the spectral auto exposure area with start pixel plus number of pixels further for the spectrometer in the NIR range autoexposure struct setpoint 45000 maxlimit 60000 VIS line start 0 0 VIS_line_len 242 VIS pixel start
8. Rasmus N Jorgensen and Peter Rasmussen Risg R 1189 EN 1 Introduction Farmers have for centuries by visible inspection been able to tell how well their crop is performing at different locations within the field For example a dark green plant is as a rule of thumb performing better than a light green plant The visual evaluation performed by the farmer is to a large extent relative because the evaluation is based on the color differences within the specific field The human eye is extremely good in directly comparing color differences but is however not very good in absolute color measurements Hatfield and Pinter 1993 In other words if the farmer was asked to evaluate plants one by one without any reference his capability to separate by color would be significantly reduced The human eye and a standard color camera RGB camera are in their own way imaging spectrometers at the limited visible wavelength region which ranges in wavelength from approximately 400nm to 750nm Hagman 1996 A human eye registers the basic tristimulus values which are relatively broad bands and is capable of separating up to 36 000 different colors Likewise the RGB camera is designed to register basically the same three bands to be able to differentiate colors and imitate the human eye Spectral Imaging Ltd 1999 transmission specular reflection diffuse reflection absorption 7 upper pidermis cuticle ao S palisade mec i spon
9. Rasmussen D Plant Nutrition and Soil Fertility Laboratory Department of Agricultural Sciences The Royal Veterinary and Agricultural University DK 1871 Frederiksberg C Copenhagen Denmark Present address Plant Environment Interactions Plant Research Department Ris National Laboratory DK 4000 Roskilde Denmark e mail Rasmus Joergensen risoe dk d Plant Environment Interactions Plant Research Department Riso National Laboratory DK 4000 Roskilde Denmark February 2002 Riso National Laboratory Roskilde Abstract The objective of this report is to develop and describe the software for a Line Imaging Spectrometer Sampler LISS to perform measurements of spectra combined with a digital RGB photo of a measurant Secondly this report should enable users to perform measurements with the system The measuring devices consist of 2 16 bit cooled SBIG ST 6 CCD cameras each equipped with an Imspector imaging spectrograph for measuring reflec tance in the visual and the near infrared range respectively a digital camera and a differential global positioning device dGPS The controlling and data col lecting tasks are developed as a Graphical User Interfaced GUI hosted by Mat lab Release 12 from Mathworks This GUI enables the operator to perform measurements from all devices simultaneously together with notes specific for the measurant and store all the data in one Matlab data structure The software includes dynamic exposure
10. click 3 2 5 Camera 1 and Camera 2 The right half of the LISS interface is reserved to the CCD cameras with the VIS and NIR linespectrometers in the upper and lower part of the area respec tively The layout for both CCD cameras is similar and will be reviewed to gether There is one text box named Exposure time 1 100s showing the exposure used for the specific CCD camera in 1 100 seconds The operator can prior to a measurement alter these exposure values and they will be used at the next measurement if auto exposure is disabled Risg R 1189 EN At the right side of the text box is an info box telling the user the actual com port used baud rate and the CCD header offset In the left part is another info box with measuring information about the CCD temperature and the maximal and minimal pixel values The measuring image is presented at the middle where the vertical and horizontal axis is the spatial and spectral axis respectively Next to the image is a legend showing the color representing the different pixel values 3 3 Initialisation of the LISS software The main flow of actions and decisions is described in section 2 2 1 and in Figure 5 on page 11 This section is more specific in the description of the ini tialisation procedure of the LISS software The first step before initiating LISS is to control that all the measuring units are connected to the computer and turned on If a GPS unit is connected this uni
11. information from equally spaced points on this line In this way the other dimension of the CCD array retains its spatial function but the other axis is used to record the spectrum in each spatial point The resulting image from the target is certainly limited just one narrow line but still it is much more than just a point Spectral Imaging Ltd 1999 Riso R 1189 EN Ss Pa a WN N icn a hy Figure 3 Imaging spectroscopy spectral imaging across a two dimensional plate An area detector is able to register position in the X coordinate axis wavelength and intensity The Y axis of the plate is measured by acquiring line images while moving the plate in the Y direction Spectral Imaging Ltd 1997 Equipment using the latter principles has been designed and manufactured by VTT Finland now commercialised by Spectral Imaging Ltd www specim fi Hyvarinen et al 1998 The Department of Agricultural Sciences Agricultural Engineering RVAU Denmark has since 1995 been working with the line imaging spectrometer technology from Specim The main focus has been on crop weed discrimination and it has been proven that this technology is capable of this kind of discrimina tion Bennedsen and Rasmussen 2001 Borregaard 1997 Borregaard et al 2000 In 1998 R N Jorgensen RVAU Denmark started a Ph D project aim ing on the line imaging spectrometer equipment at the Department of Agricul tural Sciences RVAU with the aim of es
12. of actions when overruling the standard dark frame in terpolation Gr1 Manual activation of the procedure collecting a new set of dark frames estimating the parameters for a new dark frame estimation Gr2 Procedure collecting dark frames from the CCD imaging cameras after a sequence of exposure time given in the setup definition file After collecting the dark frames linear regressions are performed for each pixel estimating the rela tionship between exposure and dark noise when no light is entering the well Gr3 User decides if the structure containing the intercept and slope for each pixel from dark frame regression should be saved Gr4 The structure containing the intercept and slope for each pixel from dark frame regression is saved as a Matlab structure into a folder containing the sam pling data 2 2 4 Dark blue Flow of actions when auto sampling are enabled dB1 Test if the CCD temperature is 1 C from the pre set value If the tem perature is valid the measuring procedure is performed dB2 Measuring procedure is performed dB3 All the measured data is saved in one file as a Matlab structure contain ing all data of value for later data processing Risg R 1189 EN dB4 Test if the auto sampling procedure is at the end of the measuring se quence given in the setup definition file If not the procedure continuous with dB1 when the time since the last measurement was initiated exceeds the sam pli
