Depends On: Initial Setup; Mosaic Settings; Med Filter
PURPOSE
Detect performs image segmentation (see §8.3). In the Mosaic pipeline, it is one of four modules that, together, carry out Dual Outlier detection (see §8.2.3). The other three modules are Mosaic Projection, Mosaic Dual Outlier and Level. Detect produces bright object detection maps.
INPUT
Detection Max Area: (int) The maximum area of a single outlier identified in the detection map. If the outlier is larger than this threshold, it will be considered an extended object.
Detection Min Area: (int)The minimum area of a pixel cluster to be detected.
Detection Threshold: (float) The number of sigma above the mean for pixels included in clusters. Default value is 3.
Threshold Type: There are three different ways to decide if a cluster of bright pixels will be segmented into several clusters. The choices are simple, combo and peak (see §8.3 Image Segmentation).
Detect output subdirectory: The subdirectory of <output_dir> that you wish to use for the output files. Default is Detect-mosaic.
COMMAND LINE INPUT
&DETECT
Detection_Max_Area = 1000,
Detection_Min_Area = 0,
Detection_Threshold = 3,
Threshold_Type = 'simple',
&END
In Global Parameters:
DETECT_DIR = Detect-mosaic
OUTPUT
Generated Fits Files (*detmap.fits): The stack of detection maps produced from interpolated BCD images. Sources with positive signals, including both real science sources and outliers from cosmic rays and bad pixels, are set to the number of the cluster membership. The background is zero in the segmentation map.
DISCUSSION
In the Mosaicpipeline, Detect is part of Dual Outlier detection (see §8.2.3). The detection of bright pixels is performed on the background-subtracted images (the output of the Med Filter module), which are in original pixels. The output is a set of mask images, one per input image, in which pixels corresponding to detected objects are set to a positive value. The mask images are saved in the Detect output subdirectory unless the Delete Intermediate Files option was set in the Mosaic Settings module. Following this module, Mosaic Projection takes the output maps and projects them to a common reference frame (the FIF) before the later modules identify true outliers.
