I would think that it is not possible to do so unless i have the actual size of the object, but I am asking about finding the distance to a certain object (i.e. car) that is detected but without knowing the size of that object, is there a certain technique that is applied in this case that makes it possible?
You are quite right. With one camera you need real parameters of the object so as to restore the distance. If you have no them, you need 2 cameras or move you camera around with constant speed.
I think these papers may be of help to you:
Rahman, Ashfaqur, et al. "An image based approach to compute object distance." International Journal of Computational Intelligence Systems 1.4 (2008): 304-312.
and
section 11 of
Sharma, Sachin Umesh, and Dharmesh J. Shah. "A Practical Animal Detection and Collision Avoidance System Using Computer Vision Technique." IEEE Access 5 (2017): 347-358
The latter paper has done this:
Image resolution is 640 × 480
X range is 0 to 640
Y range is 0 to 480
Let the right bottom coordinate of the detected cow be (x, y). Then the distance of cow from the lower edge (car/camera) is 480 – y.
They have then used a formula derived from the first paper to convert distance in pixels to meters
Asked: 2018-03-11 20:09:42 -0600
Seen: 4,432 times
Last updated: Mar 12 '18