13. to alter Riso R 1189 EN 1 1 Aim The objective of this report is to develop software for a Line Imaging Spec trometer Sampler LISS to perform measurements of spectra combined with a digital RGB photo of a measurant Secondly this report should enable users to perform measurements with the system Risg R 1189 EN 2 Material and Method This section will briefly describe the equipment that the LISS software operates from and the hardware needed to build the whole system Furthermore the over all mode of action for the whole system is explained with main focus on the LISS software 2 1 Material The hardware and the software from which the LISS software operates together with the sampling units are illustrated to the right and at the front page The whole system can be separate into three categories The hardware platform the software platform and the measuring units The content of these is described below 2 1 1 The Hardware Platform The computer hosting the software is a Chi cony MP 979 PII 266 with 192 MB Ram The laptop has one 9 pin RS 232 port and 2 USB ports A Xircom USB Port Expansion is connected to the USB port giving access to additional 4 RS 232 ports and 1 USB port extra The 4 additional RS 232 ports were setup with serial port id s from COM3 till COM6 2 1 2 The Software platform The operating system is windows 2000 ver sion 5 0 2195 or higher The Software plat form for the LISS softwar
14. were sent to the Nav Guide through pin 8 in the RS 232 port The dGPS strings were sent to the Nav Guide from a CSI MBX 3 Beacon dGPS receiver 2 2 Method The flow of actions and decisions normally carried out when using the LISS software will be described in the following There is no need to know the actual user interface of LISS In fact this section should give a good overview and ba sis to understanding the overall structure of the LISS software Figure 5 is a flow chart illustrating the flow of decisions and actions during a measuring session with the LISS software The flow of decisions and actions needed are grouped in boxes of different colors The colors represent the fol lowing groups e Green The normal flow of actions when using LISS e Red Flow of actions if the setup procedure fails e Grey Flow of actions when overruling the standard dark frame in terpolation e Dark blue Flow of actions when auto measuring is enabled e Light blue Free inspection of the measurement performed The following text will refer to the boxes in Figure 5 according to their ID s placed in the upper part of each box For example G1 refers to the first green box with the content Start LISS from Matlab Risg R 1189 EN Gr1 Gr2 Take Sequence of Activate Take Dark Frames and Dark Frame perform linear regression A Save Frame gt m NG Gr4 Save Dark os o Frame p A regr
15. 145 165 Hagman O 1996 On reflections of wood Wood quality features modelled by means of multivariate image projections of latent structures in multispectral images Lule University of Technology 1 281 Hatfield J L and P J Pinter 1993 Remote sensing for crop protection Crop Protec tion 12 403 413 Holst G C 1998 CCD arrays cameras and displays JCD Publishing Hyvarinen T E Herrala and A Dall Ava 1998 Direct sight imaging spectrograph a unique add on component brings spectral imaging to industrial applications SPIE symposium on Electronic Imaging 3302 165 175 Nielsen H P Rasmussen R Devantier and B S Bennedsen 1999 Potentials for measuring crop data by analysis of spectral information Institut f r Geod sie und Geoinformatik Fachbereich Landeskultur und Umweltschutz Universitat Rostock Interner Bericht Heft Bill R Grenzd rfer G and Schmidt F vol 12 83 92 Santa Barbara Instrument Group 1993 Universal CPU command structure Supports ST 4X ST 5 and ST 6 cameras Santa Barbara California Santa Barbara California 1 31 Santa Barbara Instrument Group 1998 CCDOPS Version 4 0 for Windows 95 Mac intosh and DOS Santa Barbara Instrument Group Santa Barbara Santa Bar bara Instrument Group 1 104 Searcy S W 1989 Machines see red and so much more Agricultural Engineering November December 10 15 Spectral Imaging Ltd 1997 Imspector imaging spectrograph user ma
16. Downloaded from orbit dtu dk on Dec 15 2015 Technical University of Denmark DIU o Development and design of a line imaging spectrometer sampler LISS A user manual Jorgensen Rasmus Nyholm Rasmussen P Publication date 2002 Document Version Publisher final version usually the publisher pdf Link to publication Citation APA J rgensen R N amp Rasmussen P 2002 Development and design of a line imaging spectrometer sampler LISS A user manual Denmark Forskningscenter Risoe Risoe R No 1189 EN General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights e Users may download and print one copy of any publication from the public portal for the purpose of private study or research e You may not further distribute the material or use it for any profit making activity or commercial gain e You may freely distribute the URL identifying the publication in the public portal If you believe that this document breaches copyright please contact us providing details and we will remove access to the work immediately and investigate your claim Riso R 1189 EN Development and Design of a Line Imaging Spectrometer Sampler LISS A User Manual Rasmus Nyholm Jorgensen and Peter
17. R 1189 EN 23 24 Rie Ala FB TE Equation 1 Estimation of dark frame nparg depending on exposure integra tion time tq A is a 242 x 375 array containing the slope and B is an array holding the intercept for each pixel is assumed Vnpark and independent of El Jy vee ee Em Figure 10 Illustration of the range and distribution of the intercepts left and the slopes right of a dark frame estimation based on Equation 1 and per formed by LISS The temperature of the CCD array was 10 C The red circle shows the position of a hot pixel When studying Figure 10 it is clear that the major contributor to the dark frame is the systematic error caused during the readout The image showing the slopes seems entirely low and blue but the colorbar showing the range of the slopes tells that a few pixels have a significant dark current These pixels could cause significant error to the measurement if the slope parameter is not taken into account 3 5 2 Procedure for using dark frames in LISS When using the estimated dark frames during a measuring sequence with LISS it is important to control and evaluate if the standard dark frame estimate suites the current settings of the CCD cameras The operator must ensure that the fol lowing parameters and settings match each other e The range of exposures of the dark frame covers the expected range of the measuring sequence The image mode or resolution The CCD temperature The CCD he
