如何在 Android 中对矩阵进行求逆和乘法？

Question

Given 2 matrices:

public float[] mRi = new float[16];  
public float[] mR = new float[16];  

These will be the outputs of two readings from

• SensorManager.getRotationMatrix(mR, x, y, z) and
• SensorManager.getRotationMatrix(mRi, x, y, z)

So there will be two 4x4 matrices which,

I want to get the result of the following equation:

• ResultMtrix=inverse(mRi)*mR

In fact I have the idea whether it works with invertM() and multiplyMM() but I don't have a clue on how to do it with the matrices.

Can you help?

Answer 1

Hey guy, I assume they are 4x4 matrix (16 elements)... You can use Gauss-Jordan elimination http://en.wikipedia.org/wiki/Gauss%E2%80%93Jordan_elimination for inversion. Matrix multiplication is described about everywhere , even more than matrix inversion.

Answer 2

The matrices you are describing are actually uni-dimensional vectors , so I am assuming that what you call inverse is actually transpose . The calculation in that case is quite simple:

Methods

// 1 row * 1 column
public static float scalarMultiplication (float[] m1, float[] m2) {
    if (m1.length != m2.length)
        throw new IllegalArgumentException("Vectors need to have the same length");
    float m = 0;
    for (int i=0; i<m1.length; i++)
        m += (m1[i]*m2[i]);
    return m;
}

// N rows * N columns
public static float[][] vectorMultiplication (float[] m1, float[] m2) {
    if (m1.length != m2.length)
        throw new IllegalArgumentException("Vectors need to have the same length");
    float[][] m = new float[m1.length][m1.length];
    for (int i=0; i<m1.length; i++)
        for (int j=0; j<m1.length; j++)
            m[i][j] = (m1[i]*m2[j]);
    return m;
}

Test

            float[] m1 = new float[16];
            float[] m2 = new float[16];

            for (int i=0; i<m1.length; i++) {
                m1[i]=i;
                m2[i]=i*i;
            }

            System.out.println ("Multiple is " + scalarMultiplication(m1, m2));
            float[][] m = vectorMultiplication(m1, m2);
            for (int i=0; i<m[0].length; i++) {
                for (int j=0; j<m[0].length; j++) {
                    System.out.print (m[i][j] +" ");
                }
                System.out.println();
            }

Output

Multiple is 14400.0
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 
0.0 1.0 4.0 9.0 16.0 25.0 36.0 49.0 64.0 81.0 100.0 121.0 144.0 169.0 196.0 225.0 
0.0 2.0 8.0 18.0 32.0 50.0 72.0 98.0 128.0 162.0 200.0 242.0 288.0 338.0 392.0 450.0 
0.0 3.0 12.0 27.0 48.0 75.0 108.0 147.0 192.0 243.0 300.0 363.0 432.0 507.0 588.0 675.0 
0.0 4.0 16.0 36.0 64.0 100.0 144.0 196.0 256.0 324.0 400.0 484.0 576.0 676.0 784.0 900.0 
0.0 5.0 20.0 45.0 80.0 125.0 180.0 245.0 320.0 405.0 500.0 605.0 720.0 845.0 980.0 1125.0 
0.0 6.0 24.0 54.0 96.0 150.0 216.0 294.0 384.0 486.0 600.0 726.0 864.0 1014.0 1176.0 1350.0 
0.0 7.0 28.0 63.0 112.0 175.0 252.0 343.0 448.0 567.0 700.0 847.0 1008.0 1183.0 1372.0 1575.0 
0.0 8.0 32.0 72.0 128.0 200.0 288.0 392.0 512.0 648.0 800.0 968.0 1152.0 1352.0 1568.0 1800.0 
0.0 9.0 36.0 81.0 144.0 225.0 324.0 441.0 576.0 729.0 900.0 1089.0 1296.0 1521.0 1764.0 2025.0 
0.0 10.0 40.0 90.0 160.0 250.0 360.0 490.0 640.0 810.0 1000.0 1210.0 1440.0 1690.0 1960.0 2250.0 
0.0 11.0 44.0 99.0 176.0 275.0 396.0 539.0 704.0 891.0 1100.0 1331.0 1584.0 1859.0 2156.0 2475.0 
0.0 12.0 48.0 108.0 192.0 300.0 432.0 588.0 768.0 972.0 1200.0 1452.0 1728.0 2028.0 2352.0 2700.0 
0.0 13.0 52.0 117.0 208.0 325.0 468.0 637.0 832.0 1053.0 1300.0 1573.0 1872.0 2197.0 2548.0 2925.0 
0.0 14.0 56.0 126.0 224.0 350.0 504.0 686.0 896.0 1134.0 1400.0 1694.0 2016.0 2366.0 2744.0 3150.0 
0.0 15.0 60.0 135.0 240.0 375.0 540.0 735.0 960.0 1215.0 1500.0 1815.0 2160.0 2535.0 2940.0 3375.0