Image Orientation Detection
Arthur Ardeshir Goshtasby · 2017
The orientation of an image depends on the geometric layout of the pattern within the image. An image's (global) orientation is calculated from the aggregate of its local orientations, and a local orientation is determined from intensity gradients or geometric gradients in a small neighborhood. To reliably determine the orientation of an image, two problems must be solved. First, the image should be converted into a form that is independent of absolute intensities and so insensitive to changes in scene lighting. Second, the digital domain of an image should be taken to a continuous one to remove/reduce geometric noise. To achieve this, this chapter introduces the concepts of geometric gradient and geometric smoothing. To compare the sensitivities of geometric gradients and intensity gradients to variations in intensities and geometry of an image, the intensities and geometry of an image are varied and the stabilities of image orientations determined by the two methods are compared.