hashFromImage: hashFromImage

In haimeh/finFindR: Fin Identification

Usage

curl -v http://localhost:8004/ocpu/library/finFindR/R/hashFromImage/json\
-F "imageobj=@C:/Users/jathompson/Documents/dolphinTestingdb/jensImgs/test2.jpg"

hashFromImage(imageobj = "yourfile1.jpg")

Arguments

imageobj

character vector which denots image file "directory/finImage.JPG"

Details

traceFromImage wrapper for use through opencpu. opencpu passes temp object name to function followed by traceToHash Processes an image(cimg) containing a fin. First the image undergoes cleanup through a variety of filters and glare removal via constrainSizeFinImage and fillGlare These processes help enhance edge clarity. The trailing edge is highlighted via neural network. The image is then cropped down to the trailing edge for efficiency purposes. The canny edges are then extracted from the crop and passed to traceFromCannyEdges which isolates coordinates for the trailing edge. These coordinates are then passed to extractAnnulus which collects image data used for identification. Both the coordinates and the image annulus are then returned.

Value

Value of type list containing: "hash" vector specifying an individual "coordinates" a matrix of coordinates

haimeh/finFindR documentation built on July 17, 2021, 12:56 a.m.