18. ad offset If one of these parameters does not fit each other when comparing the dark frame data file with the CCD camera settings a new dark frame data file must be generated before measuring The operator can inspect the conditions under which the active dark frame data file was estimated This is done by selecting Darkframe in the menu scrollbar and hereafter selecting Show Darkframe info and then compare them with the actual setting of the CCD cameras The dark frame info is presented by LISS as shown in Figure 11 Risg R 1189 EN Figure No 2 d sampling darkframe_26 Apr 2001 11 50 44 mat File Edit View Insert Tools Window Help Caml VIS VIS_intercept cpu date time 26 Apr 2001 11 50 44 HEAD_OFFSET 166 Temperature_setpoint 10 Exposuretimes 1 2 5 10 40 100 200 Cam2 NIR cpu date time 26 Apr 2001 11 50 44 HEAD_OFFSET 167 Temperature_setpoint 10 Exposuretimes 1 2 5 10 40 100 200 EES T a Gad yr BBY Od od Sd Ag o 5g wy OY BY Y LSC RODMS OORS tos Figure 11 Screen dump of LISS presenting the dark frame in use 3 5 3 Procedure for recording a new darkframe Before initiating the dark frame procedure in LISS it is important to give the CCD cameras time to cool down and stabilise after the setup procedure has been processed The different exposures are set in the setup definition file for exam ple with the following string darkframe_exposures 1 5 101 101 5 1 se section 3 4 page 20
19. d Note that the Sw does not verify if the camera is setup to return JPEG This interface has been made by engineer Johannes Nyholm Jorgensen Copen hagen 2001 Risg R 1189 EN 31 Bibliographic Data Sheet Rise R 1189 EN Title and authors Development and Design of a Line Imaging Spectrometer Sampler LISS A User Manual Rasmus Nyholm Jorgensen and Peter Rasmussen ISBN ISSN 87 550 2706 7 87 550 2708 3 Internet 0106 2840 Department or group Date Plant Research Department 6 February 2002 Groups own reg number s Project contract No s 1310116 0 9 Sponsorship The Ministry of Food Agriculture and Fisheries Denmark Pages Tables Illustrations References 31 0 14 14 Abstract max 2000 characters The objective of this report is to develop and describe the software for a Line Imaging Spectrometer Sampler LISS to perform measurements of spectra combined with a digital RGB photo of a measurant Secondly this report should enable users to perform measurements with the system The measuring devices consist of 2 16 bit cooled SBIG ST 6 CCD cameras each equipped with an Imspector imaging spectrograph for measuring reflectance in the visual and the near infrared range respectively a digi tal camera and a differential global positioning device dGPS The controlling and data collecting tasks are developed as a Graphical User Interfaced GUI hosted by Matlab Release 12 from Mathworks This GUI enables the operator to pe
20. d the LISS interface are set according to setup file G6 Controls if the setup procedure succeeded without any error G7 If no errors occurred during the setup procedure the user needs to decide if the standard dark frame interpolation should be used or another dark frame file should be loaded or if new parameters for the dark frame interpolation must be estimated Further the user decides if dark frames are subtracted from the linespectrometer measurements by setting a radio button on or off G8 Activates the measuring procedure manually G9 Tests if auto sampling is enabled If not the measuring procedure for one measurement is started G10 Tests if the CCD temperature is 1 C from the pre set value If the tem perature is valid the measuring procedure is performed G11 User decides if further inspection of the data measured is needed G12 User decides if data should be saved G13 All the measured data is saved in one file as a Matlab structure contain ing all data of value for later data processing If autosave is enabled this is per formed automatically after point G10 2 2 2 Red Flow of action if the setup procedure fails R1 Displays an error message if there is any explaining the user what went wrong during the setup procedure R2 The user decides whether to control the measuring devices or alter the setup definition file R3 Program waits 2 seconds before continuing 2 2 3 Grey Flow
21. d the cool ing procedure is enabled For detailed information se Santa Barbara Instrument Group 1993 1998 In case the digital RGB camera is enabled the pre set zoom together with other parameters preparing the unit is set When the setup procedure has been performed with no errors the dark frame and measuring related buttons are enabled and LISS is now ready to perform measurements Risg R 1189 EN 19 20 3 4 The Setup Definition File The setup definition file is a Matlab function which constructs a structure con taining all the settings necessary to make the LISS software work Below is a copy of the function as it will appear when opened for editing by LISS in the default Matlab editor function liss_struct liss_setup SThis file is a Matlab function which returns an initialisation structure with the properties required for the LISS program The main structure which is assembled from other substructures is Galled Liss struct An example of a substructure is sampling_info below This struc ture defines the default values of for the text boxes in sampling data area of LISS The syntax of building a structure is first to name it and then to fill in the content For TRUE FALSE conditions 1 TRUE and 0O FALSE sampling_info struct operator Rasmus Nyholm Jgrgensen terial tDemonstration yass VEE LALeoUnE Oyie nobel LISS manualy ass notez notes T noted
22. d the max value is max zoom wide end RELEASE Take a picture e g myCanon release Note that when you have issued a release command then you HAVE TO give a store command before you can make a new release Constraint due to PowerShot software limitations SETPATH Set the path where the picture information shall be stored e g myCanon store C temp Note that the path shall exist else an error is returned when you issue a store command STORE Store the picture with a specified file name Do not specify a file extention e g myCanon store picture Extention cannot be changed It is jpg GETRELPAR GET RELease PARameters Those are static and is defined in PowerShot 3 5 documentation Search after REMOTE RELEASE PARAMS The struct is also defined in the PowerShot file INC ReleaseCtrl h Example of command x y myCanon GetRelPar x and y is mandatory By static I mean they do not change after the initialization of the camera after you have given the first myCanon commana The first value in y is the first in the struct DEBUGINFO Get debug information Used to debug software e g my Canon debuginfo All commands return at least one parameter which is the return code The value can be used to evaluate the success of the command execution The values are defined in file globalDefines h Example of command x myCanon Release x is optional unless noted for the comman
23. e and GPS info RGB photo Control panel Camera 1 Camera 2 Riso R 1189 EN 15 16 LISS Line Imaging Spectrum Sampler Yersion 4 0 e sh SD 1 4 E A 24 SO tel x File Edit view Insert Tools Window Help Darkframe posure time 1 100 s f 27 976 COM4 115200 Baud Head_offset 166 Rasmus Nyholm J rgensen temperature 9 36370 max pens 40247 3418 LISS manual P d sampling Browse cpu date time 12Jul2001 15 34 14 Uitine 122219 Exposure time 1 100 s f 16 244 COM6 115200 Baud Head_offset 167 Latitude 5540 15893325 N H 6 3 Longitude 1218 51758190 E GPS fix GPS fix min 149 2433 Ready Take image finished file Mstart 7 E H a Amana BIUss Line maging 5p Microsoft Photo Edtor ESET UDRAS 1535 Figure 8 Screen dump of the LISS graphic user interface right after a measu rement has been performed The individual content of these subsections is described below 3 2 1 Sampling data The sampling data area placed in the upper left corner of the screen is used for typing trial and or trial specific information This information is stored together with the data from the measuring units in order to ease the organising and ana lytical task after the measuring session The purpose of each text box is listed below Text box name Purpose Operator Here the person doing the measuring type his her name Trial Type a unique name which can ident
24. e case then disable the dark frame subtraction and perform the operation later during the analytical phase If the operator enables the auto save feature LISS saves the measuring data right after the measurement has ended and the operator only has to select the Take Image button between each measurement If auto save is disabled the op erator must activate the Save Images button between each measurement 3 6 2 Single shot measuring procedure Single shot measuring is used when the operator wants to control the measure ment manually When measuring in plot trials where the content in the text boxes of the data changes frequently single shot is the optimal procedure Fur ther it is possible to inspect the quality of the measuring result before saving to the hard disk The measuring procedure is relatively simple When ready to measure select the Take Image button in the control panel to initiate the measuring procedure When the measurement is finished the operator can inspect the data directly within the Matlab environment The data from the last successful measurement is available from a global variable named imdata The variable imdata is a structure similar to the structure which will be saved as the measuring data file Riso R 1189 EN except for additional three substructures These substructures is autoexposure darkframeexposures and autosample which are not of importance for analysing the measuring data Please see section 3 7 below fo
25. e is Matlab release 12 from Mathworks Inc equipped with an Extended Real Time Toolbox version 3 1 from Humusoft enabling Matlab to access the RS 232 ports 2 1 3 The Measuring Units The sampling units are 2 SBIG ST 6A CCD Imaging Cameras from Santa Barbara In strument Group with a resolution of 375 by 242 pixels The pixel size is 23 x 27 mi crons and the total CCD array area is 8 6 x 6 5 mm The readout electronics utilise a 16 bit analogue to digital A D converter Figure 4 Picture of the LISS measuring devises 2 SBIG CCD cameras equipped with a NIR and a VIS imspector imag ing spectrograph respectively and a Canon Digital IXUS lo cated between these and double correlated sampling to achieve very high sensitivity and photometric accuracy Each of these cameras is mounted with an Imspector Line Imaging Spectrograph for the visual range model V7 and the Near Infrared range Riso R 1189 EN 10 model N10 respectively The orientation of the CCD Imaging Cameras is 242 pixels along the spatial axis and 375 pixels along the spectral axis The digital color camera for the documentation photo is a Canon digital IXUS with 2 1 Mpixels supporting remote control coded in C via the USB port The GPS receiver is a modified Nav Guide sending out a GPGGA NMEA string at 4800 baud containing the position of the measuring equipment The GPS receiver was modified so the RDS functionality was disabled and the dGPS string
26. e main folder which has seven subfolders as shown in Figure 6 H Liss_v4 Backup_of_initial_LISS_settings Canon_IxUs 5 GPs LISS Manual SBIG_CameraControl SBIG_CameraControl_subroutines Temp Figure 6 Illustration of the folder structure for the LISS application e The main folder LISS v4 contains main files for the LISS application plus the setup definition file and the standard dark frame regression file e The subfolder Backup_of initial_LISS settings contains a backup of the setup definition file and the standard dark frame regression file e The interfaces files between Matlab and the Canon Digital IXUS camera are placed in the Canon_IXUS folder The manual is placed in the subfolder LISS Manual in PDF format The GPS subfolder is a Matlab interface for picking up a NMEA GPS string from a RS 232 port and desiccating the GPGGA string into positioning in formation e Subfolder CameraControl and CameraControl_subroutines contain a Mat lab interface to the cpu unit controlling the line imaging spectrometers A detailed description of the commands is given in the cpu unit manual Santa Barbara Instrument Group 1993 e The Temp subfolder is used by the Canon IXUS procedure Riso R 1189 EN 3 1 2 Placing the LISS folder and setting up Matlab The LISS main folder and all subfolders must be copied to the Matlab folder on a computer with Matlab R12 and the extended real time tools box version 3 1 installed By using
27. enting the date and time or an image and a third might hold the actual temperature of the CCD array Figure 13 shows the full content of the structure holding the data from a meas urement To read such a file into Matlab type load path and filename and a data structure named resultstruct is loaded Risg R 1189 EN 2 28 operator trial trialcount note1 note2 autosave note3 autoexposure note4 enabledarkframes path enable cpu_datetime etimestamp exposure_time ee line_start line_len pixel_star enable pixel_len HEAD_OFFSET enable_dcs dc_restore abg_state enable abg_period setpoint dest_buffer auto_dark samp_rate p_gain i_gain reset_brownout enable gps_port raw_signal 9ps time time_formatted enable latitude gps zoom latitude_formatted PARAMS longitude longitude_formatted fix_quality satellites darkframe_exposures Figure 13 Diagram showing the content of structure in the Matlab measuring data file The green colored boxes represents structures and the transparent boxes contain arrays of different Matlab formats readout_mode open_shutter IIHI After the structure has been loaded into Matlab information regarding for ex ample the actual CCD temperature for the CCD array in the visual range at measuring are fetched by typing resultstruct
