在Java 編程中,顏色量化在優化圖像或GIF 文件的調色板方面起著至關重要的作用。此過程涉及減少顏色數量,同時保持原始影像的視覺可接受的表示。
問題陳述:
提供的程式碼在減少顏色方面似乎效率低下有效地。當將超過 256 種顏色的影像減少到 256 種顏色時,會產生明顯的錯誤,例如紅色變成藍色。這表明該演算法很難識別和保留圖像中的重要顏色。
推薦演算法:
範例實作:
以下是Java 中中值切割演算法的範例實作:
import java.util.Arrays; import java.util.Comparator; import java.awt.image.BufferedImage; public class MedianCutQuantizer { public static void quantize(BufferedImage image, int colors) { int[] pixels = image.getRGB(0, 0, image.getWidth(), image.getHeight(), null, 0, image.getWidth()); Arrays.sort(pixels); // Sort pixels by red, green, and blue channel values // Create a binary tree representation of the color space TreeNode root = new TreeNode(pixels); // Recursively divide the color space and create the palette TreeNode[] palette = new TreeNode[colors]; for (int i = 0; i < colors; i++) { palette[i] = root; root = divide(root); } // Replace pixels with their corresponding palette colors for (int i = 0; i < pixels.length; i++) { pixels[i] = getClosestColor(pixels[i], palette); } image.setRGB(0, 0, image.getWidth(), image.getHeight(), pixels, 0, image.getWidth()); } private static TreeNode divide(TreeNode node) { // Find the median color value int median = node.getMedianValue(); // Create two new nodes, one for each half of the color range TreeNode left = new TreeNode(); TreeNode right = new TreeNode(); // Divide the pixels into two halves for (int i = node.start; i < node.end; i++) { if (node.pixels[i] <= median) { left.addPixel(node.pixels[i]); } else { right.addPixel(node.pixels[i]); } } return left.count > right.count ? left : right; } private static int getClosestColor(int pixel, TreeNode[] palette) { int minDistance = Integer.MAX_VALUE; int closestColor = 0; for (TreeNode node : palette) { int distance = getDistance(pixel, node.getAverageValue()); if (distance < minDistance) { minDistance = distance; closestColor = node.getAverageValue(); } } return closestColor; } // Utility methods private static int getDistance(int color1, int color2) { int r1 = (color1 >> 16) & 0xFF; int g1 = (color1 >> 8) & 0xFF; int b1 = color1 & 0xFF; int r2 = (color2 >> 16) & 0xFF; int g2 = (color2 >> 8) & 0xFF; int b2 = color2 & 0xFF; return (r1 - r2) * (r1 - r2) + (g1 - g2) * (g1 - g2) + (b1 - b2) * (b1 - b2); } private static class TreeNode { int start; int end; int count; int[] pixels; Integer averageValue; public TreeNode() { this(new int[0], 0, 0); } public TreeNode(int[] pixels, int start, int end) { this.pixels = pixels; this.start = start; this.end = end; count = end - start; } public int getMedianValue() { return pixels[(start + end) / 2]; } public int getAverageValue() { if (averageValue == null) { int r = 0; int g = 0; int b = 0; for (int i = start; i < end; i++) { int pixel = pixels[i]; r += (pixel >> 16) & 0xFF; g += (pixel >> 8) & 0xFF; b += pixel & 0xFF; } averageValue = (r / count) << 16 | (g / count) << 8 | b / count; } return averageValue; } public void addPixel(int pixel) { int[] newPixels = new int[pixels.length + 1]; System.arraycopy(pixels, start, newPixels, start, end); newPixels[end] = pixel; pixels = newPixels; end++; count = end - start; averageValue = null; } } }
使用此實作或其他類似演算法可以顯著改進Java 應用程式中的顏色量化過程,將影像顏色減少到256或更少時,可以獲得視覺上可接受的結果。
以上是為什麼提供的用於顏色量化的 Java 程式碼很難有效地減少顏色,特別是在將顏色超過 256 的圖像減少到 256 時,導致明顯的錯誤,例如 re的詳細內容。更多資訊請關注PHP中文網其他相關文章!