k-centroid-scaler 1.2.3 → 1.3.0

package/k_centroid_scaler.d.ts

@@ -1,9 +1,23 @@
 /* tslint:disable */
 /* eslint-disable */
 export function k_centroid_resize(image_data: Uint8Array, original_width: number, original_height: number, target_width: number, target_height: number, centroids: number, iterations: number): ImageResult;
+/**
+ * Stage 1: Fast quantization to 256 colors or less using k-means++ in RGB space
+ * This stage is fast and doesn't alter tints significantly
+ */
+export function quantize_colors_stage1(image_data: Uint8Array, width: number, height: number, target_colors: number): ImageResult;
+/**
+ * Stage 2: Perceptually accurate quantization using CIELAB color space
+ * Best for final color reduction where human perception matters
+ */
+export function quantize_colors_stage2_lab(image_data: Uint8Array, width: number, height: number, target_colors: number, dithering: boolean): ImageResult;
+/**
+ * Complete two-stage quantization: Stage 1 (fast) -> Stage 2 (perceptual)
+ */
+export function quantize_colors_two_stage(image_data: Uint8Array, width: number, height: number, target_colors: number, dithering: boolean): ImageResult;
 export function quantize_colors_median_cut(image_data: Uint8Array, width: number, height: number, num_colors: number, dithering: boolean): ImageResult;
-export function quantize_colors_kmeans(image_data: Uint8Array, width: number, height: number, num_colors: number, iterations: number, dithering: boolean): ImageResult;
-export function extract_palette(image_data: Uint8Array, num_colors: number): ColorPalette;
+export function quantize_colors_kmeans(image_data: Uint8Array, width: number, height: number, num_colors: number, _iterations: number, dithering: boolean): ImageResult;
+export function extract_palette(image_data: Uint8Array, max_colors: number): ColorPalette;
 export function analyze_colors(image_data: Uint8Array, max_colors: number, sort_method: string): ColorAnalysis;
 export function get_dominant_colors(image_data: Uint8Array, num_colors: number, min_coverage: number): ColorAnalysis;
 export function process_image(data: Uint8Array, width: number, height: number, target_width: number, target_height: number, centroids: number, iterations: number): any;
@@ -65,6 +79,8 @@ export interface InitOutput {
   readonly __wbg_set_coloranalysis_total_pixels: (a: number, b: number) => void;
   readonly coloranalysis_colors: (a: number, b: number) => void;
   readonly k_centroid_resize: (a: number, b: number, c: number, d: number, e: number, f: number, g: number, h: number) => number;
+  readonly quantize_colors_stage1: (a: number, b: number, c: number, d: number, e: number) => number;
+  readonly quantize_colors_stage2_lab: (a: number, b: number, c: number, d: number, e: number, f: number) => number;
   readonly quantize_colors_median_cut: (a: number, b: number, c: number, d: number, e: number, f: number) => number;
   readonly quantize_colors_kmeans: (a: number, b: number, c: number, d: number, e: number, f: number, g: number) => number;
   readonly extract_palette: (a: number, b: number, c: number) => number;
@@ -73,6 +89,7 @@ export interface InitOutput {
   readonly process_image: (a: number, b: number, c: number, d: number, e: number, f: number, g: number, h: number) => number;
   readonly __wbg_set_imageresult_width: (a: number, b: number) => void;
   readonly __wbg_set_imageresult_height: (a: number, b: number) => void;
+  readonly quantize_colors_two_stage: (a: number, b: number, c: number, d: number, e: number, f: number) => number;
   readonly __wbg_get_colorpalette_count: (a: number) => number;
   readonly __wbg_get_imageresult_height: (a: number) => number;
   readonly __wbg_get_imageresult_width: (a: number) => number;

package/k_centroid_scaler.js

@@ -233,6 +233,55 @@ export function k_centroid_resize(image_data, original_width, original_height, t
     return ImageResult.__wrap(ret);
 }
 