28. ession Figure 5 Flow chart illustrates the decisions and flow of action during a meas uring session with the LISS software The green colored boxes show normal flow of actions The red boxes show when something goes wrong The dark blue boxes illustrate automated sampling The light grey boxes show the decisions needed concerning dark frames and the light blue box illustrates the flow for performing manual inspection of data sampling before saving 2 2 1 Green The normal flow of actions when using LISS G1 The first step is to start the LISS software from the Matlab software which serves as host application G2 The next step is to select the path or folder to which the measuring data files should be saved into G3 When LISS has been started the setup definition file will be copied to the folder into which the measuring data is saved if the folder does not already con tain any setup file Hereafter the standard setup definition file is opened to con trol if any setting needs to be changed Riso R 1189 EN 11 12 This step can be overruled if an existing setup file is already in the selected path e g if continuing a measurement series G4 When the setup definition file has been saved and closed the setup proce dure for the whole system can be initiated manually G5 The setup procedure tests if measuring units which are enabled in the setup file are working Furthermore the measuring units an
29. gy parenchyma inter cellular space respiration hole stomata Figure 1 Schematic representation of the interaction of incoming radiation with leaf tissues Guyot et al 1992 The color of a plant leaf is the end result after the light from the leaf has en tered the eye which integrates the incoming light in three broad spectral bands as mentioned earlier The light from the leaf is mainly reflected light from an external light source like the sun Some of the incoming radiation is transmitted and some is absorbed in the leaf depending on the structure and composition of the leaf as illustrated in Figure 1 Therefore the composition of the reflected light entering the eye is a code telling something about the leaf composition Risg R 1189 EN 400 500 600 700 800 nm Figure 2 Schematic illustration of a typical reflection spectrum of a green plant in the visual VIS and near infrared NIR range of light when illumi nated by the sun The three graphs R G B illustrate how an RGB camera in tegrates light from three broad bands centred red 650 nm green 530 nm and blue 460 nm simulating the human color vision After Guyot et al 1992 Searcy 1989 However studying Figure 2 it becomes clear that the information contained in the three broad bands are rather limited when compared to the character of a leaf reflectance spectra in the visual and near infrared light One way to obtain the true reflectance spectru
30. ify the specific trial Trial count Trial count is a counter identifying the number of the measurement After each measurement the number is automatically increased by one The number can be changed during a trial session but be careful not to over write prior measurements if decreasing the numbers Note 1 4 These boxes are meant for sampling specific information like ID of the area being measured sampling conditions and or comment to when something noticeable occurs Path Here the path to the folder where measurements and setup files should be stored is given The button beneath the text box named Browse can be used for interactively selecting the path When using the browse facility identify the folder where measurements and setup files should be stored and press to OK button Note Remember to have at the end of the path string Riso R 1189 EN 3 2 2 Time and GPS info This section gives information about the date and time according to the com puter performing the measurements If the GPS unit is enabled positioning info is collected at the time of sampling and stored in the data file The GPS data stored is UTC time Latitude Longitude GPS fix and the number of satellites used The latitude and longitude are in WGS84 format GPS fix tells if position is using dGPS Dif GPS fix only GPS GPS fix or if the position couldn t been found no fix 3 2 3 RGB photo The RedGreenBlue photo is a conventional d
31. igital photo of the area including the narrow line of the measurant that the line spectrometers measures The photo is mainly used to ease the human understanding of the response from the linespectrometer The picture is stored in a separate file in JPEG format with a similar name as the Matlab file containing all the other measured data 3 2 4 Control panel The control panel as the name indicates is the place where the operator interacts with LISS during the measuring session Just next to the title of this sub section LISS informs the operator the status and the result of an operation Below a brief description of each control function is given Control name Purpose and Action Auto save on off When enabled LISS saves the measured data right after the measurement Therefore the operator only has to activate Take image between each measurement Auto exposure on off When enabled LISS takes a pre set exposure and based on the highest pixel value obtained LISS calcu lates the optimal exposure so that the highest pixel value in the measurement is close to the set point given in the setup definition file This function minimises the risk of overexposed measurements and secures at the same time optimal use of the binary resolution of the CCD cameras The pre set values for both linespectrometers ex auto exposure area auto exposure time maximal pixel response etc are defined in the setup definition file Edit Create setup_file After se
32. in the Matlab dark frame data file The green colored box represents a structure and the transpar ent box contains arrays of different Matlab formats 3 6 Measuring with LISS When the setup definition file has been altered for the specific measuring ses sion and the LISS setup procedure has ended successfully the program is ready for measuring It is important to have a robust procedure which must be strictly followed in order to minimise operator errors 3 6 1 Preparations and decisions prior to measuring Before initiating single shot or multi shot measurements type all vital informa tion concerning the actual measurement into the text boxes within the sampling data area of LISS During the structuring and analytical phase of the collected data these information can be of vital importance The operator must decide whether to enable the auto exposure functionality or not In most cases it is advisable to enable this feature in order to obtain maxi mal resolution of the pixels values from the 16 bit CCD cameras especially un der varying light conditions Under stable light conditions it is possible to in crease the measuring rate a little by disabling auto exposure It must also be decided if the dark frame should be subtracted the measured images from the CCD cameras The normal procedure is to enable this feature provided the parameters for the dark frame estimation suits the actual sampling settings of the CCD cameras If this is not th
