还在手动录入身份证信息？一行行输姓名、身份证号、地址，不仅费眼还容易错，每天录上百条能累到手腕疼！在网上找到了这款吾爱大神写的工具，从 V1.0.3 开源到现在更新到 V1.0.5，核心解决 "识别慢、字段错、依赖显卡 / 网络" 的问题，纯 CPU 运行、离线可用，Win10/11 直接跑，现在分享给有同样需求的同行！（文末附下载地址）
版本迭代：每一次更新都直击使用痛点
从最初版本到 V1.0.5，工具一直在优化核心体验，尤其解决了大家反馈的关键问题：
V1.0.3（开源版）：基础功能拉满，效率先提档
- 核心升级：优化启动速度（打开软件从 10 秒缩到 3 秒），新增识别字段（补全民族、出生日期、有效期等易漏字段）；
- 实用功能：图片重命名（可选配置，识别后自动按 "姓名 - 身份证号" 命名图片）、Excel 导出可选（仅填图片路径 / 直接嵌入图片）；
- 代码开源：核心逻辑已放出，有开发能力的伙伴可下载源码自定义修改（附件自取），适配自家业务场景。
V1.0.4：解决使用体验 "小麻烦"
- 模型本地化：模型随程序分发，初次启动不用再下载几百 M 的模型包，开箱即用；
- 终端弹窗修复：彻底解决启动后黑框终端关不掉的问题，界面更清爽，不用手动杀进程。
V1.0.5：精准度再升级
- 字段匹配优化：重构识别字段的匹配算法，姓名、身份证号、地址等核心字段识别准确率提升 15%，尤其解决了 "地址含生僻字识别错""身份证号末尾 X 识别成 x" 的问题；
- 无额外依赖变更：依旧纯 CPU 运行，不用折腾显卡配置，老电脑也能扛。
工具已经打包好：
链接：https://pan.quark.cn/s/423f3761c712
源码分享
python
# 在gui.py文件中添加以下代码
import re<p></p>
<p>from PySide6.QtCore import QThread, Signal
import traceback
from openpyxl.workbook import Workbook
from openpyxl.drawing.image import Image as XLImage
from openpyxl.styles import Alignment</p>
<p>import cv2
import numpy as np
from PIL import Image</p>
<p>'''
ocr实际开始工作的线程
需要将前边加载好的模型传递过来
'''
class OCRWorker(QThread):</p>
<h2 id="53093959_定义信号，用于通知主线程处理进度和结果">定义信号，用于通知主线程处理进度和结果</h2>
<pre><code data-highlighted="yes" class="hljs language-python">progress_updated = Signal(<span class="hljs-built_in">int</span>, <span class="hljs-built_in">int</span>) <span class="hljs-comment"># 当前进度，总数量</span>
finished_signal = Signal() <span class="hljs-comment"># 处理完成信号</span>
error_occurred = Signal(<span class="hljs-built_in">str</span>) <span class="hljs-comment"># 错误信息信号</span>
<span class="hljs-keyword">def</span> <span class="hljs-title function_">__init__</span>(<span class="hljs-params">self, file_paths, export_options, ocr</span>):
<span class="hljs-built_in">super</span>().__init__()
<span class="hljs-variable language_">self</span>.file_paths = file_paths
<span class="hljs-variable language_">self</span>.export_options = export_options
<span class="hljs-variable language_">self</span>.ocr = ocr
<span class="hljs-variable language_">self</span>._should_terminate = <span class="hljs-literal">False</span> <span class="hljs-comment"># 添加终止标志</span>
<span class="hljs-keyword">def</span> <span class="hljs-title function_">run</span>(<span class="hljs-params">self</span>):
<span class="hljs-keyword">try</span>:
<span class="hljs-comment"># 处理所有文件</span>
<span class="hljs-variable language_">self</span>.process_files(<span class="hljs-variable language_">self</span>.file_paths)
<span class="hljs-keyword">except</span> Exception <span class="hljs-keyword">as</span> e:
error_msg = <span class="hljs-string">f"处理过程中发生错误: <span class="hljs-subst">{<span class="hljs-built_in">str</span>(e)}</span>\n<span class="hljs-subst">{traceback.format_exc()}</span>"</span>
<span class="hljs-variable language_">self</span>.error_occurred.emit(error_msg)
<span class="hljs-keyword">def</span> <span class="hljs-title function_">process_files</span>(<span class="hljs-params">self, file_paths</span>):
wb = Workbook()
ws = wb.active
ws.append([<span class="hljs-string">"图片"</span>, <span class="hljs-string">"姓名"</span>, <span class="hljs-string">"性别"</span>, <span class="hljs-string">"民族"</span>, <span class="hljs-string">"出生日期"</span>, <span class="hljs-string">"住址"</span>, <span class="hljs-string">"身份证号"</span>, <span class="hljs-string">"有效期限"</span>])
row_idx = <span class="hljs-number">2</span>
