Image gradient sampling grid

2016 1 7
1 个回答
更新时间：2016 1 8
3 次查看（30 天）

0 个投票

Hi, everyone. I am having a difficulty with my project in MATLAB. I really do not understand this concept of sampling grid and from the circle to R. Thank you for any help in advance! Here is the question:
Id is a discrete grayscale image of size N^2, thus an element of R N^N. This discrete image Id can be viewed as sampling on regular grid of a function I from the square [-1,1]^2 to R, with the grid spacing being h=2/(N-1) in both directions. How to compute I from Id?

请先登录，再回答此问题。

回答（1 个）

Image Analyst
Image Analyst 2016-1-7

0 个投票

You never said what the badly-named "I" is, so I don't know how to compute it from Id.
All the rest of the question is just a bunch of mathematical mumbo jumbo that really doesn't amount to anything except to allow the author to impress himself. Trust me, you can safely ignore it.

2 个评论
显示 无 隐藏 无

Hi, thank you for such quick response. I don't understand the theory behind this mumbo jumbo though. Clarifying the I and Id, In fact Id is the discrete image, the matrix, with indices i,j between 1 and N, and I is the equivalent image expressed at coordinates x,y. Basically, from what I understood, they want me to make an Image I the borders of which will be (-1, 1) and which will represent the values from the Image Id (which is a grayscale image), but only converting the matrix numbers of Id image (which has values b/w 0-255) to a matrix of I image (which will have values b/w -1,1 and will be bordered in dimension as a square -1, 1). I don't know if this is my correct understanding .
Image Analyst
Image Analyst 2016-1-8
Not sure what that means. A digital image is an array with rows and columns. A real image is just a bunch of light (or whatever). Not sure if you want to interpolate a different number of pixels, or if you want some kind of spatial calibration system that keeps track of the real-world (x,y) given the (column, row) indexes. For example element (1,1) corresponds to x=-1 and y=-1, and element (480, 1) (last row of a 480x640 image) corresponds to x=-1, y=+1, and element (480,640) corresponds to x=1, y=1. If so, you'll have to keep track of the spatial calibration factor(s) separately whenever you want to convert from real world coordinates to indexes.
See my spatial calibration demo, attached.

请先登录，再进行评论。

类别

在 帮助中心 和 File Exchange 中查找有关 Neighborhood and Block Processing 的更多信息

产品

标签

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by