33. l language in providing a flexible environment for technical computing The open architecture makes it possible to use Matlab and its companion toolboxes to explore data create algo rithms and create custom tools Cand Agro R N Jorgensen has carried out the main idea and the design of the overall structure of LISS as a part of his Ph D project B Sc Hort P Rasmussen has coded and debugged the LISS application The Department of Agricultural Science Agricultural Engineering at The Royal Veterinary and Agricultural University Denmark is gratefully acknowl edged for lending out their line imaging system full time during the months of development We also want to thank the Canon PowerShot Developer Pro gramme for the technical support during the coding of the remote control for the Canon Digital IXUS camera Furthermore we are grateful to engineer Johannes Nyholm Jorgensen for programming the interface file enabling remote control of the Canon Digital IXUS camera directly from the Matlab command prompt Finally we especially thank all those who read draft copies of the manuscript Professor D Sc Ph D Niels Erik Nielsen Research Professor Simon Black more Ph D Stud Lene Krol Christensen and Associated Professor Henning Nielsen The Royal Veterinary and Agricultural University Department of Ag ricultural Sciences Professor D Sc Gunnar Gissel Nielsen Ris National Labo ratory Plant Research Department Thursday 21 February 2002
34. lecting the folder into which data should be stored activation of this function is the next step in the initialisation of LISS described in section 3 3 When activated LISS looks for setup definition file within the data folder and opens the file in the default Matlab editor This enables the operator to control and alter the settings before activating the setup procedure Af ter editing the file in the editor remember to save and close the file before continuing If LISS doesn t find a setup definition file in the data folder it copies a standard setup definition file and opens it for editing Setup When activating the setup procedure LISS tests if the enabled measuring units are present and initialises these according to the setup definition file in the work Risg R 1189 EN 17 18 Control name Purpose and Action Save Image Enable dark Take dark frames Save dark frames Take image ing folder If this procedure succeeds LISS is ready to measure after the CCD cameras have cooled down their CCD temperature to the pre set temperature Note LISS can work with both spectrometers dis abled Selecting this button activates the saving procedure LISS constructs a filename based on the content of text boxes Trial Trial Count and Note 1 plus a string identifying the enabled units and separates them with an underscore letter _ When all measuring units are enabled the string content is VNGD alias Visual
35. m is to use a spectrophotometer which is able to measure both VIS and NIR in very narrow bands A standard spectrophotometer is usually only able to measure from a small integrated area at a time This means that if one wants to measure the optical spectrum from different spatial locations of an object either the measuring instrument or the target has to be moved or scanned Spectral Imaging Ltd 1999 By using a spectrometer valu able spectral information can be achieved but the spatial or imaging informa tion is lost Therefore an imaging spectrometer is needed Then how to realise an imaging spectrometer in practise Basically some kind of component capable of separating different wavelengths prism grating or filter optics which are used to image spatial features and a detector for measuring the acquired image The possibilities are limited considering the pre sent available components and their properties Technically it is not possible by using a standard 2D CCD matrix element a semiconductor chip with a grid of light sensitive elements used for converting light images as in a television camera into electrical signals to measure simultaneously the 2 dimensional surface as a matrix plane X Y plane the spectrum components with high resolution and the intensity of each spectral element However it is possible to measure the spatial informa tion from a line with specified length and small but finite width and the spec tral
36. ng interval given in the setup definition file 2 2 5 Light blue Free inspection of the measurement performed Lb Allows the user to inspect the data before saving This can be done by running various Matlab routines outside the LISS application analysing the data before deciding whether to save the result or not Risg R 1189 EN 13 14 3 LISS software User s Manual This section is build up as a regular user manual which will enable a novice operator to use the LISS software This manual prerequisite the operator is fa miliar with using Matlab Further a minimum knowledge regarding program ming terminology of Matlab is needed The section explains and deals with the following subjects e Description of how to install and setting up LISS in order to be able to op erate within the Matlab environment e A detailed description of the LISS user interface with concern to individual functions and how to operate these e Point by point explanation of the setup definition file How to perform the initialisation procedure Discussing the use of dark frames and how to generate new parameters for dark interpolation e Finally the actual measuring procedure is described in details 3 1 Installing and setting up Matlab to host LISS The LISS application does not have any automatic installation procedure Therefore the installation must be done manually 3 1 1 The LISS folders All routines and data necessary are contained in on
37. nual Spectral Imaging Ltd Finland Spectral Imaging Ltd 1999 Imaging Spectrometry http www specim fi aboutim html Spectral Imaging Ltd Risg R 1189 EN 30 5 Appendix Technical info The LISS program itself is based on a Graphic User Interface GUI in form of a Matlab figure named Liss fig and is loaded by the m file iss m The maincode and the control centre are placed in the m file called liss_action m which is called from Liss fig by use of callbacks Several sub m files are used by the liss_action m in controlling the SBIG CCD cameras located in the subdirectory SBIG_CameraControl An overview of all the routines used by LISS in Matlab is shown in Figure 14 LISS_v4 liss m Te at toe et Tole rane fos Taare LISS fig etae a EE 8 ed liss_darkframe m ERT EET T a EN U EASES E E EE na GPS gps_gpgga_find m gps_gpgga_find m Canon_IXUS myCanon dil Canon specific files liss_action m SBIG_CameraControl get_activity_status m sa agp bg eg rl get_cpu_info m get_line m ERR ARRAN get_readout_peak m A hell get_rom_version m mw r get_uncompressed_line m zts A ppp read_blank_video m aN Pewee read_thermistor m AN A hhh regulate_temp m at a Ph set_com_baud m ANa R Wh se