total_files = <span class="hljs-built_in">len</span>(<span class="hljs-variable language_">self</span>.file_paths)
processed_count = <span class="hljs-number">0</span>
<span class="hljs-keyword">for</span> i, path <span class="hljs-keyword">in</span> <span class="hljs-built_in">enumerate</span>(file_paths):
<span class="hljs-comment"># 检查是否收到终止请求</span>
<span class="hljs-keyword">if</span> <span class="hljs-variable language_">self</span>._should_terminate:
<span class="hljs-built_in">print</span>(<span class="hljs-string">"收到终止请求，正在保存已处理的数据..."</span>)
<span class="hljs-keyword">break</span>
<span class="hljs-comment"># 发送进度更新信号</span>
<span class="hljs-variable language_">self</span>.progress_updated.emit(i + <span class="hljs-number">1</span>, total_files)
info = <span class="hljs-variable language_">self</span>.extract_info_from_image(path)
<span class="hljs-comment"># 检查线程是否被中断</span>
<span class="hljs-keyword">if</span> <span class="hljs-variable language_">self</span>.isInterruptionRequested():
<span class="hljs-keyword">break</span>
<span class="hljs-keyword">if</span> info:
ws.cell(row=row_idx, column=<span class="hljs-number">2</span>, value=info[<span class="hljs-string">"姓名"</span>])
ws.cell(row=row_idx, column=<span class="hljs-number">3</span>, value=info[<span class="hljs-string">"性别"</span>])
ws.cell(row=row_idx, column=<span class="hljs-number">4</span>, value=info[<span class="hljs-string">"民族"</span>])
ws.cell(row=row_idx, column=<span class="hljs-number">5</span>, value=info[<span class="hljs-string">"出生日期"</span>])
ws.cell(row=row_idx, column=<span class="hljs-number">6</span>, value=info[<span class="hljs-string">"住址"</span>])
ws.cell(row=row_idx, column=<span class="hljs-number">7</span>, value=info[<span class="hljs-string">"身份证号"</span>])
ws.cell(row=row_idx, column=<span class="hljs-number">8</span>, value=info[<span class="hljs-string">"有效期限"</span>])
<span class="hljs-comment"># 根据导出选项决定如何处理图片</span>
export_option = <span class="hljs-variable language_">self</span>.export_options.get(<span class="hljs-string">"export_option"</span>, <span class="hljs-string">"image_path"</span>)
<span class="hljs-comment"># 处理重命名（如果配置了重命名选项）</span>
<span class="hljs-keyword">if</span> <span class="hljs-variable language_">self</span>.should_rename_file(info):
new_path = <span class="hljs-variable language_">self</span>.rename_file(path, info)
<span class="hljs-comment"># 更新图片路径为重命名后的路径</span>
<span class="hljs-keyword">if</span> export_option == <span class="hljs-string">"image_path"</span>:
ws.cell(row=row_idx, column=<span class="hljs-number">1</span>, value=new_path)
<span class="hljs-comment"># 更新返回数据中的图片路径</span>
info[<span class="hljs-string">"图片路径"</span>] = new_path
<span class="hljs-keyword">if</span> export_option == <span class="hljs-string">"image_file"</span>:
<span class="hljs-comment"># 直接嵌入图片文件</span>
<span class="hljs-keyword">try</span>:
img = XLImage(info[<span class="hljs-string">"图片路径"</span>])
img.width = <span class="hljs-number">500</span>
img.height = <span class="hljs-number">300</span>
ws.row_dimensions[zxsq-anti-bbcode-row_idx].height = img.height
ws.add_image(img, <span class="hljs-string">f"A<span class="hljs-subst">{row_idx}</span>"</span>)
ws.column_dimensions[<span class="hljs-string">'A'</span>].width = img.width * <span class="hljs-number">0.14</span>
<span class="hljs-keyword">except</span> Exception <span class="hljs-keyword">as</span> e:
