program留学生代做、Python编程设计调试、data代写、Python语言代做

日期：2020-05-23 11:23

Computer Vision Coursework

1. Topic

Image stitching: In this project, you are required to write an image stitching program that can

successfully stitch at least three images into one panorama image.

You need to capture your own images (at least two sets) for testing your program. The trick to

capture a good set of images is to only rotate your body while taking images without moving

yourself away.

Python is chosen as the language for development and PyCharm is used as the IDE. You can use

any libraries that can help you to achieve your tasks such as OpenCV, as long as it will not directly

give you the output of the project you are working on. (不能直接用 OpenCV 里的图像拼接函数，要

自己实现图像拼接的功能) You cannot directly use the code you found online or from the lab sample

code.

2. Program Design Outline

这里的步骤只是一个参考，可修改

a) Collect input images

b) Exact the features of images

The feature exaction will use SIFT algorithms to implement.

c) Implement the image patching

SIFT algorithm obtains the feature points in the image and the corresponding feature

description. Then, the KNN algorithm will be used to obtain the K data points closest to each

other in the space and classify these data points into a class.

d) Calculate the Homography

With the above steps, many matching points are got. However, it still needs Homography to

delete the unwanted corners that do not belong to the overlap area. In this step, RANSAC

algorithm will be used for robust parameter estimation by fitting mathematical models from

observed data sets that may contain outliers.

e) Implement the image wrapping and blending

f) Repeat the steps b) to e)

The steps b) to e) will be repeated for all input images.

以上是及格的要求，这个作业是想要高分的；高分要自己想 ideas，实现更多其他的特性，

可能的特性如下：

? Stitch more than three images into one panorama image.

? Various algorithms can be tried to get the best possible results. And critically evaluates your

method on the basis of those results; what are its strengths and weaknesses?

? Not all images of the same set are input sequentially, different input order can be tried.

? It is supposed that the given images have clear gap or lacking part with each other.