38. of the two CCD cameras ensuring optimal use of 16 bit range under unstable illumination conditions A routine handling dark frame subtraction in a robust manner minimising the effect of hot pixels is also included This report enables a novice user to perform measurements with LISS rela tively easy enabling the user to put the main focus at the analytical phase and not on the measuring phase Keywords imaging spectrograph spectrometer Imspector Matlab reflectance ISBN 87 550 2706 7 87 550 2708 3 Internet ISSN 0106 2840 Print Pitney Bowes Management Services Denmark 2002 Contents PREFACE oo suscsisc3ias sovest tact sc3ias seceassunesatets to scetecddets E E duds secsasdiceustedeces 4 i INTRODUCTION rsccescecsesscssvscciccd stess enrere or Seoane onroro oea iesceseswees seusessetecescsascesastes 5 1 1 AM a ar ETE A EE EERE E ET A E N ARATE Orv 8 2 MATERIAL AND METHOD e eessesssesesocsssoeosoesssoossoososoesesososoesesoosoeososoesesossses 9 2 1 MATERIA Lists cities itodo o E E es cases ledge seit E E ERS EEE T EE EE 9 241 TheHardware Platforma a A ade 9 22 The Software Play Ov sian ccscsstesceliesseg Suk caahcas etn ia eactedeen assented 9 QS ThesMeasuring Ons it iicteiiecetias egstesk evehcon apiece einen lak eet tec ation tebes 9 2 2 METHOD Toreo E as ev eset ace es tates EAE E R EEE A 10 2 2 1 Green The normal flow of actions when using LISS H 2 2 2 Red Flow of action if the setup procedure fails a n
39. osure and normally assumed Poisson distributed Holst 1998 The streaky pattern caused by systematic errors during the readout by the build up of dark current in the storage area of the CCD and by hot pixels in the stor age area 3 5 1 How to reduce the dark frame effect to the quality Since dark frame noise is always present when measuring it should be sub tracted before further analysis The simplest approach is to record a dark frame just before the actual measurement with common exposure and then subtract this frame from the measured frame This is however a rather time consuming procedure since the CCD camera have to perform two exposures every time the measuring exposure time changes As mentioned above sudden random hot pixels can occur in the CCD array The effect of these pixels can be reduced significantly by averaging several dark frames or by recording several dark frames with different exposures and then make a linear regression depending on exposure In preliminary investigations it was shown that the dark frames in fact could be described well with linear re gression for each pixel This is in fact the standard procedure for LISS when estimating the dark frame for the CCD camera according to Equation 1 How ever the variation within some of the pixels is so high that negative slopes can be obtained If this happens the slope for the actual pixel is set to zero and the intercept is set to the average value of all exposures Riso
40. r a detailed description of the imdata structure In order to get access to imdata in Matlab after measuring type global imdata Having access to imdata it is possible to run additional functions analysing the data before saving the result When ready to store the measurement at the hard disc select the Save Images button and save the data 3 6 3 Multi shot sampling procedure When the operator for example wishes to do several measurements from one area through time multi shot sampling is the function to use As shown in sec tion 3 4 above this feature is set in the setup definition file and initiated by the setup procedure The measuring data file is similar to the single shot file Since the measurements are done over a larger time span it is advisable to enable the auto exposure feature to overcome varying light conditions When the setup procedure has finished the auto sampling is started by select ing the Take Image button During the measuring LISS will display the time left till next measurement and the sequence number out of the total measurements to perform 3 7 The structure of the measuring data All data of importance for a measurement is stored in a structure A structure in Matlab is array of data containers which each is a field having a fieldname The fields of a structure can contain any kind of data For example one field might contain a text string representing the operator name another might con tain an array repres
41. rform measurements from all devices simultaneously together with notes specific for the measurant and store all the data in one Matlab data structure The software includes dynamic exposure of the two CCD cameras ensuring optimal use of 16 bit range under unstable illumination conditions A routine handling dark frame subtraction in a robust manner minimising the effect of hot pixels is also included This report enables a novice user to perform measurements with LISS relatively easy enabling the user to put the main focus at the analytical phase and not on the measuring phase Descriptors INIS EDB Available on request from Information Service Department Riso National Laboratory Afdelingen for Informationsservice Forskningscenter Ris P O Box 49 DK 4000 Roskilde Denmark Telephone 45 4677 4004 Telefax 45 4677 4013
42. rk frames in LISS ccccccccscssseeeeseetessseeeenseesenseesenaeeaes 24 3 5 3 Procedure for recording a new darkframe cssc 25 3 5 4 The structure of the dark frame data file onn 25 3 6 MEASURING WITH LISS soii EE E E E Ei s 26 3 6 1 Preparations and decisions prior to measuring 26 3 6 2 Single shot measuring procedure esseere sere 26 3 6 3 Multi shot sampling procedure 27 3T THE STRUCTURE OF THE MEASURING DATA s s ssssesssseserterestssesestrresesesseseserten 27 3 8 SHUTTING DOWN LISS AND EQUIPMENT c cccccccssesesessesceeeneeseseeecssseseeseneeees 28 4 REFERENCES cidos oeseri cacleudeussesesstoracsSdeestsessscdeesdecesesesisegresdabesesecsnese 29 5 APPENDIX TECHNICAL INFO seseseseseereroeseseccseerosoesesecoseesosoesececosrosorseseeeee 30 5 1 APPENDIX USING CANON IXUS DIRECTLY FROM MATLAB ccseesseeeeees 31 Risg R 1189 EN Preface This manual should enable a novice user to perform measurements with the Line Imaging Spectrometer Sampler system LISS relatively easy This will further enable the user to put the main focus on the analytical phase and not on the measuring phase Further the reader will be able to build the system from scratch if needed But most important it should inspire the reader to build a new and better system based on the main principle of the LISS system Matlab was selected as the host application for LISS due to its integration of mathematical computing visualization and powerfu
