Locates and centroids image features


This routine processes a list of images, locating and centroiding image features (such as stars) which have groups of connected pixels above threshold values.

Connected groups of pixels are accepted as objects if they have more than a minimum number of pixels. Such groups may be rejected if they contact the edges of the data array.

Threshold estimation is performed using either a percentage data point (i.e. the value for which this percentage of pixels have a lower value) or by using a standard deviation and background value determined by fitting a gaussian to the data histogram.


findobj in minpix outlist


If this parameter is TRUE then a threshold determined by this routine for each of the images will be used. If FALSE then you will be prompted for a threshold value for each image. [TRUE]
The minimum fraction of the image area (expressed as a percentage) which is required in the peak bin when forming the histogram. Ensuring that at least one bin contains this fraction of counts is intended to make sure that the image histogram is well sampled. This increases the robustness of mode estimates made from the histogram but decreases the accuracy. Only used if USEPER is FALSE. [2.5]
On exit this parameter contains a list of the number of objects detected in each input image. This may be useful in scripts where the values can be accessed using the KAPPA (SUN/95) PARGET command.
A list of image names which contain the data components to be scanned for image features. The image names should be separated by commas and may include wildcards.
Name of the CCDPACK logfile. If a null (!) value is given for this parameter then no logfile will be written, regardless of the value of the LOGTO parameter.

If the logging system has been initialised using CCDSETUP, then the value specified there will be used. Otherwise, the default is "CCDPACK.LOG". [CCDPACK.LOG]

Every CCDPACK application has the ability to log its output for future reference as well as for display on the terminal. This parameter controls this process, and may be set to any unique abbreviation of the following:
  • TERMINAL – Send output to the terminal only

  • LOGFILE – Send output to the logfile only (see the LOGFILE parameter)

  • BOTH – Send output to both the terminal and the logfile

  • NEITHER – Produce no output at all

If the logging system has been initialised using CCDSETUP then the value specified there will be used. Otherwise, the default is "BOTH". [BOTH]

The minimum number of non-BAD pixels which must be present in a connected group for acceptance as an image feature. [6]
The name of a file to contain the names of the output position lists. The names written to this file are those generated using the expression given to the OUTLIST parameter. The file may be used in an indirection expression to input all the position lists output from this routine into another routine. [FINDOBJ.LIS]
The number of standard deviations above the background that should be used as the threshold. This parameter is only accessed if the USEPER parameter is FALSE and a gaussian is being fitted to the background. [5]
The names of the output lists.

These may be specified as list of comma separated names, using indirection if required, OR, as a single modification element (of the input image names). The simplest modification element is the asterisk "" which means call each of the output lists the same name as the corresponding input images. So:

IN >

signifies that all the images in the current directory should be used and the output lists should have the same names.

Other types of modification can also occur, such as:

OUTLIST > _objs.dat

which means call the position lists the same as the input images but put "_objs.dat" after the names. Replacement of a specified string with another in the output file names can also be used:

OUTLIST > |_debias|_images.dat|

this replaces the string "_debias" with "_images.dat" in any of the output names.

If wildcarded names for the input images are used then is it recommended that wildcards are also used for the position list names (the order of input names is not guaranteed).

The output files contain a integer index for each image feature followed by the X and Y centroid (formed using all the intensity information) and finally the mean intensity of pixels in the group. [.DAT]

If TRUE then it is not a fatal error to detect no objects on an image. In this case the output list of positions will not be written and the value in the COUNTS parameter will be set to 0. [FALSE]
An oversampling factor which is used when forming the initial histogram (greater than 1). The oversample is estimated by making the initial histogram mean count OVERSAMP times smaller than the mean count which would give BINFRAC in every bin. Increasing the oversample will increase the probability that only one bin will meet the BINFRAC criterion. Only used if USEPER is FALSE. [5]
The percentage point in the data histogram which is to be used as the threshold estimate. For data which has a significant background count this value should always be much greater than 50 (the median) and probably greater than the upper quantile (75). Only used if USEPER is TRUE. [96]
The threshold which is to be used for detecting image features. Connected pixel groups above this threshold form image features. This parameter is only used if the AUTOTHRESH parameter is set FALSE. In this case a value may be supplied for each image which is being processed. [Dynamic default]
If TRUE then pixel groups may contact the edges of the data array. Contact is defined as any pixel in the connected group of pixels being on the first or last column or row of the actual data array (not including any image origin information). Setting this FALSE decreases the probability of incomplete pixel groups being centroided which would result in inaccurate positions. [FALSE]
If TRUE then a percentage point (of the total counts) in the histogram will be used to estimate the threshold. Otherwise a gaussian fit to the data histogram will be used to estimate the background value. [TRUE]


findobj in= minpix=10 outlist=.find
In this example FINDOBJ processes all the images in the current directory locating objects with connected pixel groups which have more than 9 pixels above the threshold.
findobj "image1,image2,image10" 6 "obj1.dat,obj2.dat,obj3.dat" useper=false nsigma=3
In this example FINDOBJ estimates the threshold using the mode value in the histogram of data values as an estimate of the background and the fit of a gaussian to this to estimate the background standard deviation. The threshold used for each image is then 3 times the standard deviation above the estimated background.

See also

Automated registration”.


The algorithm used for calculating the values of percentiles for threshold determination should give good results even in the presence of pixel values which lie very far away from the bulk of the data. However, the sampling of the histogram used to estimate the mode and standard deviation may be poor in the presence of extreme outliers. If there are extreme outliers therefore, the percentile method (USEPER set to TRUE) of determining the threshold should be used.

The histogram used by FINDOBJ when USEPER is FALSE is formed by (if necessary) re-binning until the BINFRAC criterion is met, it is expected that this will always result in a well sampled histogram. The background value is the mode of this histogram and is not refined during the gaussian fitting. The gaussian fitting just estimates the standard deviation of the background and uses a fixed peak value and position (the mode of the histogram) and iterates rejecting bins whose counts fall below 20 percent of the peak value, stopping when either 3 iterations have been performed or the standard deviation does not change by more than one bin width in data values.

FINDOBJ is optimised to determine a reliable detection threshold and is not concerned with the accurate determination of the background value on a frame (as it performs no photometric measurements). For this reason the histogram which it uses to determine the background value is made in such a way that it is usually very well sampled (probably oversampled, for most other purposes). FINDOBJ should not be used in a manner for which it is not suited without understanding how if differs from other more specialized routines.

On exit the CURRENT_LIST items in the CCDPACK extensions (.MORE.CCDPACK) of the input NDFs are set to the names of the appropriate output lists. These items will be used by other CCDPACK position list processing routines to automatically access the lists.

CCDPACK format - Position lists in CCDPACK are formatted files whose first three columns are interpreted as the following.

The column one value must be an integer and is used to identify positions which may have different locations but are to be considered as the same point. Comments may be included in the file using the characters # and !. Columns may be separated by the use of commas or spaces.

In all cases the coordinates in position lists are pixel coordinates.

Behaviour of parameters

Most parameters retain their current value as default. The "current" value is the value assigned on the last run of the application. If the application has not been run then the "intrinsic" defaults, as shown in the parameter help, apply. The exceptions to this rule are:

Retaining parameter values has the advantage of allowing you to define the default behaviour of the application but does mean that additional care needs to be taken when re-using the application after a break of sometime. The intrinsic default behaviour of the application may be restored by using the RESET keyword on the command line.

Certain parameters (LOGTO and LOGFILE) have global values. These global values will always take precedence, except when an assignment is made on the command line. Global values may be set and reset using the CCDSETUP and CCDCLEAR commands.

Implementation Status: