#!/usr/bin/env python3 """ 自动检测 m02.png 中的小标题位置并在下方写入新内容 基于 easyocr 实现自动文字定位 """ import json import sys import os from pathlib import Path try: import easyocr import numpy as np from PIL import Image, ImageDraw, ImageFont except ImportError as e: print(f"缺少依赖: {e}") print("请安装: pip install easyocr pillow numpy") sys.exit(1) FONT_DIR = Path("/usr/share/fonts/truetype/dejavu") FONT_TEXT = os.path.join(FONT_DIR, "DejaVuSans.ttf") FONT_BOLD = os.path.join(FONT_DIR, "DejaVuSans-Bold.ttf") def _find_font(size, bold=False): try: path = FONT_BOLD if bold else FONT_TEXT return ImageFont.truetype(str(path), size) except Exception: return ImageFont.load_default() def hex_to_rgb(hex_color): h = hex_color.lstrip('#') if len(h) == 3: h = ''.join(c*2 for c in h) return tuple(int(h[i:i+2], 16) for i in (0, 2, 4)) def detect_text_regions(image_path): """使用 easyocr 检测图片中的所有文字区域""" print(" 正在初始化 OCR 模型并扫描图片...") reader = easyocr.Reader(['ch_sim', 'en'], gpu=False, verbose=False) img = Image.open(image_path).convert('RGB') results = reader.readtext(np.array(img)) detections = [] for (bbox, text, conf) in results: xs = [point[0] for point in bbox] ys = [point[1] for point in bbox] x0, y0 = int(min(xs)), int(min(ys)) x1, y1 = int(max(xs)), int(max(ys)) cx = (x0 + x1) // 2 cy = (y0 + y1) // 2 detections.append({ "text": text.strip(), "bbox": [x0, y0, x1, y1], "conf": float(conf), "center": (cx, cy), "left": x0, "top": y0, "right": x1, "bottom": y1, "width": x1 - x0, "height": y1 - y0 }) print(f" 检测到 {len(detections)} 个文本区域。") return detections def sort_text_regions(regions, y_tolerance=20): """对文本区域进行排序：先按行（Y轴），再按列（X轴）""" return sorted(regions, key=lambda r: (r["center"][1], r["center"][0])) def auto_fill_text(template_path, output_path, new_contents, font_size=24, text_color="#000000", y_offset=15, debug=False): """自动检测小标题并在下方写入新内容""" try: detections = detect_text_regions(template_path) if not detections: print(" 错误: 未检测到任何文本区域") return False sorted_detections = sort_text_regions(detections) # 过滤 filtered = [d for d in sorted_detections if 30 <= d["width"] <= 800 and 10 <= d["height"] <= 50] print(f" 过滤后剩余 {len(filtered)} 个有效文本区域。") if len(new_contents) > len(filtered): new_contents = new_contents[:len(filtered)] elif len(new_contents) < len(filtered): filtered = filtered[:len(new_contents)] img = Image.open(template_path).convert("RGB") draw = ImageDraw.Draw(img) font = _find_font(font_size, bold=True) coord_mapping = [] for idx, (det, new_text) in enumerate(zip(filtered, new_contents)): left_x = det["left"] bottom_y = det["bottom"] target_x = left_x target_y = bottom_y + y_offset img_w, img_h = img.size target_x = max(0, min(target_x, img_w - 1)) target_y = max(0, min(target_y, img_h - 1)) coord_mapping.append({ "index": idx, "detected_text": det["text"], "new_text": new_text, "detection_bbox": det["bbox"], "write_position": [target_x, target_y], "detection_confidence": det["conf"] }) if debug: bbox = det["bbox"] draw.rectangle([bbox[0], bbox[1], bbox[2], bbox[3]], outline="#0000FF", width=1) draw.line([target_x-5, target_y, target_x+5, target_y], fill="#FF0000", width=1) draw.line([target_x, target_y-5, target_x, target_y+5], fill="#FF0000", width=1) draw.text((target_x, target_y - 20), str(idx), fill="#FF00FF", font=_find_font(12)) draw.text((target_x, target_y), new_text, font=font, fill=text_color) print(f" [标题 {idx+1:2d}]: '{det['text'][:15]:<15}' -> 在 ({target_x:4d}, {target_y:4d}) 写入: '{new_text}'") img.save(output_path, dpi=(300, 300), quality=95) print(f"\n 处理完成！结果已保存至: {output_path}") # 保存坐标映射 JSON json_path = output_path.rsplit('.', 1)[0] + '_coords.json' with open(json_path, 'w', encoding='utf-8') as f: json.dump({"template": template_path, "output": output_path, "settings": {"font_size": font_size, "text_color": text_color, "y_offset": y_offset}, "mapping": coord_mapping}, f, ensure_ascii=False, indent=2) print(f" 坐标映射已保存至: {json_path}") return True except Exception as e: print(f" 错误: {e}") import traceback traceback.print_exc() return False def main(): import argparse parser = argparse.ArgumentParser(description="自动检测 m02.png 中的小标题并在下方写入新内容") parser.add_argument("--template", "-t", required=True, help="模板图片路径") parser.add_argument("--output", "-o", required=True, help="输出图片路径") parser.add_argument("--contents", "-c", required=True, help="新内容的 JSON 文件路径或 JSON 字符串") parser.add_argument("--font-size", "-f", type=int, default=24, help="字号 (默认: 24)") parser.add_argument("--color", "-cl", default="#000000", help="颜色 (十六进制, 默认: #000000)") parser.add_argument("--y-offset", "-y", type=int, default=15, help="垂直偏移 (默认: 15)") parser.add_argument("--debug", "-d", action="store_true", help="调试模式") args = parser.parse_args() try: if os.path.isfile(args.contents): with open(args.contents, 'r', encoding='utf-8') as f: new_contents = json.load(f) else: new_contents = json.loads(args.contents) if not isinstance(new_contents, list): print(" 错误: 输入必须是一个字符串列表") sys.exit(1) except Exception as e: print(f" 错误: 无法解析新内容: {e}") sys.exit(1) if not os.path.isfile(args.template): print(f" 错误: 模板文件不存在: {args.template}") sys.exit(1) print("=" * 60) print(" 自动检测并填充文字工具") print("=" * 60) print(f" 模板: {args.template} | 输出: {args.output}") print(f" 内容数: {len(new_contents)} | 字号: {args.font_size} | 偏移: {args.y_offset}px") print(f" 调试: {'开' if args.debug else '关'}") print("-" * 60) success = auto_fill_text(args.template, args.output, new_contents, args.font_size, args.color, args.y_offset, args.debug) sys.exit(0 if success else 1) if __name__ == "__main__": main()