43. rt number to which the CCD camera are connected for the visual caml and the near infrared cam2 respectively enable TRUE FALSE caml port COM2 caml enable 1 cam2 port COM3 cam2 enable 1 oe SDONT CHANGE IN THE FOLLOWING LINES All substructure are collected into the main structure liss_struct struct sampling_info sampling_info eont rol inte control into aw VIS_take_image_conditions VIS_take_image_conditions NIR_take_image_conditions NIR_take_image_conditions temperature_regulation temperature_regulation autoexposure autoexposure darkframe darkframe GPS GPS autosample autosample digital ixus digital 1 xus ras reami Caml panics cam2 cam2 oe Risg R 1189 EN 3 5 Use of Dark Frames Taking an image with the CCD cameras in total darkness the readout is ex pected be zero for all pixels in the CCD array This is however not the case as illustrated in Figure 9 i 4 SER LAN es PT A d li 50 100 150 200 250 300 350 Figure 9 A dark frame recorded with the CCD camera in total darkness with an exposure of 100 msec The CDD was cooled down to 10 TC There are several reasons for the readout not to be zero The main reason to the noisy pattern of the dark frame is dark current electrons which are thermally generated The number of thermally generated dark current electrons is propor tional with exp
44. t head offsetmi SBIG_CameraControll_subroutines take_image m ail sy add_checksum m byte_extract m decompress m dissection m packet_check m jq read_packet m Figure 14 Schematic Illustration of the LISS application and its routines and GUIs The names of the directories and subdirectories are in red The Matlab functions are given in the white boxes For example liss_action m use all the functions in the directory SBIG_CameraControl and a few function in SBIG_CameraControl_ subroutines Within the SBIG_ directories only read_packet m call packet_check m which is located within the same directory level Risg R 1189 EN 5 1 Appendix Using Canon IXUS directly from Matlab The digital Canon IXUS camera can be used to take pictures directly via the command line The current directory in Matlab must be set to the Canon_IXUS directory containing the interface file mycanon dil Furthermore the camera must be turned on before executing the latter file The syntax for the Canon in terface file can be called by typing mycanon help in the Matlab command prompt as shown below gt gt mycanon help Info Camera ready Mex file Sw version 01 00 01 PowerShot Sw version used 3 5 Commands HELP Get help regarding list and syntax of commands e g myCanon help STATUS Get the status e g myCanon status ZOOM Valid range is 0 4 e g myCanon zoom 1 Where 0 is tele end an
45. t should be turned on a few minutes before starting LISS allowing it time to find the satellites and the dGPS signal needed for the positioning Start up Mat lab R12 equipped with the extended real time toolbox version 3 1 and the driv ers for Canon Digital IXUS installed and type iss Provided that LISS installa tion has been done properly the LISS software should now be running Select the path into which the measured data should be stored Hereafter open the setup definition file from the control panel and check if all connected measuring units are enabled Change the setup definition file so it fits to the purpose of the specific measuring session Save the file with the suggested path and filename When all measuring units are connected and turned on and the setup definition file has been prepared the setup procedure can be started This procedure per forms several tasks which will be described The first step is to load the whole setup definition file into the memory of the computer Hereafter it is checked if LISS can establish contact with all enabled sampling units For the two CCD cameras several settings is altered within the camera CPU which always act as slave The first step is to increase the baud rate communica tion of camera CPU from 9600 baud to 115 kilo baud Then the electronic head offset of the current CCD array is being determined for each camera The CCD temperature set point and the parameters for the regulation are set an
46. tablishing a nitrogen index in winter wheat grown under natural conditions Preliminary results made in corporation with the Department of Agricultural Sciences RVAU indicate great potential within this area Nielsen et al 1999 During this work it was realised that the software included with the cameras mounted at the two linespectrometer units could only control one camera unit at the time Furthermore it was not possible to take a digital photo simultaneously with the linespectrometer since all sam pling units used individual controlling software Finally a GPS position stored together with the measurements could ease the organisation of the data signifi cantly Based on the latter it was decided to design a new application named Line Im aging Spectrometer Sampler LISS which should be able to fulfil the follow ing demands e It should be possible to control and collect data from all measuring units simultaneously e Possible to add specific information for each measurement into the data file e Data from all units containing the measured data and settings should be stored in one structured file e Should be able to control the following sampling units e 2 CCD cameras ST 6 from SBIG with RS 232 connection e digital color camera with USB connection which can be interfaced to Matlab e GPS unit with RS 232 connection e Windows2000 compatible e Able to run at a laptop with only 1 USB port e Software should be flexible and easy
47. the Matlab Path Browser the main folder shown in Figure 7 is added to the default path of Matlab Add Folder to Path Liss_v4 H Canon_IXUS X gt cms Figure 7 The LISS folder named LISS _v4 which must be added to the path of Matlab 3 1 3 Setting up the PC enabling Matlab to interface the Canon IXUS After the drivers delivered with the Canon Digital IXUS camera have been in stalled a few additional steps are needed to enable the interface from Matlab to the IXUS drivers The first step is to set the registry information necessary from the Matlab command line that can make Windows call to launch the IXUS drivers In the Current Directory bar set the path to C matlab Liss_v4 Canon_IXUS Then type RUNDLL32 EXE PSDKREG DLL Install in the command window For further information on how to control the Canon Digital IXUS camera di rectly from Matlab see section 5 1 on page 31 When the settings described above have been done the LISS application can be executed be typing liss in the Matlab command window 3 2 The LISS User Interface This section will give a detailed description and overview of the LISS user in terface shown in Figure 8 The LISS interface is optimised to a screen resolution of 1024 x 768 pixels enabling the user to see all options at the same time The LISS screen is split up in six subsections according to their function The sub sections are Sampling data Tim
Download Pdf Manuals
Related Search