Investigating Role of SIFT and SURF in Panorama Generation

Image stitching (Mosaicing) is considered an active area of computer vision and computer graphics research. Image stitching is about combining two or more images of the same scene into a panoramic image called a high- resolution one. Techniques for image stitching can be categorized into two general

Project Title

Investigating Role of SIFT and SURF in Panorama Generation

Project Area of Specialization

Augmented and Virtual Reality

Project Summary

Image stitching (Mosaicing) is considered an active area of computer vision and computer graphics research. Image stitching is about combining two or more images of the same scene into a panoramic image called a high- resolution one. Techniques for image stitching can be categorized into two general approaches: called direct and functional. Direct techniques compare all the pixel intensities of the images with each other, while functional techniques aim to determine the relationship between the images by means of separate features extracted from the images processed. The last approach has the advantage of being faster against scene movement and has the ability to discover the overlapping relationships between an unordered set of images automatically. The aim of this project is to present a survey on the image stitching based on the feature. The main picture stitching components are described. A framework for a complete system of image stitching based on functional approaches is introduced. 

Project Objectives

Objective of the propose system is to investigate role of different feature descriptors such as SIFT and SURF for panorama generation and then generate a 360 degree panorama video which is use for survillance system.

Project Implementation Method

First step of the propose system is Image Acquisition . In any visual system the most important phase is images acquisition. There is different source available for the retrieval and acquisition of images.When we keep the optical center at fixed position or a simple camera by using these methods we can move a camera in parallel to the scene and rotate a camera over its Y-axis. for this pupose we need a high resolution camera.

Second step is eatures extraction the features extraction is a  method which extract and match different things from images such as points of an image, lines in an image, corners of an image and edges or geometrical pixels. Strong characteristics detector contains invariant pictures having noise, invariant measures, invariant translation and rotation transformations. The image stitching is achieved using the image features obtained from the corresponding images taken from different viewpoints.we extract SIFT and SURF features for our propose system.

Third step is Homography Using RANSAC Once we have the feature information for all images, then we match images using that feature information. In image matching, first we find the neighbors of all images and then we find a set of correct matching after that match all features. RANSAC algorithm is use for computing homography. After that stitch images with each each other to make a panorama image or output image.

Benefits of the Project

There is alot of Benifits of this project but the most importent is that , that it is use in survillance system. Digital video surveillance has become commonly used in public and private places such as government buildings, military bases, car parks, and banks, and so forth. Traditional monitoring cameras can only cover a limited area, leading to “blind spots.” Developments in panoramic imaging technology offer significant advantages over traditional surveillance systems. They can monitor an area that covers 180° or 360° and so replace several traditional cameras.

Some other benifits are  business inspection, computer graphics, ?lm, entertainment, mobile devices and security, etc.

Technical Details of Final Deliverable

Python

OpenCV

C#

2 High resolution cameras

Final Deliverable of the Project

Software System

Type of Industry

IT

Technologies

Augmented & Virtual Reality

Sustainable Development Goals

Quality Education

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com