I have an image of intensity values. On top of this image are objects that travel various trajectories. The trajectories are traced, and it is noticed that they have a higher average speed when they are closer to the higher image intensity values.
I would like to create some type of surface or manifold on top of this image that can approximate the effect of spatial location and the objects speed.
The kinds of data sets I am working with are: The Image of intensity values: a black and white picture, about 1500x1500.
And Object Speeds: a 200x1 vector of average speeds corresponding to 200 different objects trajectories that started at different locations on the image
I have tried things like plotting a heat map over the image using meshCanopy: http://www.mathworks.com/matlabcentral/fileexchange/29485-meshcanopy
which is nice, but because of the speed data I have, it just looks like I am tracing out an objects trajectory, and the height of that trace is based on the average speed. This is an okay approach, but I would like to think of a way to do an approximation of speeds of an objects trajectory based on its proximity to image intensities. So maybe I could plot some sort of color gradient, where the color intensity is associated with the speed of the object, and the colors become more intense as they get closer to my original image intensity values. Does this make sense? Essentially using 200 objects trajectories, and using that as a starting point to approximate speed of regions where I didn't actually measure an object.
I am not really sure what kinds of tools are available out there to do something similar, and this is perhaps not possible, but I wanted to see if maybe people have done similar things to this?
Thank you to whoever responds to this.
