It appears that the “sensitivity” variable controls how “close” the color of pixels appear to the target sample. There appears to be some minimum number of the detections grouped before the SIG is identified. Is there a way to lower this threshold. From further back on the field, the flags are so small they frequently do not get detected.