What is wrong with my perlin noise generator? - c++

Question

I am trying to generate perlin noise for a math essay for school, and i have some difficulties figuring out the math behind it. This is my perlin class. The perlin noise function generates ( should generate) a number between 0 and 1, that i then multiply by 255 to apply color to every pixel on the screen, please help!
#include "perlinnoise.h"
perlinnoise::perlinnoise()
{
srand(time(NULL));
double random = rand() % 1000;
for (int i = 0; i < (651 * 2); i = i + 2)
{
random = (rand() % 1000);
vecGrad[i] = random / 1000;
vecGrad[i + 1] = vecGrad[i];
vecGrad[i] = cos(vecGrad[i] * 2 * 3.1416);
vecGrad[i + 1] = sin(vecGrad[i + 1] * 2 * 3.1416);
}
}
int perlinnoise::perlinNoise(int x, int y)
{
//20 pixel in each case
//30 boxes in width and 20 boxes in height
//651 vectors to create
sf::Vector2i boxXY;
boxXY.x = ((x / 20));
boxXY.y = ((y / 20));
sf::Vector2i displacement1; displacement1.x = x - boxXY.x * 20; displacement1.y = y - boxXY.y * 20;
sf::Vector2i displacement2; displacement2.x = x - (boxXY.x * 20 + 20); displacement2.y = y - boxXY.y * 20;
sf::Vector2i displacement3; displacement3.x = x - boxXY.x * 20; displacement3.y = y - (boxXY.y * 20 + 20);
sf::Vector2i displacement4; displacement4.x = x - (boxXY.x * 20 + 20); displacement4.y = y - (boxXY.y * 20 + 20);
/*std::cout << displacement1.x << std::endl; std::cout << displacement1.y << std::endl;
std::cout << displacement2.x << std::endl; std::cout << displacement2.y << std::endl;
std::cout << displacement3.x << std::endl; std::cout << displacement3.y << std::endl;
std::cout << displacement4.x << std::endl; std::cout << displacement4.y << std::endl;*/
double dotP1 = (vecGrad[((boxXY.y * 30) + boxXY.x)] * displacement1.x) + (vecGrad[(boxXY.y * 30) + boxXY.x + 1] * displacement1.y);
double dotP2 = (vecGrad[((boxXY.y * 30) + boxXY.x + 3)] * displacement2.x) + (vecGrad[(boxXY.y * 30) + boxXY.x + 4] * displacement2.y);
double dotP3 = (vecGrad[((boxXY.y * 30 + 1) + boxXY.x)] * displacement3.x) + (vecGrad[(boxXY.y * 30) + boxXY.x + 1] * displacement3.y);
double dotP4 = (vecGrad[((boxXY.y * 30 + 1) + boxXY.x + 3)] * displacement4.x) + (vecGrad[(boxXY.y * 30) + boxXY.x + 4] * displacement4.y);
This is where i have some troubles ( I think)
int intensity = 0;
double Sx = (3 * (x - boxXY.x * 20) * (x - boxXY.x * 20)) - (2 * (x - boxXY.x * 20) * (x - boxXY.x * 20) * (x - boxXY.x * 20));
double Sy = (3 * (y - boxXY.y * 20) * (y - boxXY.y * 20)) - (2 * (y - boxXY.y * 20) * (y - boxXY.y * 20) * (y - boxXY.y * 20));
double a = dotP1 + (Sx * (dotP2 - dotP1));
double b = dotP3 + (Sx * (dotP4 - dotP3));
double aa = dotP1 + (Sy * (dotP2 - dotP1));
double bb = dotP3 + (Sy * (dotP4 - dotP3));
intensity = (a+b+aa+bb)/4;
//Should generate number between 0 and 1, but doesn't :/
return intensity;
}
perlinnoise::~perlinnoise()
{
}
I've been reading lots of articles, and they are all very unclear about the math used.I ended up generating a grid with 20*20 pixels in each, with each cross section in the grid having a randomly generated gradient vector. I then calculate the displacement vectors and then do the dot product on the four corners with displacement and gradient vectors. This first part is a bit messy as i am not very experienced, but the last part is a bit more straightforward. I use a smoothing function on the x and y axis and use that number to generate a, b, aa and bb, and i then take the average of that. This is what i thought i understood from the articles i read, but apparently it's wrong :/ Any help please?
Thanks in advance!