An image processing method and apparatus for determining depth in an original image captured by a light field image capture device, in which a light field analysis algorithm is applied a plurality of times to the original image, changing the focus setting each time, so as to generate a respective plurality of scene images focused at different depths; edge detection is performed in respect of each of the scene images to generate a respective plurality of edge detected images; area identification is performed in respect of each edge detected image to generate a respective plurality of area identification images indicative of areas of respective edge detected images in which edges have been detected; and the area identification images are applied to respective scene images so as to extract from the scene images respective image segments corresponding to the areas in which edges have been detected.