<span class="hljs-built_in">print</span>(<span class="hljs-string">f"无法插入图片 <span class="hljs-subst">{path}</span>: <span class="hljs-subst">{e}</span>"</span>)
<span class="hljs-keyword">else</span> :
<span class="hljs-comment"># 仅保存图片路径</span>
ws.cell(row=row_idx, column=<span class="hljs-number">1</span>, value=info[<span class="hljs-string">"图片路径"</span>])
<span class="hljs-keyword">for</span> col <span class="hljs-keyword">in</span> <span class="hljs-built_in">range</span>(<span class="hljs-number">1</span>, <span class="hljs-number">9</span>):
cell = ws.cell(row=row_idx, column=col)
cell.alignment = Alignment(horizontal=<span class="hljs-string">'center'</span>, vertical=<span class="hljs-string">'center'</span>)
row_idx += <span class="hljs-number">1</span>
processed_count += <span class="hljs-number">1</span>
<span class="hljs-keyword">for</span> col <span class="hljs-keyword">in</span> <span class="hljs-built_in">range</span>(<span class="hljs-number">1</span>, <span class="hljs-number">9</span>):
header_cell = ws.cell(row=<span class="hljs-number">1</span>, column=col)
header_cell.alignment = Alignment(horizontal=<span class="hljs-string">'center'</span>, vertical=<span class="hljs-string">'center'</span>)
output_path = <span class="hljs-string">"身份证识别结果.xlsx"</span>
wb.save(output_path)
<span class="hljs-keyword">if</span> <span class="hljs-variable language_">self</span>._should_terminate:
<span class="hljs-built_in">print</span>(<span class="hljs-string">f"处理已终止，已完成 <span class="hljs-subst">{processed_count}</span>/<span class="hljs-subst">{total_files}</span> 个文件，结果已保存到 <span class="hljs-subst">{output_path}</span>"</span>)
<span class="hljs-keyword">else</span>:
<span class="hljs-built_in">print</span>(<span class="hljs-string">f"处理完成，共处理 <span class="hljs-subst">{processed_count}</span> 个文件，结果已保存到 <span class="hljs-subst">{output_path}</span>"</span>)
<span class="hljs-comment"># 发送完成信号</span>
<span class="hljs-variable language_">self</span>.finished_signal.emit()
<span class="hljs-keyword">def</span> <span class="hljs-title function_">extract_info_from_image</span>(<span class="hljs-params">self, image_path</span>):
<span class="hljs-string">"""从图片中提取信息（优化版文本处理）"""</span>
<span class="hljs-keyword">try</span>:
<span class="hljs-comment"># 检查文件是否存在和可读</span>
<span class="hljs-comment"># import os</span>
<span class="hljs-comment"># if not os.path.exists(image_path):</span>
<span class="hljs-comment"># raise FileNotFoundError(f"图片文件不存在: {image_path}")</span>
<span class="hljs-comment">#</span>
<span class="hljs-comment"># if not os.access(image_path, os.R_OK):</span>
<span class="hljs-comment"># raise PermissionError(f"没有权限读取图片文件: {image_path}")</span>
<span class="hljs-comment"># # 检查是否需要预处理身份证图片</span>
<span class="hljs-comment"># if self.export_options.get("preprocess_id_card", True):</span>
<span class="hljs-comment"># processed_image_path = self.preprocess_id_card_image(image_path)</span>
<span class="hljs-comment"># else:</span>
<span class="hljs-comment"># processed_image_path = image_path</span>
<span class="hljs-comment">#</span>
<span class="hljs-comment"># result = self.ocr.ocr(processed_image_path, cls=True)</span>
result = <span class="hljs-variable language_">self</span>.ocr.ocr(image_path, cls=<span class="hljs-literal">True</span>)
<span class="hljs-comment"># 1. 先整体拼接所有文本</span>
all_text = <span class="hljs-string">""</span>
