Static Object Depth Estimation Using Defocus Blur Levels Features

Depth estimation of an object in a scene from images intensity is an essential problem in machine vision applications. Here, In order to obtain the distance between object and camera, a novel method presented for depth estimation from a single defocus blurred image. Unlike other methods for depth estimation which are based on the characteristics of the three-dimensional information, here, only constant camera parameters are considered. Thus, the change of distance of the object from the camera has direct relation with amount of object´s blurring in image. In fact, the amount of blurring will be published with a direct comparison in front of the camera and can be seen with the gray level changes around the edges of objects. The proposed method is considered to be Blind method in which image in a focus state is not required. Using only one frame, the relative uniform depth map can be estimated.