Hello,
We are working with the segmentation results provided by the function "fcvImageSegmentationRegionGrow" of the Qualcomm FastCV Library for GREY images and obtaining very poor results. Following the instructions from https://developer.qualcomm.com/docs/fastcv/api/group__image__processing.html#gafa2329cc821def360dc514feba9d8685 we have tested almost all the combinations for the thresholds of the split and merge stages. The results mostly were very poor and only a few had just some sense.
Is there any document where this function is explained in depth? Is there any problem working with grey images?
I would appreciate a toy example of this function with a well-known image of the literature.