<span class="hljs-keyword">for</span> res <span class="hljs-keyword">in</span> result:
<span class="hljs-keyword">for</span> line <span class="hljs-keyword">in</span> res:
text = line[zxsq-anti-bbcode-<span class="hljs-number">1</span>][zxsq-anti-bbcode-<span class="hljs-number">0</span>]
<span class="hljs-keyword">if</span> text:
all_text += text
<span class="hljs-comment"># 2. 去除"中华人民共和国居民身份证"标题</span>
all_text = re.sub(<span class="hljs-string">r'中华人民共和国居民身份证'</span>, <span class="hljs-string">''</span>, all_text)
<span class="hljs-comment"># 3. 去除所有空格和特殊空白字符</span>
all_text = re.sub(<span class="hljs-string">r'\s+'</span>, <span class="hljs-string">''</span>, all_text)
<span class="hljs-comment"># 4. 在关键字段前添加换行符</span>
keywords = [<span class="hljs-string">'姓名'</span>, <span class="hljs-string">'性别'</span>, <span class="hljs-string">'民族'</span>, <span class="hljs-string">'出生'</span>, <span class="hljs-string">'住址'</span>, <span class="hljs-string">'公民身份号码'</span>, <span class="hljs-string">'签发机关'</span>, <span class="hljs-string">'有效期限'</span>]
<span class="hljs-keyword">for</span> keyword <span class="hljs-keyword">in</span> keywords:
all_text = re.sub(<span class="hljs-string">f'(<span class="hljs-subst">{keyword}</span>)'</span>, <span class="hljs-string">r'\n\1'</span>, all_text)
<span class="hljs-built_in">print</span>(<span class="hljs-string">f"处理后的文本: <span class="hljs-subst">{all_text}</span>"</span>)
<span class="hljs-comment"># 初始化提取结果</span>
name = gender = nation = birth = address = id_number = expire = <span class="hljs-string">""</span>
<span class="hljs-comment"># 提取各字段信息</span>
<span class="hljs-comment"># 提取身份证号</span>
<span class="hljs-comment"># 直接匹配17位数字+1位校验码（数字或X）</span>
id_match = re.search(<span class="hljs-string">r'[\d]{17}[\dXx]'</span>, all_text)
<span class="hljs-keyword">if</span> id_match:
id_number = id_match.group().strip()
<span class="hljs-comment"># 移除身份证号码干扰</span>
all_text = all_text.replace(id_match.group(), <span class="hljs-string">''</span>)
<span class="hljs-comment"># 提取姓名</span>
name_match = re.search(<span class="hljs-string">r'姓名(.+?)(?=\n|$)'</span>, all_text)
<span class="hljs-keyword">if</span> name_match:
name = name_match.group(<span class="hljs-number">1</span>).strip()
<span class="hljs-comment"># 提取性别</span>
gender_match = re.search(<span class="hljs-string">r'性别(男|女)'</span>, all_text)
<span class="hljs-keyword">if</span> gender_match:
gender = gender_match.group(<span class="hljs-number">1</span>).strip()
<span class="hljs-comment"># 提取民族</span>
nation_match = re.search(<span class="hljs-string">r'民族(.+?)(?=\n|$)'</span>, all_text)
<span class="hljs-keyword">if</span> nation_match:
nation = nation_match.group(<span class="hljs-number">1</span>).strip()
<span class="hljs-comment"># 提取出生日期</span>
birth_match = re.search(<span class="hljs-string">r'出生(.+?)(?=\n|$)'</span>, all_text)
<span class="hljs-keyword">if</span> birth_match:
birth = birth_match.group(<span class="hljs-number">1</span>).strip()
<span class="hljs-comment"># 提取住址</span>
address_match = re.search(<span class="hljs-string">r'住址(.+?)(?=\n|$)'</span>, all_text)
<span class="hljs-keyword">if</span> address_match:
address = address_match.group(<span class="hljs-number">1</span>).strip()
<span class="hljs-comment"># 提取有效期限</span>
expire_match = re.search(<span class="hljs-string">r'有效期限(.+?)(?=\n|$)'</span>, all_text)
<span class="hljs-keyword">if</span> expire_match:
expire = expire_match.group(<span class="hljs-number">1</span>).strip()
data = {
<span class="hljs-string">"姓名"</span>: name,
<span class="hljs-string">"性别"</span>: gender,
<span class="hljs-string">"民族"</span>: nation,
<span class="hljs-string">"出生日期"</span>: birth,
<span class="hljs-string">"住址"</span>: address,
<span class="hljs-string">"身份证号"</span>: id_number,
<span class="hljs-string">"有效期限"</span>: expire,
<span class="hljs-string">"图片路径"</span>: image_path
}
<span class="hljs-built_in">print</span>(<span class="hljs-string">f"data == <span class="hljs-subst">{data}</span>"</span>)
<span class="hljs-keyword">return</span> data
<span class="hljs-keyword">except</span> Exception <span class="hljs-keyword">as</span> e:
<span class="hljs-built_in">print</span>(<span class="hljs-string">f"处理 <span class="hljs-subst">{image_path}</span> 失败: <span class="hljs-subst">{e}</span>"</span>)
<span class="hljs-keyword">return</span> <span class="hljs-literal">None</span>
<span class="hljs-keyword">def</span> <span class="hljs-title function_">should_rename_file</span>(<span class="hljs-params">self, info</span>):
<span class="hljs-string">"""检查是否需要重命名文件"""</span>
rename_options = <span class="hljs-variable language_">self</span>.export_options.get(<span class="hljs-string">"rename_options"</span>, [])
<span class="hljs-keyword">return</span> <span class="hljs-built_in">len</span>(rename_options) > <span class="hljs-number">0</span>
<span class="hljs-keyword">def</span> <span class="hljs-title function_">rename_file</span>(<span class="hljs-params">self, original_path, info</span>):
<span class="hljs-string">"""根据配置重命名文件"""</span>
<span class="hljs-keyword">if</span> <span class="hljs-keyword">not</span> <span class="hljs-variable language_">self</span>.should_rename_file(info):
<span class="hljs-keyword">return</span> original_path
rename_options = <span class="hljs-variable language_">self</span>.export_options.get(<span class="hljs-string">"rename_options"</span>, [])
separator = <span class="hljs-variable language_">self</span>.export_options.get(<span class="hljs-string">"separator"</span>, <span class="hljs-string">"_"</span>)
<span class="hljs-comment"># 构建新的文件名部分</span>
name_parts = []
<span class="hljs-keyword">for</span> option <span class="hljs-keyword">in</span> rename_options:
<span class="hljs-keyword">if</span> option == <span class="hljs-string">"name"</span> <span class="hljs-keyword">and</span> info.get(<span class="hljs-string">"姓名"</span>):
name_parts.append(info[<span class="hljs-string">"姓名"</span>])
<span class="hljs-keyword">elif</span> option == <span class="hljs-string">"id"</span> <span class="hljs-keyword">and</span> info.get(<span class="hljs-string">"身份证号"</span>):
name_parts.append(info[<span class="hljs-string">"身份证号"</span>])
<span class="hljs-keyword">elif</span> option == <span class="hljs-string">"nation"</span> <span class="hljs-keyword">and</span> info.get(<span class="hljs-string">"民族"</span>):
name_parts.append(info[<span class="hljs-string">"民族"</span>])
<span class="hljs-keyword">elif</span> option == <span class="hljs-string">"sex"</span> <span class="hljs-keyword">and</span> info.get(<span class="hljs-string">"性别"</span>):
name_parts.append(info[<span class="hljs-string">"性别"</span>])
<span class="hljs-keyword">elif</span> option == <span class="hljs-string">"address"</span> <span class="hljs-keyword">and</span> info.get(<span class="hljs-string">"住址"</span>):
name_parts.append(info[<span class="hljs-string">"住址"</span>])
<span class="hljs-keyword">if</span> <span class="hljs-keyword">not</span> name_parts:
<span class="hljs-keyword">return</span> original_path
<span class="hljs-comment"># 构造新文件名</span>
new_name = separator.join(name_parts)
<span class="hljs-comment"># 保持原始文件扩展名</span>
<span class="hljs-keyword">import</span> os
dir_name = os.path.dirname(original_path)
file_ext = os.path.splitext(original_path)[zxsq-anti-bbcode-<span class="hljs-number">1</span>]
new_path = os.path.join(dir_name, new_name + file_ext)
<span class="hljs-comment"># 重命名文件</span>
<span class="hljs-keyword">try</span>:
os.rename(original_path, new_path)
<span class="hljs-keyword">return</span> new_path
<span class="hljs-keyword">except</span> Exception <span class="hljs-keyword">as</span> e:
<span class="hljs-built_in">print</span>(<span class="hljs-string">f"重命名文件失败 <span class="hljs-subst">{original_path}</span> -> <span class="hljs-subst">{new_path}</span>: <span class="hljs-subst">{e}</span>"</span>)
<span class="hljs-keyword">return</span> original_path
<span class="hljs-comment"># 图片灰度处理, 处理成扫描件, 下面还没写好 不要用</span>
<span class="hljs-comment"># def preprocess_id_card_image(self, image_path):</span>
<span class="hljs-comment"># """对身份证图片进行校正、裁剪并转换为黑白扫描件"""</span>
<span class="hljs-comment"># try:</span>
<span class="hljs-comment"># # 读取图片</span>
<span class="hljs-comment"># img = cv2.imread(image_path)</span>
<span class="hljs-comment"># if img is None:</span>
<span class="hljs-comment"># return image_path</span>
<span class="hljs-comment">#</span>
<span class="hljs-comment"># # 1. 转换为灰度图</span>
<span class="hljs-comment"># gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)</span>
<span class="hljs-comment">#</span>
<span class="hljs-comment"># # 2. 使用中值滤波代替高斯模糊</span>
<span class="hljs-comment"># denoised = cv2.medianBlur(gray, 3)</span>
<span class="hljs-comment">#</span>
<span class="hljs-comment"># # 3. 使用自适应阈值</span>
<span class="hljs-comment"># binary = cv2.adaptiveThreshold(</span>
<span class="hljs-comment"># denoised, 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C,</span>
<span class="hljs-comment"># cv2.THRESH_BINARY, 15, 3</span>
<span class="hljs-comment"># )</span>
<span class="hljs-comment">#</span>
<span class="hljs-comment"># # 4. 可选：轻微平滑处理</span>
<span class="hljs-comment"># smoothed = cv2.medianBlur(binary, 1)</span>
<span class="hljs-comment">#</span>
<span class="hljs-comment"># # 5. 保存处理后的图片</span>
<span class="hljs-comment"># import os</span>
<span class="hljs-comment"># dir_name = os.path.dirname(image_path)</span>
<span class="hljs-comment"># file_name = os.path.splitext(os.path.basename(image_path))[zxsq-anti-bbcode-0]</span>
<span class="hljs-comment"># file_ext = os.path.splitext(image_path)[zxsq-anti-bbcode-1]</span>
<span class="hljs-comment"># processed_path = os.path.join(dir_name, f"{file_name}_processed{file_ext}")</span>
<span class="hljs-comment">#</span>
<span class="hljs-comment"># cv2.imwrite(processed_path, smoothed)</span>
<span class="hljs-comment">#</span>
<span class="hljs-comment"># return processed_path</span>
<span class="hljs-comment"># except Exception as e:</span>
<span class="hljs-comment"># print(f"身份证图片预处理失败 {image_path}: {e}")</span>
<span class="hljs-comment"># return image_path</span>
<span class="hljs-comment">#</span>
<span class="hljs-comment"># def order_points(self, pts):</span>
<span class="hljs-comment"># """对四个点进行排序：左上、右上、右下、左下"""</span>
<span class="hljs-comment"># rect = np.zeros((4, 2), dtype="float32")</span>
<span class="hljs-comment">#</span>
<span class="hljs-comment"># # 计算坐标和</span>
<span class="hljs-comment"># s = pts.sum(axis=1)</span>
<span class="hljs-comment"># rect[zxsq-anti-bbcode-0] = pts[np.argmin(s)] # 左上角点（坐标和最小）</span>
<span class="hljs-comment"># rect[zxsq-anti-bbcode-2] = pts[np.argmax(s)] # 右下角点（坐标和最大）</span>
<span class="hljs-comment">#</span>
<span class="hljs-comment"># # 计算坐标差</span>
<span class="hljs-comment"># diff = np.diff(pts, axis=1)</span>
<span class="hljs-comment"># rect[zxsq-anti-bbcode-1] = pts[np.argmin(diff)] # 右上角点（坐标差最小）</span>
<span class="hljs-comment"># rect[zxsq-anti-bbcode-3] = pts[np.argmax(diff)] # 左下角点（坐标差最大）</span>
<span class="hljs-comment">#</span>
<span class="hljs-comment"># return rect</span>
<span class="hljs-comment">#</span>
<span class="hljs-comment"># def four_point_transform(self, image, pts):</span>
<span class="hljs-comment"># """四点透视变换"""</span>
<span class="hljs-comment"># # 获取排序后的坐标</span>
<span class="hljs-comment"># rect = self.order_points(pts)</span>
<span class="hljs-comment"># (tl, tr, br, bl) = rect</span>
<span class="hljs-comment">#</span>
<span class="hljs-comment"># # 计算新图像的宽度和高度</span>
<span class="hljs-comment"># width_a = np.sqrt(((br[zxsq-anti-bbcode-0] - bl[zxsq-anti-bbcode-0]) ** 2) + ((br[zxsq-anti-bbcode-1] - bl[zxsq-anti-bbcode-1]) ** 2))</span>
<span class="hljs-comment"># width_b = np.sqrt(((tr[zxsq-anti-bbcode-0] - tl[zxsq-anti-bbcode-0]) ** 2) + ((tr[zxsq-anti-bbcode-1] - tl[zxsq-anti-bbcode-1]) ** 2))</span>
<span class="hljs-comment"># max_width = max(int(width_a), int(width_b))</span>
<span class="hljs-comment">#</span>
<span class="hljs-comment"># height_a = np.sqrt(((tr[zxsq-anti-bbcode-0] - br[zxsq-anti-bbcode-0]) ** 2) + ((tr[zxsq-anti-bbcode-1] - br[zxsq-anti-bbcode-1]) ** 2))</span>
<span class="hljs-comment"># height_b = np.sqrt(((tl[zxsq-anti-bbcode-0] - bl[zxsq-anti-bbcode-0]) ** 2) + ((tl[zxsq-anti-bbcode-1] - bl[zxsq-anti-bbcode-1]) ** 2))</span>
<span class="hljs-comment"># max_height = max(int(height_a), int(height_b))</span>
<span class="hljs-comment">#</span>
<span class="hljs-comment"># # 目标点</span>
<span class="hljs-comment"># dst = np.array([</span>
<span class="hljs-comment"># [0, 0],</span>
<span class="hljs-comment"># [max_width - 1, 0],</span>
<span class="hljs-comment"># [max_width - 1, max_height - 1],</span>
<span class="hljs-comment"># [0, max_height - 1]], dtype="float32")</span>
<span class="hljs-comment">#</span>
<span class="hljs-comment"># # 计算透视变换矩阵并应用</span>
<span class="hljs-comment"># M = cv2.getPerspectiveTransform(rect, dst)</span>
<span class="hljs-comment"># warped = cv2.warpPerspective(image, M, (max_width, max_height))</span>
<span class="hljs-comment">#</span>
<span class="hljs-comment"># return warped</span>
<span class="hljs-comment"># 中断处理, 此处不要直接中断线程, 可能导致excel未能处理完毕线程就退出了</span>
<span class="hljs-comment"># 我们应该保证excel</span>
<span class="hljs-keyword">def</span> <span class="hljs-title function_">request_termination</span>(<span class="hljs-params">self</span>):
<span class="hljs-string">"""请求终止处理过程"""</span>
<span class="hljs-variable language_">self</span>._should_terminate = <span class="hljs-literal">True</span></code></pre>