Detailed explanation of common methods of filtering matrix (array) elements in Matlab

I. Introduction

When using Matlab matrices or arrays, it is sometimes necessary to filter some elements, that is, to replace elements that meet certain conditions with other data.In this paper, several relatively simple methods are given for the filtering and application of matrix (array) elements.

Second, matrix (array) element filtering method

1. Use find
Example 1: Get elements in the matrix that meet certain conditions to form a new array.

a = [ -1, 2; 3, -1 ]b = a( find(a > 0) )

There is an output:
a =
-1 2
3 -1
b =
3
2

At this point b is a column vector.Since the matlab storage matrix is ​​stored in the column direction, 3 is in the front and 2 is in the back.
2. Use logical expressions
Example 2: Set the elements greater than zero in the matrix to infinity, and other elements are changed according to a certain algorithm.

a = [ -1, 2; 3, -1 ]b( a > 0 ) = Inf;b( a <= 0 ) = a( a <= 0 ).^2 + 1;b % Now b is a column vector

Output result:
a =
-1 2
3 -1
b =
2
Inf
Inf
2
If you want bIf it has the same dimension as a, there are two methods as follows:
1) Method 1: First generate an all-zero matrix with the same dimension as a

a = [ -1, 1; 1, -1; -1, 1 ];b= zeros( size(a) );b( a > 0 ) = Inf;b( a <= 0 ) = a( a <= 0 ).^2 + 1;b % At this time, b is an array of the same dimension as a

Output result:
b =
2 Inf
Inf 2
2 Inf
2) Method 2: Use reshape to change the column vector to a matrix

a = [ -1, 1; 1, -1; -1, 1 ];b( a > 0 ) = Inf;b( a <= 0 ) = a( a <= 0 ).^2 + 1;b = reshape( b, size(a) )

Output result:
b =
2 Inf
Inf 2
2 Inf
3, using logical AND
Example 3: If you change a row or column of the matrix to another value, the operation is very simple. For example, setting the first column of matrix B to 0, you can do the following:

B = [ 1, -1; 2, -2; -3, 3 ]C = B;C( :, 1 ) = 0

Output result:
B =
1 -1
2 -2
-3 3
C =
0 -1
0 -2
03
Example 4: If you want to set some elements at specified positions to 0, you can perform a bitwise AND operation similar to the C language.First specify a matrix with a non-zero target position, then do a logical AND with the given matrix, and then dot multiply the given matrix

B = [ 1, -1; 2, -2; -3, 3 ];E = [ 1,1; 0,0; 1,0 ] % logical matrixD = (B & E) .* B

Output result:
E =
1 1
0 0
1 0
D =
1 -1
0 0
-3 0
p>

Three, simple application of matrix (array) element filtering

1. Piecewise function evaluation
Example 5: Simple piecewise function

Reference code:

t = [ -1 : 0.1 : 1 ] * pi;y( t <= 0 ) = -t( t <= 0 );y( t > 0 ) = sin( t( t > 0 ) ).*2 + t( t > 0 );plot( t, y )

Run result:

Example 6: A bit more complicated piecewise function

Reference code:

x = [ 1 : 20 ];y( x >= 1 & x <= 5 ) = 6 - x(x >= 1 & x <= 5 );y( x >= 6 & x <= 15 ) = 1;y( x >= 16 & x <= 20 ) = x( x >= 16 & x <= 20 ) - 15;xlen = length( x )ylen = length( y )plot( x, y, 'b' );axis( [ 0, 21, 0, 6 ] );

Run result:

2. Use a known array to filter other arrays
Example 7:

a = [ -1, 1; 1, -1 ]b = pascal( 2 )b( a > 0 ) = 5b( a+3 < b ) = 3b( b < 2 ) = b( b < 2 ) + 2

Running result:
a =
-1 1
1 -1
b =
1 1
1 2
b =
1 5
5 2
b =
1 3
3 2
b =
3 3
3 2