International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395 -0056
Volume: 04 Issue: 04 | Apr -2017
p-ISSN: 2395-0072
www.irjet.net
STEREO VISION S.Teja, Sk.Asif, K.Bhavani,B.Charan,B.L.N.Phaneendra *U.G,Student, #Assistant professor, Dept of IT, VR Siddhartha Engineering College ---------------------------------------------------------------------***--------------------------------------------------------------------Abstract - Stereoscopic vision delivers a sense of depth based on binocular information but additionally acts as a mechanism for achieving correspondence between patterns arriving at the left and right eyes. We analyse quantitatively the cortical architecture for stereoscopic vision in two areas of macaque visual cortex. Stereo Vision is an area of study in the field of Machine Vision that attempts to recreate the human vision system by using two or more 2D views of the same scene to derive 3D depth information about the scene. Depth information can be used to track moving objects in 3D space, gather distance information for scene features, or to construct a 3D spatial model of a scene. As an emerging technology, Stereo Vision algorithm are constantly being revised and developed, and as we will discuss in this project, many alternative approaches exist for implementation of a Stereo Vision system. This project introduces Stereo Vision as an area of current international research, presents some interesting applications of Stereo Vision, and discusses implementation of a Stereo Vision system. Key Words: Stereoscopic, Visual cortex, Stereo Vision, Correspondence, Depth information.
We can even perceive and measure "empty" space with our eyes and brains. 1.2 Literature Survey
1.INTRODUCTION
4
Calculating the distance of various points in the scene relative to the position of the camera is one of the important tasks for a computer vision system. A common method for extracting such depth information from intensity images is to acquire a pair of images using two cameras displaced from each other by a known distance. As an alternative, two or more images taken from a moving camera can also be used to compute depth information. In contrast to intensity images, images in which the value at each pixel is a function of the distance of the corresponding point in the scene from the sensor are called range images or density stereo map. Such images are acquired directly using range imaging systems. Two of the most commonly used principles for obtaining such range images are radar and triangulation. In addition to these methods in which the depth information is computed directly, 3-D information can also be estimated indirectly from 2-D intensity images using image cues such as shading and texture. These methods are described briefly in this chapter. 1.1 Objectives With stereo vision, we can see WHERE objects are in relation to our own bodies with much greater precision. We can see a little bit around solid objects without moving our heads. © 2017, IRJET
|
Impact Factor value: 5.181
|
Numerous methods of implementation for stereo vision disparity mapping have been established in the past few years. This can be observed from the table listed in below. S.no
Method
Year
Author
Technique
1
Sum of squared differences (SSD) Normalised cross correlation (NCC) Rank Transform (RT) Census Transform (CT)
1994
Fusiello et al.
Sum of squared differences
1997
Satoh
Correspondence
2002
Gac al.
2004
Humen berger et al.
2
3
et
Rank Difference Hamming Distance
Chart-1: Existing Methods Of Stereo Vision 1.2 Algorithm 1.3.1 Semi-Global Matching(SGM) Semi-Global Matching (Hirschmüller, 2005 and 2008) successfully combines concepts of global and local stereo methods for accurate, pixel-wise matching at low runtime. The core algorithm considers pairs of images with known intrinsic and extrinsic orientation. The method has been implemented for rectified and unrectified images. In the latter case, epipolar lines are efficiently computed and followed explicitly while matching (Hirschmüller et al., 2005). 1.3.2 Advantages of SGM algorithm Semi-global matching allows high-density point clouds of the first reflective surface to be rapidly and automatically extracted from stereo-imagery. One advantage of SGM is that these point clouds look and feel much like lidar point clouds. SGM models can be colorized from the available imagery
ISO 9001:2008 Certified Journal
|
Page 636