#!/usr/bin/env python3 """ OCR验证工具的布局管理模块 包含标准布局、滚动布局、紧凑布局的实现 """ import streamlit as st from pathlib import Path from PIL import Image from typing import Dict, List, Optional import plotly.graph_objects as go from ocr_validator_utils import ( convert_html_table_to_markdown, parse_html_tables, draw_bbox_on_image, rotate_image_and_coordinates ) class OCRLayoutManager: """OCR布局管理器""" def __init__(self, validator): self.validator = validator self.config = validator.config self._rotated_image_cache = {} # 缓存旋转后的图像 def get_rotation_angle(self) -> float: """从OCR数据中获取旋转角度""" if self.validator.ocr_data: for item in self.validator.ocr_data: if isinstance(item, dict) and 'rotation_angle' in item: return item['rotation_angle'] return 0.0 def load_and_rotate_image(self, image_path: str) -> Optional[Image.Image]: """加载并根据需要旋转图像""" if not image_path or not Path(image_path).exists(): return None # 检查缓存 rotation_angle = self.get_rotation_angle() cache_key = f"{image_path}_{rotation_angle}" if cache_key in self._rotated_image_cache: return self._rotated_image_cache[cache_key] try: image = Image.open(image_path) # 如果需要旋转 if rotation_angle != 0: st.info(f"🔄 检测到文档旋转角度: {rotation_angle}°,正在自动旋转图像...") # 收集所有bbox坐标 all_bboxes = [] text_to_bbox_map = {} # 记录文本到bbox索引的映射 bbox_index = 0 for text, info_list in self.validator.text_bbox_mapping.items(): text_to_bbox_map[text] = [] for info in info_list: all_bboxes.append(info['bbox']) text_to_bbox_map[text].append(bbox_index) bbox_index += 1 # 旋转图像和坐标 rotated_image, rotated_bboxes = rotate_image_and_coordinates( image, rotation_angle, all_bboxes ) # 更新bbox映射 - 使用映射关系确保正确对应 for text, bbox_indices in text_to_bbox_map.items(): for i, bbox_idx in enumerate(bbox_indices): if bbox_idx < len(rotated_bboxes) and i < len(self.validator.text_bbox_mapping[text]): self.validator.text_bbox_mapping[text][i]['bbox'] = rotated_bboxes[bbox_idx] # 缓存结果 self._rotated_image_cache[cache_key] = rotated_image return rotated_image else: # 无需旋转,直接缓存原图 self._rotated_image_cache[cache_key] = image return image except Exception as e: st.error(f"❌ 图像加载失败: {e}") return None def render_content_section(self, layout_type: str = "standard"): """渲染内容区域 - 统一方法""" st.header("📄 OCR识别内容") # 显示旋转信息 # rotation_angle = self.get_rotation_angle() # if rotation_angle != 0: # st.info(f"📐 文档旋转角度: {rotation_angle}°") # 文本选择器 if self.validator.text_bbox_mapping: text_options = ["请选择文本..."] + list(self.validator.text_bbox_mapping.keys()) selected_index = st.selectbox( "选择要校验的文本", range(len(text_options)), format_func=lambda x: text_options[x][:50] + "..." if len(text_options[x]) > 50 else text_options[x], key=f"{layout_type}_text_selector" ) if selected_index > 0: st.session_state.selected_text = text_options[selected_index] else: st.warning("没有找到可点击的文本") def render_md_content(self, layout_type: str): """渲染Markdown内容 - 统一方法""" if not self.validator.md_content: return None, None # 搜索功能 search_term = st.text_input( "🔍 搜索文本内容", placeholder="输入关键词搜索...", key=f"{layout_type}_search" ) display_content = self.validator.md_content if search_term: lines = display_content.split('\n') filtered_lines = [line for line in lines if search_term.lower() in line.lower()] display_content = '\n'.join(filtered_lines) if filtered_lines: st.success(f"找到 {len(filtered_lines)} 行包含 '{search_term}'") else: st.warning(f"未找到包含 '{search_term}' 的内容") # 渲染方式选择 render_mode = st.radio( "选择渲染方式", ["HTML渲染", "Markdown渲染", "DataFrame表格", "原始文本"], horizontal=True, key=f"{layout_type}_render_mode" ) return display_content, render_mode def render_content_by_mode(self, content: str, render_mode: str, font_size: int, layout_type: str): """根据渲染模式显示内容 - 统一方法""" if content is None or render_mode is None: return if render_mode == "HTML渲染": content_style = f""" """ st.markdown(content_style, unsafe_allow_html=True) st.markdown(f'
{content}
', unsafe_allow_html=True) elif render_mode == "Markdown渲染": converted_content = convert_html_table_to_markdown(content) content_style = f""" """ st.markdown(content_style, unsafe_allow_html=True) st.markdown(f'
{converted_content}
', unsafe_allow_html=True) elif render_mode == "DataFrame表格": if ' 30 else text_options[x], key="compact_quick_text_selector" # 使用不同的key ) if selected_index > 0: st.session_state.selected_text = text_options[selected_index] # 自定义CSS样式 st.markdown(f""" """, unsafe_allow_html=True) # 处理并显示OCR内容 if self.validator.md_content: # 高亮可点击文本 highlighted_content = self.validator.md_content for text in self.validator.text_bbox_mapping.keys(): if len(text) > 2: # 避免高亮过短的文本 css_class = "highlight-text selected-highlight" if text == st.session_state.selected_text else "highlight-text" highlighted_content = highlighted_content.replace( text, f'{text}' ) st.markdown( f'
{highlighted_content}
', unsafe_allow_html=True ) with right_col: # 修复的对齐图片显示 self.create_aligned_image_display(zoom_level, "compact") def create_aligned_image_display(self, zoom_level: float = 1.0, layout_type: str = "aligned"): """创建与左侧对齐的图片显示""" # 精确对齐CSS st.markdown(f""" """, unsafe_allow_html=True) st.markdown(f'
', unsafe_allow_html=True) st.header("🖼️ 原图标注") # 图片缩放控制 col1, col2 = st.columns(2) with col1: current_zoom = st.slider("图片缩放", 0.3, 2.0, zoom_level, 0.1, key=f"{layout_type}_zoom_level") with col2: show_all_boxes = st.checkbox("显示所有框", value=False, key=f"{layout_type}_show_all_boxes") # 使用新的图像加载方法 image = self.load_and_rotate_image(self.validator.image_path) if image: try: # 根据缩放级别调整图片大小 new_width = int(image.width * current_zoom) new_height = int(image.height * current_zoom) resized_image = image.resize((new_width, new_height), Image.Resampling.LANCZOS) # 计算选中的bbox - 注意bbox已经是旋转后的坐标 selected_bbox = None if st.session_state.selected_text and st.session_state.selected_text in self.validator.text_bbox_mapping: info = self.validator.text_bbox_mapping[st.session_state.selected_text][0] # bbox已经是旋转后的坐标,只需要应用缩放 bbox = info['bbox'] selected_bbox = [int(coord * current_zoom) for coord in bbox] # 创建交互式图片 fig = self.create_resized_interactive_plot(resized_image, selected_bbox, current_zoom, show_all_boxes) st.plotly_chart(fig, use_container_width=True, key=f"{layout_type}_plot") # 显示选中文本的详细信息 if st.session_state.selected_text and st.session_state.selected_text in self.validator.text_bbox_mapping: st.subheader("📍 选中文本详情") info = self.validator.text_bbox_mapping[st.session_state.selected_text][0] bbox = info['bbox'] info_col1, info_col2 = st.columns(2) with info_col1: st.write(f"**文本内容:** {st.session_state.selected_text[:30]}...") st.write(f"**类别:** {info['category']}") # 显示旋转信息 rotation_angle = self.get_rotation_angle() if rotation_angle != 0: st.write(f"**旋转角度:** {rotation_angle}°") with info_col2: st.write(f"**位置:** [{', '.join(map(str, bbox))}]") if len(bbox) >= 4: st.write(f"**大小:** {bbox[2] - bbox[0]} x {bbox[3] - bbox[1]} px") # 错误标记功能 col1, col2 = st.columns(2) with col1: if st.button("❌ 标记为错误", key=f"{layout_type}_mark_error"): st.session_state.marked_errors.add(st.session_state.selected_text) st.rerun() with col2: if st.button("✅ 取消错误标记", key=f"{layout_type}_unmark_error"): st.session_state.marked_errors.discard(st.session_state.selected_text) st.rerun() except Exception as e: st.error(f"❌ 图片处理失败: {e}") st.error(f"详细错误: {str(e)}") else: st.error("未找到对应的图片文件") if self.validator.image_path: st.write(f"期望路径: {self.validator.image_path}") st.markdown('
', unsafe_allow_html=True) def create_resized_interactive_plot(self, image: Image.Image, selected_bbox: Optional[List[int]], zoom_level: float, show_all_boxes: bool) -> go.Figure: """创建可调整大小的交互式图片 - 修复图像显示和bbox对齐问题""" fig = go.Figure() # 添加图片 - 修正图像定位,确保与工具栏距离一致 fig.add_layout_image( dict( source=image, xref="x", yref="y", x=0, y=image.height * zoom_level, # 修正:图片左上角位置 sizex=image.width * zoom_level, sizey=image.height * zoom_level, sizing="stretch", opacity=1.0, layer="below" ) ) # 显示所有bbox if show_all_boxes: for text, info_list in self.validator.text_bbox_mapping.items(): for info in info_list: bbox = info['bbox'] if len(bbox) >= 4: # bbox已经是旋转后的坐标,需要应用缩放并转换坐标系 x1, y1, x2, y2 = bbox[:4] # 应用缩放 scaled_x1 = x1 * zoom_level scaled_y1 = y1 * zoom_level scaled_x2 = x2 * zoom_level scaled_y2 = y2 * zoom_level # 转换为plotly坐标系(原点在左下角) plot_x1 = scaled_x1 plot_y1 = (image.height * zoom_level) - scaled_y2 # 翻转Y坐标 plot_x2 = scaled_x2 plot_y2 = (image.height * zoom_level) - scaled_y1 # 翻转Y坐标 color = "rgba(0, 100, 200, 0.2)" if text in self.validator.marked_errors: color = "rgba(255, 0, 0, 0.3)" fig.add_shape( type="rect", x0=plot_x1, y0=plot_y1, x1=plot_x2, y1=plot_y2, line=dict(color=color.replace('0.2', '0.8').replace('0.3', '1.0'), width=1), fillcolor=color, ) # 高亮显示选中的bbox if selected_bbox and len(selected_bbox) >= 4: x1, y1, x2, y2 = selected_bbox[:4] # 转换为plotly坐标系(selected_bbox已经是缩放后的坐标) plot_x1 = x1 plot_y1 = (image.height * zoom_level) - y2 # 翻转Y坐标 plot_x2 = x2 plot_y2 = (image.height * zoom_level) - y1 # 翻转Y坐标 fig.add_shape( type="rect", x0=plot_x1, y0=plot_y1, x1=plot_x2, y1=plot_y2, line=dict(color="red", width=3), fillcolor="rgba(255, 0, 0, 0.3)", ) # 计算合适的显示尺寸 aspect_ratio = image.width / image.height display_height = min(800, max(400, image.height // 2)) display_width = int(display_height * aspect_ratio) # 设置布局 - 确保图像完全可见,使用缩放后的尺寸 fig.update_layout( width=display_width, height=display_height, margin=dict(l=0, r=0, t=0, b=0), showlegend=False, plot_bgcolor='white', dragmode="pan", # X轴设置 - 使用缩放后的图像尺寸 xaxis=dict( visible=False, range=[0, image.width * zoom_level], constrain="domain", fixedrange=False, autorange=False ), # Y轴设置 - plotly坐标系(原点在左下角) yaxis=dict( visible=False, range=[0, image.height * zoom_level], constrain="domain", scaleanchor="x", scaleratio=1, fixedrange=False, autorange=False ) ) return fig