+/**
+ * Stage 1: Fast quantization to 256 colors or less using k-means++ in RGB space
+ * This stage is fast and doesn't alter tints significantly
+ * @param {Uint8Array} image_data
+ * @param {number} width
+ * @param {number} height
+ * @param {number} target_colors
+ * @returns {ImageResult}
+ */
+export function quantize_colors_stage1(image_data, width, height, target_colors) {
+    const ptr0 = passArray8ToWasm0(image_data, wasm.__wbindgen_export_1);
+    const len0 = WASM_VECTOR_LEN;
+    const ret = wasm.quantize_colors_stage1(ptr0, len0, width, height, target_colors);
+    return ImageResult.__wrap(ret);
+}
+
+/**
+ * Stage 2: Perceptually accurate quantization using CIELAB color space
+ * Best for final color reduction where human perception matters
+ * @param {Uint8Array} image_data
+ * @param {number} width
+ * @param {number} height
+ * @param {number} target_colors
+ * @param {boolean} dithering
+ * @returns {ImageResult}
+ */
+export function quantize_colors_stage2_lab(image_data, width, height, target_colors, dithering) {
+    const ptr0 = passArray8ToWasm0(image_data, wasm.__wbindgen_export_1);
+    const len0 = WASM_VECTOR_LEN;
+    const ret = wasm.quantize_colors_stage2_lab(ptr0, len0, width, height, target_colors, dithering);
+    return ImageResult.__wrap(ret);
+}
+
+/**
+ * Complete two-stage quantization: Stage 1 (fast) -> Stage 2 (perceptual)
+ * @param {Uint8Array} image_data
+ * @param {number} width
+ * @param {number} height
+ * @param {number} target_colors
+ * @param {boolean} dithering
+ * @returns {ImageResult}
+ */
+export function quantize_colors_two_stage(image_data, width, height, target_colors, dithering) {
+    const ptr0 = passArray8ToWasm0(image_data, wasm.__wbindgen_export_1);
+    const len0 = WASM_VECTOR_LEN;
+    const ret = wasm.quantize_colors_median_cut(ptr0, len0, width, height, target_colors, dithering);
+    return ImageResult.__wrap(ret);
+}
+
 /**
  * @param {Uint8Array} image_data
  * @param {number} width
@@ -253,26 +302,26 @@ export function quantize_colors_median_cut(image_data, width, height, num_colors
  * @param {number} width
  * @param {number} height
  * @param {number} num_colors
- * @param {number} iterations
+ * @param {number} _iterations
  * @param {boolean} dithering
  * @returns {ImageResult}
  */
-export function quantize_colors_kmeans(image_data, width, height, num_colors, iterations, dithering) {
+export function quantize_colors_kmeans(image_data, width, height, num_colors, _iterations, dithering) {
     const ptr0 = passArray8ToWasm0(image_data, wasm.__wbindgen_export_1);
     const len0 = WASM_VECTOR_LEN;
-    const ret = wasm.quantize_colors_kmeans(ptr0, len0, width, height, num_colors, iterations, dithering);
+    const ret = wasm.quantize_colors_kmeans(ptr0, len0, width, height, num_colors, _iterations, dithering);
     return ImageResult.__wrap(ret);
 }
 
 /**
  * @param {Uint8Array} image_data
- * @param {number} num_colors
+ * @param {number} max_colors
  * @returns {ColorPalette}
  */
-export function extract_palette(image_data, num_colors) {
+export function extract_palette(image_data, max_colors) {
     const ptr0 = passArray8ToWasm0(image_data, wasm.__wbindgen_export_1);
     const len0 = WASM_VECTOR_LEN;
-    const ret = wasm.extract_palette(ptr0, len0, num_colors);
+    const ret = wasm.extract_palette(ptr0, len0, max_colors);
     return ColorPalette.__wrap(ret);
 }
 

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "k-centroid-scaler",
   "type": "module",
-  "version": "1.2.3",
+  "version": "1.3.0",
   "files": [
     "k_centroid_scaler_bg.wasm",
     "k_centroid_scaler.js",