Centroider

class sonic.Centroider

Method Summary

static COB(imgObj, thresh, options)

Estimate centers of brightness within an image by removing background noise, and using the center of brightness equation

Inputs:

• imgObj (1x1 sonic.Image): Sonic image to detect centroids in

• thresh (1x1 double): threshold value for noise usually determined by running img.estNoiseRand or img.estNoiseSort

• minStarSize (1x1 double): Optional cutoff for minimum cluster size. Good for removing hot pixels.

• maxStarSize (1x1 double): Optional cutoff for maximum cluster size.

• nonMaxSupDist (1x1 double): Optional minimum distance between cluster centroids. Will always save centroids with largest DN within this area. Used for non maximum suppression.

• buffer (1x1 double): Optional no centroids found to be within this many pixels from the edge will be considered. Often used when a median filter has been applied.

Outputs:

• centroids (1x1 sonic.Points2 or empty 0x0 sonic.Points2): Points2 object containing the centroids found in the image. If no centroids are detected, an empty Points2 will be output.

• comps (nx1 cell array): Contains the linear indices of the connected components corresponding to the centroids

Last revised: 3/7/24 Last author: Ava Thrasher