let cam;

let prevCam;

let rowArray;

let noiseValue = 2;

let song;

let isHitting = false;

let hittingTime = 0;

let imageModelURL = 'https://teachablemachine.withgoogle.com/models/9Zbro_T2N/';

let classifier;

let label = "";

let confidence = "";

​

function preload(){

   classifier = ml5.imageClassifier(imageModelURL + 'model.json', modelReady);

 song = loadSound('whitenoise.mp3');

  song.loop();

}

function setup() {

  song.setVolume(0.0);

  song.play();

  rowArray = [];

  createCanvas(640, 480);

​

  cam = createCapture(VIDEO);

  //cam.size(640, 480);

  cam.hide();

  for (let i = 0; i < height; i++)

    {

      rowArray[i] = random(255);

      //console.log(rowArray[i]);

    }

  img = createImage(width, height);

  classifyVideo();

}

​

function draw() {

  background(0);

​

  cam.loadPixels();

  img.loadPixels();

  if (label == "flap") {

    // do something 

   isHitting = true;

    //if (hittingTime)

      hittingTime += 1;

    //console.log(hittingTime);

  }

  else

    {

      //console.log(hittingTime);

      isHitting = false;

      hittingTime = 0;

    }

  if (hittingTime > 50)

    {

      //console.log(hittingTime);

      noiseValue = 2;

    }

  //console.log(deltaTime);

  if (noiseValue > 0)

    {

       song.setVolume(1 - (noiseValue/2));

    noiseValue -= deltaTime/20000;

    }

  //console.log(cam.pixels.length); // w * h * 4(RGBA)

let tempRowArray = [];

  tempRowArray[0] = rowArray[height-1];

 for (let i = 1; i < height; i++)

    {

      tempRowArray[i] = rowArray[i-1];

      //console.log(rowArray[i]);

    }

  rowArray = tempRowArray;

  for (let y = 0; y < img.height; y++) {

    for (let x = 0; x < img.width; x++) {

​

      let index = (x + y * img.width) * 4;

      let r = 0;

      let g = 0;

       let b = 0;

      if (random(2) < noiseValue)

        {

          let offset = floor(random((2 - noiseValue)* 20)) * 4;

         r =  cam.pixels[index + offset + 0];

      g = cam.pixels[index + 1];

       b = cam.pixels[index + offset + 2];

        }

      else{

        let tr = (random() < 0.5)? rowArray[y]:255;

         r = tr; //cam.pixels[index + 0];

      g = tr;//cam.pixels[index + 1];

      b = tr;//cam.pixels[index + 2];

      }

​

      // original color from the cam

      img.pixels[index + 0] = r; // R

      img.pixels[index + 1] = g; // G

      img.pixels[index + 2] = b; // B

      img.pixels[index + 3] = 255; // A

    }

  }

​

  img.updatePixels();

  image(img, 0, 0);

​

  // store the current cam image (one frame)

}

​

function classifyVideo() {

  classifier.classify(cam, gotResult);

}

​

function gotResult(error, results) {

  // If there is an error

  if (error) {

    console.error(error);

    return;

  }

  // The results are in an array ordered by confidence.

  // console.log(results[0]);

  label = results[0].label;

  confidence = results[0].confidence;

  //console.log(results);

  // Classifiy again!

  classifyVideo();

}

​

function modelReady() {

  console.log("Model Loaded!");

}