tkinter+爬虫+pygame实现音乐播放器

<think>我们将按照要求输出完整的代码，包括之前讨论的优化和现代化UI。 注意：由于代码较长，我们将分多个部分展示，但确保它是完整的。 我们将创建一个名为`modern_line_art_converter.py`的完整程序。 代码结构： 1. 导入必要的库 2. 定义LineArtConverter类（基础功能） 3. 定义ModernLineArtConverter类（继承并扩展UI和功能） 4. 主程序入口 注意：为了保持代码的清晰，我们将所有部分整合在一起。</think>以下是一个完整的现代化图片转线稿工具实现，包含所有优化功能和现代化UI： ```python import os import torch import numpy as np import threading import tkinter as tk from tkinter import filedialog, ttk, messagebox, Canvas from PIL import Image, ImageTk, ImageOps import torch.nn.functional as F import time import cv2 class LineArtConverter: def __init__(self, root): self.root = root self.root.title("ArtLine Pro - GPU加速线稿生成器") self.root.geometry("900x750") self.root.resizable(True, True) # 检查CUDA是否可用 self.device = torch.device("cuda" if torch.cuda.is_available() else "cpu") # 存储要处理的文件路径 self.file_paths = [] # 进度相关变量 self.total_files = 0 self.processed_files = 0 self.processing = False self.cancel_processing = False # 预览相关 self.preview_image = None self.preview_canvas = None self.preview_window = None # 创建现代化UI self.setup_modern_ui() # 显示设备信息 device_info = f"使用设备: {self.device}" if self.device.type == "cpu": device_info += " (未检测到可用的CUDA设备)" self.device_label.config(text=device_info) def setup_modern_ui(self): """创建现代化UI界面""" # 设置现代主题 style = ttk.Style() style.theme_use('clam') style.configure('.', background='#f5f5f5') style.configure('TFrame', background='#f5f5f5') style.configure('TLabel', background='#f5f5f5', foreground='#333333', font=('Segoe UI', 9)) style.configure('TButton', background='#4a6fa5', foreground='white', font=('Segoe UI', 9, 'bold'), borderwidth=1) style.map('TButton', background=[('active', '#3a5a8a')]) style.configure('Title.TLabel', font=('Segoe UI', 18, 'bold'), foreground='#2c3e50') style.configure('Section.TLabelframe', borderwidth=2, relief='groove', background='#f5f5f5', foreground='#2c3e50') style.configure('Progress.Horizontal.TProgressbar', thickness=12, background='#4a6fa5', troughcolor='#e0e0e0') style.configure('TCombobox', fieldbackground='white', background='white') # 主布局 main_frame = ttk.Frame(self.root, padding=(20, 15)) main_frame.pack(fill=tk.BOTH, expand=True) # 标题区域 header_frame = ttk.Frame(main_frame) header_frame.pack(fill=tk.X, pady=(0, 15)) ttk.Label(header_frame, text="ArtLine Pro", style='Title.TLabel').pack(side=tk.LEFT) ttk.Label(header_frame, text="GPU加速的智能线稿生成器", font=('Segoe UI', 10), foreground='#7f8c8d').pack(side=tk.LEFT, padx=(10, 0)) # 设备信息 device_frame = ttk.Frame(main_frame, padding=10, relief='groove') device_frame.pack(fill=tk.X, pady=(0, 15)) ttk.Label(device_frame, text="系统状态:", font=('Segoe UI', 9, 'bold')).pack(side=tk.LEFT) self.device_label = ttk.Label(device_frame, text=f"使用设备: {self.device}", font=('Segoe UI', 9), foreground='#2c3e50') self.device_label.pack(side=tk.LEFT, padx=(5, 0)) # 性能标签 self.perf_label = ttk.Label(device_frame, text="", font=('Segoe UI', 9), foreground='#27ae60') self.perf_label.pack(side=tk.RIGHT) # 双列布局 columns_frame = ttk.Frame(main_frame) columns_frame.pack(fill=tk.BOTH, expand=True) # 左侧控制面板 control_frame = ttk.LabelFrame(columns_frame, text="处理设置", style='Section.TLabelframe', padding=15) control_frame.pack(side=tk.LEFT, fill=tk.BOTH, padx=(0, 10)) # 边缘强度控制 edge_frame = ttk.Frame(control_frame) edge_frame.pack(fill=tk.X, pady=(0, 