基于Django与Tesseract OCR的模糊低对比度图像文本提取问题

项目场景 #

待处理图像特征：

• 模糊、低对比度
• 包含标题、日期、小号字体、加粗字体等多种文本块，部分文本模糊不清
• 图像尺寸约256×350像素，分辨率96 DPI

视图函数（views.py） #

import os
from django.shortcuts import render
from django.core.files.storage import FileSystemStorage
import pytesseract
import cv2
from PIL import Image
import numpy as np
import re

# 设置Tesseract命令路径
pytesseract.pytesseract.tesseract_cmd = '/usr/bin/tesseract'

def clean_text(text):
    # 清理常见OCR错误示例
    text = re.sub(r'\s+', ' ', text)  # 移除多余空白字符
    text = re.sub(r'[^\w\s,.!?;:()"\\\']', '', text)  # 保留常用标点符号
    return text.strip()

def home(request):
    return render(request, 'layout/index.html')

def preprocess_image(image_path):
    # 读取图像
    image = cv2.imread(image_path, cv2.IMREAD_COLOR)
    if image is None:
        raise ValueError(f"Error loading image: {image_path}")
    
    # 提取红色通道以提升对比度（可根据实际场景启用）
    red_channel = image[:, :, 2]

    # 转换为灰度图
    gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
    
    # 手动直方图拉伸提升对比度
    min_val = np.min(gray)
    max_val = np.max(gray)
    stretched_gray = ((gray - min_val) / (max_val - min_val) * 255).astype(np.uint8)
    
    # 应用锐化蒙版改善模糊文本
    blurred = cv2.GaussianBlur(stretched_gray, (9, 9), 10.0)
    unsharp_image = cv2.addWeighted(stretched_gray, 1.5, blurred, -0.5, 0)
    
    # 自适应阈值处理
    binary = cv2.adaptiveThreshold(unsharp_image, 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C, cv2.THRESH_BINARY, 11, 2)
    
    # 中值滤波去噪
    denoised = cv2.medianBlur(binary, 3)
    
    return denoised

def upload(request):
    if request.method == 'POST' and request.FILES.get('document'):
        document = request.FILES['document']
        fs = FileSystemStorage()
        filename = fs.save(document.name, document)
        uploaded_file_url = fs.url(filename)

        # 预处理上传的图像
        try:
            preprocessed_image = preprocess_image(fs.path(filename))
        except ValueError as e:
            return render(request, 'layout/index.html', {'error': str(e)})

        # 将处理后的图像转换为PIL格式
        preprocessed_image_pil = Image.fromarray(preprocessed_image)

        # Tesseract提取文本，自定义配置
        custom_config = r'--oem 3 --psm 12 --dpi 1500'
        text = pytesseract.image_to_string(preprocessed_image_pil, config=custom_config)

        # 清理提取的文本
        text = clean_text(text)

        context = {
            'uploaded_file_url': uploaded_file_url,
            'text': text,
        }
        return render(request, 'layout/index.html', context)
    return render(request, 'layout/index.html')

HTML模板（layout/index.html） #

<!DOCTYPE html>
<html>
<head>
    <title>文档文本提取</title>
</head>
<body>
    <h1>上传文档图像</h1>
    <form method="post" enctype="multipart/form-data" action="{% url 'upload' %}">
        {% csrf_token %}
        <input type="file" name="document">
        <button type="submit">上传并提取文本</button>
    </form>

    {% if uploaded_file_url %}
        <h2>已上传图像:</h2>
        <img src="{{ uploaded_file_url }}" alt="上传的文档图像">
        <h2>提取的文本:</h2>
        <pre>{{ text }}</pre>
    {% endif %}
</body>
</html>

URL配置（urls.py） #

from django.urls import path
from . import views

urlpatterns = [
    path('', views.home, name='home'),
    path('upload/', views.upload, name='upload'),
]

内容的提问来源于stack exchange，提问作者Vishal Upadhyay