Skip to content

hyperCSI algorithm to signature hyperspectral image by image with lots of missing pixels

Notifications You must be signed in to change notification settings

HsuChiChen/ncku-image-processing

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Image Processing

Time : 2020 fall (first half semester of sophomore)

lecture

subject teacher
影像處理 林家祥

Report


Environment

  1. OS
  • Windows 10 21H1
  1. Software
  • Matlab R2021a

How to Run

  1. download repo
git clone https://github.com/HsuChiChen/image-processing.git
  1. open matlab
matlab
  1. run the main.m in midterm or final file
    press run gui icon in matlab

Midterm

Rrequirement

input output
hyperspectral imgae signature of the 3 material

More info is in doc/midterm_rule.pdf

Algorithm

由於Github不支援latex語法建議於doc/midterm_demo.pptx觀看以獲得更好的理解。

  1. 主成分分析(PCA)降維
    • 資料平移到0
    • 解共變異數矩陣的特徵值(λ)和特徵向量(𝑣)
    • 取變異量(特徵值)最大與次之的兩主成份

  1. SPA找單形體頂點
頂點 尋找方法
頂點1 原點最遠的點
頂點2 算(頂點一到原點)的垂直投影矩陣求所有投影點離原點最遠的點
頂點3 在投影面上離投影後頂點二距離最遠的投影點

  1. 修正頂點
    • 取𝑎­2−𝑎­3的法向量(二維時(a,b)→(−b,a))
    • 取頂點二在圓範圍內與(1)做內積,求最大值之頂點p_1
    • 取頂點三在圓範圍內與(1)做內積,求最大值之頂點p_2
    • p_1,p_2之法向量b
    • p_2上在b的投影即為位移量h
    • 有法向量b與位移量h即可描述該直線,重複(1)-(6)求得其他兩直線
    • 兩直線解聯立方程式得實際頂點P

  1. 做PCA反運算得物質指紋𝑃=𝐶𝑎_𝑖+𝑑

  2. 分布圖
    利用高中觀念,求任意點a點到直線的距離/頂點P到直線的距離,即為該點所占成分比例

  • 演算法驗證
    發現助教給的數據很完美,修正前後點是一樣的。
X 頂點一 頂點二 頂點三
X點(SPA) 4.509814 0.16576 -3.65346
Y點(SPA) -0.23925 0.753292 -0.40151
X點(修正) 4.509814 0.16576 -3.65346
Y點(修正) -0.23925 0.753292 -0.40151


Final

Rrequirement

input output
hyperspectral imgae with some missing data signature of the 3 material

with error of 9.1214e-04

More info is in doc/final_rule.pdf

Algorithm

由於Github不支援latex語法建議於doc/final_demo.pptx觀看以獲得更好的理解。

  • Missing Data選取

before after
  1. 求A
input output
Y_omega A

Y_omega做HyperCSI(期中考內容)得指紋A

  1. 求S
input output
Y_omega2 Y_re

Y_omega2做PCA降維N=224→2

input output
A_omega h, b, alpha

求A_omega做PCA降維後 的點alpha以及代表Hyperplane的點h、法向量b

input output
Y_re, h, b, alpha S

點到線之距離公式

  1. 求Y
input output
A, S Y

About

hyperCSI algorithm to signature hyperspectral image by image with lots of missing pixels

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages