医学图像处理笔记(4)

Point Operation

Application: contrast enhancement; contrast stretching; gray scale transformation

pixel-by-pixel copying operation

Linear point operations

\(D_{A}=f(D_{A})=aD_{A}+b\)

a>1:contrast increased; 0<a<1:contrast reduced(灰度值差异拉伸，缩小)

a=1 & b!=0:shift gray level;(灰度平移)

a<0:reverse the contrast

Nonlinear point operations

\(f(x)=x+C_{x}(D_{m}-x)\) Dm: maximum gray level ,Cx: determine the amount of increase

Application: Photometric calibration ;Contrast enhancement ; thresholding; Contour lines; clipping

Algebraic operation:

\(C_{x, y}=A_{x, y} + B_{x, y}\)

\(C_{x, y}=A_{x, y} - B_{x, y}\)

\(C_{x, y}=A_{x, y} * B_{x, y}\)

\(C_{x, y}=A_{x, y} / B_{x, y}\)

Application

Adding:averaging for noise reduction; double-exposure effect;

Subtraction: background subtraction; motion detection; gradient magnitude

Geometric Operation:

change the spatial relationships among the objects in an image

1. spatial transformation

2. gray level interpolation

Transformation

平移(translation)、缩放（scale change）、旋转（rotation）

Interpolation Algorithm

Nearest Neighbor Interpolation

Bilinear interpolation(square interpolation)

B-spline

polynomials

Neighborhood Operation

4- neighbor \(N_{4}(P)\)

8- neighbor \(N_{8}(P)\)

Connected Component Labeling

intensity criterion (IC)

Labeling Algorithm

Image Feature

Region properties:

1: perimeter and area

perimeter

estimated by the number of pixels on the contour

calculated from the chain-code (计算边缘长度) of the contour

Area

the number of unit squares contained (Pick's formula): \(A(P)= n_{l}+ n_{B}/2 - 1\) )

2: center, radius and Diameter

center

eccentricity of a point P in F is the maximum of distance d(p,q) for all points, ecc(p) = max d(p,q)

center : the set of points P of least eccentricity

Radius: the value of the least eccentricity

Diameter: the value of the greatest eccentricity

3: centroid Moments and orientation

4: extreme points and curvature

5: intensity properties

Bonding box(区域用矩形圈起来)