如何在Processing中获取摄像头画面指定区域的平均颜色？

Hey there! No worries about messy code—we’ve all fumbled through our first Processing projects, so let’s break down why you’re only getting black for that rectangle’s average color.

Common Culprits & Fixes #

1. You’re Not Updating the Camera’s Pixel Array

This is the most frequent issue for beginners! Processing doesn’t automatically sync the camera feed’s pixels to the pixels[] array—you have to tell it to load them every time a new frame arrives.

• Fix: Right after reading a new camera frame with cam.read(), call cam.loadPixels(). This ensures you’re working with the latest pixel data, not an empty (black-by-default) array.

2. Your Pixel Index Calculation Is Wrong

Processing stores pixels in a 1D array where each row is stacked end-to-end. If you’re using the wrong formula to get the pixel at (x,y), you’ll end up grabbing pixels outside the valid area (which default to black).

• Correct Index: For a camera feed with width cam.width, the index of pixel (x,y) is y * cam.width + x.

3. No Boundary Checks for the Rectangle

If your rectangle extends outside the camera’s resolution (e.g., negative coordinates, or width/height that go beyond cam.width/cam.height), you’ll be sampling invalid pixels that return black.

• Fix: Add a quick check inside your loop to make sure you only sample pixels within the camera’s bounds.

Example Working Code #

Here’s a cleaned-up version that addresses all these issues, with comments to explain each step:

Capture cam;
// Rectangle properties
int rectX = 10;
int rectY = 10;
int rectW = 50;
int rectH = 50;
color avgRectColor;

void setup() {
  size(640, 480);
  // Initialize camera with matching window resolution (adjust if needed)
  cam = new Capture(this, width, height);
  cam.start();
}

void draw() {
  if (cam.available()) {
    // Step 1: Read new frame and load its pixels
    cam.read();
    cam.loadPixels();
    
    // Step 2: Calculate average color of the rectangle area
    avgRectColor = getAverageColor(rectX, rectY, rectW, rectH);
    
    // Step 3: Draw camera feed and rectangle
    image(cam, 0, 0);
    fill(avgRectColor); // Use the average color to fill the rectangle (optional)
    stroke(255);
    rect(rectX, rectY, rectW, rectH);
    
    // Optional: Display RGB values for debugging
    fill(255);
    text("Avg Color RGB: " + red(avgRectColor) + ", " + green(avgRectColor) + ", " + blue(avgRectColor), 10, height - 20);
  }
}

// Helper function to calculate average color in a rectangular area
color getAverageColor(int x, int y, int w, int h) {
  int totalR = 0, totalG = 0, totalB = 0;
  int validPixels = 0;
  
  // Loop through every pixel in the rectangle
  for (int px = x; px < x + w; px++) {
    for (int py = y; py < y + h; py++) {
      // Skip pixels outside the camera's bounds
      if (px >= 0 && px < cam.width && py >= 0 && py < cam.height) {
        color pixelColor = cam.pixels[py * cam.width + px];
        totalR += red(pixelColor);
        totalG += green(pixelColor);
        totalB += blue(pixelColor);
        validPixels++;
      }
    }
  }
  
  // Avoid division by zero (in case rectangle is empty/off-screen)
  if (validPixels == 0) return color(0);
  
  // Calculate and return average color
  return color(totalR/validPixels, totalG/validPixels, totalB/validPixels);
}

Quick Additional Checks #

• Make sure your rectangle’s rectW and rectH are greater than 0—if either is 0, you won’t sample any pixels!
• If your camera uses a different resolution than your window, adjust the Capture initialization to match, or resize the camera feed to window size before loading pixels (use cam.resize(width, height) after cam.read()).

内容的提问来源于stack exchange，提问作者Hope