An efficient edge detection algorithm for 2D-3D conversion

Abstract

The 2D-3D conversion requires 2D content to convert into 3D display. This conversion process first estimates the 3D structure of the scene and then rendered the scene; finally it produces 3D images. In Existing system, the Hybrid depth generation algorithm has three depth cues for depth estimation: motion information, linear perspective, and texture characteristics. To find the edge detection they are using a sobel operator. We propose a canny edge detection algorithm instead of sobel operator to find the accurate edge detection; this edge detection algorithm is used to reduce the amount of data in the image. This approach used to detect the real edge points and non edge points. It should maximize the real edge points and minimize the non edge points. These similarities to maximize the signal to noise ratio. The detected edges as close as to the real edges. The real edge should not result as the detected edge. Using a canny edge detection algorithm the visual perception of the image can be improved.