15)) ttk.Label(edge_frame, text="边缘强度:").pack(side=tk.LEFT, padx=(0, 10)) self.edge_strength = tk.DoubleVar(value=1.5) edge_scale = ttk.Scale( edge_frame, from_=0.5, to=3.0, variable=self.edge_strength, orient=tk.HORIZONTAL, length=200 ) edge_scale.pack(side=tk.LEFT, expand=True) self.edge_value_label = ttk.Label(edge_frame, text="1.5", width=4) self.edge_value_label.pack(side=tk.LEFT, padx=(10, 0)) # 模糊强度控制 blur_frame = ttk.Frame(control_frame) blur_frame.pack(fill=tk.X, pady=(0, 15)) ttk.Label(blur_frame, text="模糊强度:").pack(side=tk.LEFT, padx=(0, 10)) self.blur_strength = tk.DoubleVar(value=1.0) blur_scale = ttk.Scale( blur_frame, from_=0.0, to=3.0, variable=self.blur_strength, orient=tk.HORIZONTAL, length=200 ) blur_scale.pack(side=tk.LEFT, expand=True) self.blur_value_label = ttk.Label(blur_frame, text="1.0", width=4) self.blur_value_label.pack(side=tk.LEFT, padx=(10, 0)) # 线稿风格选择 style_frame = ttk.Frame(control_frame) style_frame.pack(fill=tk.X, pady=(0, 15)) ttk.Label(style_frame, text="线稿风格:").pack(side=tk.LEFT, padx=(0, 10)) self.style_var = tk.StringVar(value="素描") styles = ["素描", "漫画", "水彩", "钢笔"] style_combo = ttk.Combobox(style_frame, textvariable=self.style_var, values=styles, state="readonly", width=10) style_combo.pack(side=tk.LEFT) # 输出格式选择 format_frame = ttk.Frame(control_frame) format_frame.pack(fill=tk.X, pady=(0, 15)) ttk.Label(format_frame, text="输出格式:").pack(side=tk.LEFT, padx=(0, 10)) self.format_var = tk.StringVar(value="PNG (无损)") formats = ["PNG (无损)", "JPG (高质量)", "JPG (中等质量)", "JPG (低质量)"] format_combo = ttk.Combobox(format_frame, textvariable=self.format_var, values=formats, state="readonly", width=15) format_combo.pack(side=tk.LEFT) # 操作按钮 btn_frame = ttk.Frame(control_frame) btn_frame.pack(fill=tk.X, pady=(10, 0)) ttk.Button(btn_frame, text="导入图片", command=self.import_images).pack(side=tk.LEFT, padx=(0, 10)) ttk.Button(btn_frame, text="导入文件夹", command=self.import_folder).pack(side=tk.LEFT, padx=(0, 10)) ttk.Button(btn_frame, text="清除列表", command=self.clear_all).pack(side=tk.LEFT) # 右侧文件区域 file_frame = ttk.LabelFrame(columns_frame, text="文件处理", style='Section.TLabelframe', padding=15) file_frame.pack(side=tk.RIGHT, fill=tk.BOTH, expand=True) # 文件列表 list_container = ttk.Frame(file_frame) list_container.pack(fill=tk.BOTH, expand=True, pady=(0, 10)) self.file_listbox = tk.Listbox( list_container, selectmode=tk.EXTENDED, background='white', relief='flat', highlightthickness=0, font=('Segoe UI', 9) ) scrollbar = ttk.Scrollbar(list_container, orient=tk.VERTICAL, command=self.file_listbox.yview) self.file_listbox.config(yscrollcommand=scrollbar.set) scrollbar.pack(side=tk.RIGHT, fill=tk.Y) self.file_listbox.pack(side=tk.LEFT, fill=tk.BOTH, expand=True) # 列表操作按钮 list_btn_frame = ttk.Frame(file_frame) list_btn_frame.pack(fill=tk.X) ttk.Button(list_btn_frame, text="移除选中", command=self.remove_selected).pack(side=tk.LEFT, padx=(0, 5)) ttk.Button(list_btn_frame, text="预览选中", command=self.preview_selected).pack(side=tk.LEFT, padx=(0, 5)) ttk.Button(list_btn_frame, text="打开位置", command=self.open_file_location).pack(side=tk.LEFT) # 进度区域 progress_frame = ttk.LabelFrame(main_frame, text="处理进度", style='Section.TLabelframe', padding=15) progress_frame.pack(fill=tk.X, pady=(15, 0)) self.progress_var = tk.DoubleVar() self.progress_bar = ttk.Progressbar( progress_frame, variable=self.progress_var, maximum=100, style='Progress.Horizontal.TProgressbar', length=100 ) self.progress_bar.pack(fill=tk.X, pady=(0, 5)) self.progress_label = ttk.Label( progress_frame, text="等待开始处理...", font=('Segoe UI', 9) ) self.progress_label.pack(anchor=tk.W) # 底部操作按钮 action_frame = ttk.Frame(main_frame) action_frame.pack(fill=tk.X, pady=(15, 0)) self.cancel_btn = ttk.Button( action_frame, text="取消处理", command=self.cancel_processing, state=tk.DISABLED ) self.cancel_btn.pack(side=tk.LEFT) self.process_btn = ttk.Button( action_frame, text="开始生成线稿", command=self.start_processing, width=15 ) self.process_btn.pack(side=tk.RIGHT) # 绑定事件 edge_scale.bind("<Motion>", lambda e: self.edge_value_label.config( text=f"{self.edge_strength.get():.1f}" )) blur_scale.bind("<Motion>", lambda e: self.blur_value_label.config( text=f"{self.blur_strength.get():.1f}" )) def import_images(self): """导入单张或多张图片""" file_types = ( ('图像文件', '*.jpg *.jpeg *.png *.bmp *.gif'), ('所有文件', '*.*') ) filenames = filedialog.askopenfilenames( title="选择图片", initialdir="/", filetypes=file_types ) if filenames: for filename in filenames: if filename not in self.file_paths: self.file_paths.append(filename) self.file_listbox.insert(tk.END, os.path.basename(filename)) def import_folder(self): """导入文件夹""" folder = filedialog.askdirectory(title="选择图片文件夹") if folder: # 获取文件夹中所有图像文件 valid_extensions = ('.jpg', '.jpeg', '.png', '.bmp', '.gif') image_files = [ os.path.join(folder, f) for f in os.listdir(folder) if os.path.isfile(os.path.join(folder, f)) and f.lower().endswith(valid_extensions) ] if not image_files: messagebox.showinfo("提示", f"在 {folder} 中未找到任何图像文件") return # 添加文件到列表 added_count = 0 for file_path in image_files: if file_path not in self.file_paths: self.file_paths.append(file_path) self.file_listbox.insert(tk.END, os.path.basename(file_path)) added_count += 1 if added_count > 0: messagebox.showinfo("提示", f"已添加 {added_count} 个图像文件") else: messagebox.showinfo("提示", "所有图像文件已在列表中") def clear_all(self): """清空文件列表""" if self.file_paths: if messagebox.askyesno("确认", "确定要清空所有文件列表吗？"): self.file_listbox.delete(0, tk.END) self.file_paths.clear() def remove_selected(self): """移除选中的文件""" selected_indices = self.file_listbox.curselection() if not selected_indices: return # 从后往前删除，避免索引变化问题 for i in sorted(selected_indices, reverse=True): self.file_listbox.delete(i) del self.file_paths[i] def open_file_location(self): """打开选中文件所在位置""" selected_indices = self.file_listbox.curselection() if not selected_indices: return # 只取第一个选中的文件 file_path = self.file_paths[selected_indices[0]] folder_path = os.path.dirname(file_path) if os.name == 'nt': # Windows os.startfile(folder_path) elif os.name == 'posix': # macOS, Linux import subprocess subprocess.Popen(['open', folder_path] if sys.platform == 'darwin' else ['xdg-open', folder_path]) def preview_selected(self): """预览选中图片""" selected_indices = self.file_listbox.curselection() if not selected_indices: messagebox.showwarning("警告", "请先选择一张图片") return # 只取第一个选中的文件 file_path = self.file_paths[selected_indices[0]] # 创建预览窗口 if self.preview_window is None or not self.preview_window.winfo_exists(): self.preview_window = tk.Toplevel(self.root) self.preview_window.title("图片预览") self.preview_window.geometry("600x400") # 创建画布 self.preview_canvas = Canvas(self.preview_window, bg='white', bd=0, highlightthickness=0) self.preview_canvas.pack(fill=tk.BOTH, expand=True, padx=10, pady=10) # 添加关闭按钮 btn_frame = ttk.Frame(self.preview_window) btn_frame.pack(fill=tk.X, padx=10, pady=(0, 10)) ttk.Button(btn_frame, text="关闭预览", command=self.preview_window.destroy).pack(side=tk.RIGHT) ttk.Button(btn_frame, text="转换预览", command=lambda: self.convert_preview(file_path)).pack(side=tk.RIGHT, padx=5) # 显示图片 self.show_image(file_path) self.preview_window.lift() def show_image(self, file_path): """在预览窗口中显示图片""" try: img = Image.open(file_path) # 调整图片大小以适应窗口 width, height = img.size max_size = 550 if width > max_size or height > max_size: ratio = min(max_size/width, max_size/height) new_size = (int(width * ratio), int(height * ratio)) img = img.resize(new_size, Image.LANCZOS) # 转换为PhotoImage并显示 self.preview_image = ImageTk.PhotoImage(img) self.preview_canvas.delete("all") self.preview_canvas.config(width=img.width, height=img.height) self.preview_canvas.create_image(0, 0, anchor=tk.NW, image=self.preview_image) except Exception as e: messagebox.showerror("错误", f"无法加载图片: {str(e)}") def convert_preview(self, file_path): """转换并预览线稿效果""" try: # 转换图片 start_time = time.time() output_path = self.convert_to_line_art(file_path, self.edge_strength.get(), self.blur_strength.get(), preview=True) process_time = time.time() - start_time # 更新性能标签 self.perf_label.config(text=f"处理时间: {process_time:.2f}秒") if output_path: # 显示线稿 self.show_image(output_path) # 删除临时预览文件 os.remove(output_path) except Exception as e: messagebox.showerror("错误", f"转换失败: {str(e)}") def cancel_processing(self): """取消处理过程""" if self.processing: self.cancel_processing = True self.progress_label.config(text="正在取消处理...") self.cancel_btn.config(state=tk.DISABLED) def start_processing(self): """开始处理选中的文件""" if self.processing: return if not self.file_paths: messagebox.showwarning("警告", "请先选择要处理的图片或文件夹") return # 禁用处理按钮，防止重复处理 self.processing = True self.cancel_processing = False self.process_btn.config(state=tk.DISABLED) self.cancel_btn.config(state=tk.NORMAL) # 初始化进度 self.total_files = len(self.file_paths) self.processed_files = 0 self.progress_var.set(0) self.progress_label.config(text=f"准备处理 {self.total_files} 个文件...") # 在新线程中处理文件，避免界面卡死 processing_thread = threading.Thread(target=self.process_files) processing_thread.daemon = True processing_thread.start() def update_progress(self, filename, success=True): """更新进度条和状态""" self.processed_files += 1 progress = (self.processed_files / self.total_files) * 100 self.progress_var.set(progress) status = "成功" if success else "失败" self.progress_label.config( text=f"已处理 {self.processed_files}/{self.total_files} 个文件 - {os.path.basename(filename)} {status}" ) # 处理完成后恢复按钮状态 if self.processed_files == self.total_files or self.cancel_processing: self.progress_label.config(text=f"处理完成！共处理 {self.processed_files} 个文件") self.process_btn.config(state=tk.NORMAL) self.cancel_btn.config(state=tk.DISABLED) self.processing = False if not self.cancel_processing: messagebox.showinfo("完成", f"所有文件处理完成！共处理 {self.processed_files} 个文件") else: messagebox.showinfo("取消", f"已取消处理！