码迷,mamicode.com
首页 > 移动开发 > 详细

[学习笔记] CS131 Computer Vision: Foundations and Applications:Lecture 4 像素和滤波器

时间:2017-12-09 22:36:07      阅读:370      评论:0      收藏:0      [点我收藏+]

标签:bsp   main   histogram   dom   color   transform   osi   replica   sign   

Background reading: Forsyth and Ponce, Computer Vision Chapter 7

Image sampling and quantization

Types of images: binary, gray scale, color

技术分享图片

Resolution: DPI: dots per inch, spatial pixel density

技术分享图片

Image histograms: histogram of an image provides the frequency of the brightness(intensity) value in the image

Image as functions: an image is a funciton  $f$ from $R^2$ to $R^M$ 

Linear systems: Forming a new image whose pixel values are transformed from original pixel values

Goal: extract useful information from images, or transform  images into another domain where we can modify/enhance image properties. 

  • Features(edges, corners, blobs) 
  • super-resolution, in-painting, de-nosing

技术分享图片

Moving Average, image segmentation, 

技术分享图片

技术分享图片

Convolution and correlation:

Edge effect: A computer will only convolve finite support signal,at the edge:

  • zero padding
  • edge value replication
  • mirror extension

技术分享图片

 

[学习笔记] CS131 Computer Vision: Foundations and Applications:Lecture 4 像素和滤波器

标签:bsp   main   histogram   dom   color   transform   osi   replica   sign   

原文地址:http://www.cnblogs.com/vincentcheng/p/8012615.html

(0)
(0)
   
举报
评论 一句话评论(0
登录后才能评论!
© 2014 mamicode.com 版权所有  联系我们:gaon5@hotmail.com
迷上了代码!