Computer Vision libraries for trailer truck testbed using open source computer vision libraries

Computer Vision is a field that aims at understanding and analyzing images from the real world to produce numerical and symbolical data. It is a first step at duplicating the capabilities of human vision by electronically understanding the image and perceiving its features. This work aims at providing some of the features of a human eye to a trailer truck. These features include getting a 3D wireframe from continuous images and prediction of the next position of the objects in view, while the truck is moving.;The thesis has been divided into 3 sections. First section is acquiring images in real time. Second section is the preprocessing of images to achieve edges and points in the image, and the third section is converting the edges and the points into 3D wireframe and predicting of the next position of the image. OpenCV library and Point cloud libraries are used in this process to facilitate operations on 2D and 3D images.