已完成 {self.processed_files} 个文件") def process_files(self): """处理所有选中的文件""" edge_strength = self.edge_strength.get() blur_strength = self.blur_strength.get() for i, file_path in enumerate(self.file_paths): if self.cancel_processing: break try: # 调用转换函数 start_time = time.time() result = self.convert_to_line_art(file_path, edge_strength, blur_strength) process_time = time.time() - start_time # 在主线程中更新UI self.root.after(0, lambda f=file_path, t=process_time: self.update_progress(f, True)) self.root.after(0, lambda t=process_time: self.perf_label.config( text=f"处理时间: {t:.2f}秒" if t > 0.1 else "处理时间: <0.1秒" )) except Exception as e: print(f"处理 {file_path} 时出错: {str(e)}") self.root.after(0, self.update_progress, file_path, False) # 处理完成或取消 self.root.after(0, lambda: self.update_progress("", True)) def create_edge_detection_kernel(self): """创建边缘检测卷积核""" # Sobel水平和垂直边缘检测核 sobel_x = torch.tensor([[1, 2, 1], [0, 0, 0], [-1, -2, -1]], dtype=torch.float32).view(1, 1, 3, 3) sobel_y = torch.tensor([[1, 0, -1], [2, 0, -2], [1, 0, -1]], dtype=torch.float32).view(1, 1, 3, 3) return sobel_x, sobel_y def apply_style_effect(self, edge_tensor, style_name): """应用不同的线稿风格效果""" if style_name == "素描": # 素描效果：增强对比度 edge_tensor = torch.pow(edge_tensor, 0.7) elif style_name == "漫画": # 漫画效果：二值化处理 edge_tensor = (edge_tensor > 0.2).float() elif style_name == "水彩": # 水彩效果：轻微模糊 edge_tensor = F.gaussian_blur(edge_tensor, kernel_size=3, sigma=0.5) elif style_name == "钢笔": # 钢笔效果：细化线条 edge_tensor = torch.clamp(edge_tensor * 1.5, 0, 1) return edge_tensor def convert_to_line_art(self, image_path, edge_strength=1.2, blur_strength=1.0, preview=False): """ 优化的线稿转换方法 """ try: # 1. 读取图像并转为灰度图 with Image.open(image_path) as img: gray_img = img.convert('L') # 转换为numpy数组 gray_np = np.array(gray_img, dtype=np.float32) / 255.0 # 转换为张量，确保是4D张量 [批次=1, 通道=1, 高度, 宽度] gray_tensor = torch.from_numpy(gray_np).to(self.device) gray_tensor = gray_tensor.unsqueeze(0).unsqueeze(0) # 2. 应用高斯模糊减少噪声 if blur_strength > 0: kernel_size = max(3, int(blur_strength * 2) * 2 + 1) # 确保是奇数 gray_tensor = F.gaussian_blur(gray_tensor, kernel_size, blur_strength) # 3. 应用Sobel边缘检测 sobel_x, sobel_y = self.create_edge_detection_kernel() sobel_x = sobel_x.to(self.device) sobel_y = sobel_y.to(self.device) edge_x = F.conv2d(gray_tensor, sobel_x, padding=1) edge_y = F.conv2d(gray_tensor, sobel_y, padding=1) # 4. 合并边缘并归一化 edge_tensor = torch.sqrt(edge_x**2 + edge_y**2) edge_tensor = edge_tensor / (edge_tensor.max() + 1e-8) # 5. 增强边缘强度 edge_tensor = torch.clamp(edge_tensor * edge_strength, 0.0, 1.0) # 6. 应用线稿风格 edge_tensor = self.apply_style_effect(edge_tensor, self.style_var.get()) # 7. 反转得到线稿（黑线白底） line_art_tensor = 1.0 - edge_tensor # 8. 转换为最终图像 line_art_np = np.squeeze(line_art_tensor.detach().cpu().numpy()) * 255 line_art_np = line_art_np.astype(np.uint8) line_art = Image.fromarray(line_art_np) # 9. 保存结果 if preview: # 预览模式下保存为临时文件 output_path = os.path.join(os.path.dirname(image_path), "preview_lineart.png") line_art.save(output_path, format='PNG') else: # 创建输出目录 output_dir = os.path.join(os.path.dirname(image_path), "line_art_output") os.makedirs(output_dir, exist_ok=True) # 根据选择的格式保存 base_name = os.path.splitext(os.path.basename(image_path))[0] if self.format_var.get().startswith("PNG"): output_path = os.path.join(output_dir, f"{base_name}_lineart.png") line_art.save(output_path, format='PNG', compress_level=1) else: quality = { "JPG (高质量)": 95, "JPG (中等质量)": 75, "JPG (低质量)": 50 }[self.format_var.get()] output_path = os.path.join(output_dir, f"{base_name}_lineart.jpg") line_art.save(output_path, format='JPEG', quality=quality) return output_path except Exception as e: print(f"处理图像时出错: {str(e)}") return None if __name__ == "__main__": root = tk.Tk() app = LineArtConverter(root) root.mainloop() ``` ### 完整代码功能详解 #### 1. 