Multi Source Fit to Image Data

 

In case an in advance knowledge of a sample of sources is known in an image, a least square fitting of a point spread function model to the image data may be advantageous in order to determine the source parameters. If sources lie in a crowded region (and/or) do overlap in the point spread function a multiple source fit is recommended to be applied. A command FIT/MULTI_SOURCE is available to fit up to 6 overlapping sources simultanously.

The method can be applied to ROSAT PSPC and HRI data. The point spread function model used in the least-square fit is a simple Gaussian approximation. The command syntax is:

FIT/MULTI_SOURCE mlcnts edit

  A parameter file mlcnts.par is created and offered to be edited. The parameters in the parameter file mlcnts.par are listed in Table 5.11.

  
Table 5.11: Parameter file for FIT/MULTI_SOURCE

The parameters of the parameter file mlcnts.par are described in more detail:

• INPUT_IMAGE : name of the input image to which the chi-square fit is applied
• LDETECT_LIST : name of result table of local detection applied to INPUT_IMAGE
• BG_IMAGE : name of background image created from INPUT_IMAGE
• BG_SCALE : scale factor applied to the background image (usually 1)
• ENERGY_CONV : energy conversion factor in units of . Values of the energy conversion factor (ECF) for different spectra can be found e.g. in the ROSAT Call for Proposals.
• EXPOSURE_FLAG : flag to use (T) or not to use (F) the exposure image for the calculation of the source count rate and flux. If set to EXPOSURE_FLAG = F, the count rate and flux will be calculated from the TIM_INT descriptor information of the image INPUT_IMAGE
• EXPOSURE_IMAGE: name of the exposure image. This image is evaluated if the flag EXPOSURE_FLAG is set to T
• EXPOSURE_SCALE: scale factor for the exposure image (it is normally 1)
• X_IMA_CENTER : x-coordinate in image of on-axis position (ima_pix)
• Y_IMA_CENTER : y-coordinate in image of on-axis position (ima_pix)
• ENERGY_KEV : mean energy of the image data (keV)
• MIN_OFFAX : minimum off-axis angle (arcmin) of sources to be considered in the fit
• MAX_OFFAX : maximum off-axis angle (arcmin) of sources to be considered in the fit
• ML_THR : minimum maximum likelihood for source to be accepted
• RESULT_IMAGE : name of result image (background image with fit of sources added
• RESULT_TABLE : name of fit result table (cf. Table 5.12)
• FIT_POSITION : flag to keep source position free (FIT_POSITION = free) or fixed (FIT_POSITION = fixed) in the least-square fit

The RESULT_TABLE has the following information (cf. Table 5.12):

  
Table 5.12: Info from RESULT_TABLE of FIT/MULTI_SOURCE

• INDEX is a source index. It is the sequence number of the source
• ID_SRC the index of the input source
• ID_SAT the index of the satellite of the source
• BG_MP are the background counts per image pixel at the source position
• X_IMA is the source x-position (in units of image pixels) determined by least squares fit
• Y_IMA is the source y-position (in units of image pixels) determined by least squares fit
• CTS are the source counts determined by least squares fit
• EXI_ML is the likelihood of existence of the source
• XERR is the error in source x-position (in units of image pixels) determined by least squares fit
• YERR is the error in source y-position (in units of image pixels) determined by least squares fit
• CERR is the error in source counts determined by the least squares fit
• EXT is the extension of the source (in units of image pixels). It is set to EXT=0, no extension is fit to the source
• EXTERR is the error of the extension of the source (in units of image pixels). It is set to EXTERR=0, no extension is fit to the source
• FLUX is the flux of the source (in units of erg cm sec ). The flux is calculated as :RATE * ENERGY_CONV with the energy conversion factor ENERGY_CONV specified in the parameter file
• EFLUX is the error in the flux of the source (in units of erg cm sec ). The error in the flux is calculated as :ERATE * ENERGY_CONV with the energy conversion factor ENERGY_CONV specified in the parameter file.
• RATE is the source count rate (in units of counts s )
• ERATE is the error in the source count rate (in units of counts s )
• CHISQR is the chi-square of the fit
• CUTRAD is the cut out radius of the source (in units of ima_pix)
• OFFAX is the off-axis-angle of the source (in units of arc minutes)
• FWHM is the Full width at half maximum (FWHM) of the point spread function of the source