I am trying to find actual length of an object, say X, from an image. For that I'm using a reference object , say Y, ( object of known dimensions ). From image while converting pixel length of X to physical length, using reference object dimension , it is consistently resulting in error b/w 1.2 inch to 1.6 inch.
I am using one frame of 60 inch * 48 inch as a reference object and actual object is placed inside the frame. My formula is actual length of X = pixel length of X * actual length of Y / pixel length of Y.I am using overhead camera of 13 MP. what can be the reason of error ? and how to reduce this within 1 inch ?