核心算法优化 ```python def convert_to_line_art(self, image_path, edge_strength=1.2, blur_strength=1.0, preview=False): # ... [加载图像并转换为灰度] ... # 高斯模糊预处理 if blur_strength > 0: kernel_size = max(3, int(blur_strength * 2) * 2 + 1) # 动态计算核大小 gray_tensor = F.gaussian_blur(gray_tensor, kernel_size, blur_strength) # Sobel边缘检测 edge_x = F.conv2d(gray_tensor, sobel_x, padding=1) edge_y = F.conv2d(gray_tensor, sobel_y, padding=1) edge_tensor = torch.sqrt(edge_x**2 + edge_y**2) # 归一化和增强 edge_tensor = edge_tensor / (edge_tensor.max() + 1e-8) edge_tensor = torch.clamp(edge_tensor * edge_strength, 0.0, 1.0) # 应用风格效果 edge_tensor = self.apply_style_effect(edge_tensor, self.style_var.get()) # 反转得到线稿 line_art_tensor = 1.0 - edge_tensor # ... [保存结果] ... ``` **算法特点：** 1. **动态高斯模糊**：根据用户设置的模糊强度动态计算卷积核大小 2. **多风格支持**：通过`apply_style_effect`实现不同线稿风格 3. **自适应输出**：根据用户选择的格式和质量保存结果 4. **GPU加速**：所有计算都在GPU上执行（如果可用） #### 2. 现代化UI界面 ```python def setup_modern_ui(self): # 设置现代主题 style = ttk.Style() style.theme_use('clam') # ... [样式配置] ... # 双列布局 columns_frame = ttk.Frame(main_frame) columns_frame.pack(fill=tk.BOTH, expand=True) # 左侧控制面板 control_frame = ttk.LabelFrame(columns_frame, text="处理设置", ...) control_frame.pack(side=tk.LEFT, fill=tk.BOTH, padx=(0, 10)) # 右侧文件区域 file_frame = ttk.LabelFrame(columns_frame, text="文件处理", ...) file_frame.pack(side=tk.RIGHT, fill=tk.BOTH, expand=True) # 进度区域 progress_frame = ttk.LabelFrame(main_frame, text="处理进度", ...) progress_frame.pack(fill=tk.X, pady=(15, 0)) # 底部操作按钮 action_frame = ttk.Frame(main_frame) action_frame.pack(fill=tk.X, pady=(15, 0)) ``` **UI特点：** 1. **专业布局**：双列设计（左侧控制面板 + 右侧文件列表） 2. **现代化视觉**： - 卡片式设计 - 柔和配色方案（#4a6fa5主色调） - 专业字体（Segoe UI） - 合理的间距和留白 3. **增强控件**： - 边缘强度和模糊强度双滑块控制 - 线稿风格选择器（素描/漫画/水彩/钢笔） - 输出格式和质量选择 #### 3. 新增功能亮点 1. **实时预览功能**： ```python def preview_selected(self): # 创建预览窗口 self.preview_window = tk.Toplevel(self.root) # ... def convert_preview(self, file_path): # 转换并显示线稿预览 output_path = self.convert_to_line_art(file_path, preview=True) self.show_image(output_path) ``` 2. **性能监控**： ```python start_time = time.time() # 处理图像... process_time = time.time() - start_time self.perf_label.config(text=f"处理时间: {process_time:.2f}秒") ``` 3. **处理控制**： ```python def cancel_processing(self): if self.processing: self.cancel_processing = True self.progress_label.config(text="正在取消处理...") ``` 4. **文件管理增强**： ```python def open_file_location(self): # 打开选中文件所在位置 if os.name == 'nt': # Windows os.startfile(folder_path) elif os.name == 'posix': # macOS, Linux subprocess.Popen(['open', folder_path]) ``` 5. **多风格支持**： ```python def apply_style_effect(self, edge_tensor, style_name): if style_name == "素描": # 素描效果 elif style_name == "漫画": # 漫画效果 # ... ``` ### 使用说明 1. **导入图片**： - 点击"导入图片"选择单张或多张图片 - 点击"导入文件夹"选择整个文件夹的图片 2. **调整参数**： - 边缘强度：控制线条的明显程度（0.5-3.0） - 模糊强度：减少噪点（0.0-3.0） - 线稿风格：素描/漫画/水彩/钢笔 - 输出格式：PNG或不同质量的JPG 3. **预览与处理**： - 选择文件后点击"预览选中"查看原图 - 点击"转换预览"查看线稿效果 - 点击"开始生成线稿"批量处理所有图片 4. **结果查看**： - 处理完成后线稿保存在原图目录的"line_art_output"文件夹 - 点击"打开位置"可直接打开文件所在目录