An inefficient implementation of equals().
An inefficient implementation of equals(). Subclasses should override this if they can implement it more cheaply and anyone cares about such things.
Get the pixel value at (x, y, channelIdx).
Get the pixel value at (x, y, channelIdx). Channels are indexed as follows:
Returns an iterator of coordinate values based on the "natural" order of a Vectorized image.
Returns an iterator of coordinate values based on the "natural" order of a Vectorized image. That is, this returns a value of the form (x,y,channel,value) in order.
This method is optimized to avoid unnecessary memory allocation and designed
to approach the performance of an equivalent while loop over the image pixels for
speeding up things like Aggregation over an image regardless of underlying image ordering.
An important restriction is that the reference to the returned CoordinateValue
should not be modified or saved by the caller.
Put a pixel value at (x, y, channelIdx).
Put a pixel value at (x, y, channelIdx).
Returns a flat version of the image, represented as a single array.
Returns a flat version of the image, represented as a single array. It is indexed as follows: The pixel value for (x, y, channelIdx) is at channelIdx + x*numChannels + y*numChannels*xDim.
This implementation works for arbitrary image formats but it is inefficient.
Helper trait for implementing Images that wrap vectorized representations of images.