{"id":1080326,"date":"2025-01-08T12:29:04","date_gmt":"2025-01-08T04:29:04","guid":{"rendered":"https:\/\/docs.pingcode.com\/ask\/ask-ask\/1080326.html"},"modified":"2025-01-08T12:29:08","modified_gmt":"2025-01-08T04:29:08","slug":"python%e5%a6%82%e4%bd%95%e8%bd%ac%e5%8c%96%e5%9b%be%e7%89%87%e4%b8%ad%e6%96%87%e5%ad%97-2","status":"publish","type":"post","link":"https:\/\/docs.pingcode.com\/ask\/1080326.html","title":{"rendered":"python\u5982\u4f55\u8f6c\u5316\u56fe\u7247\u4e2d\u6587\u5b57"},"content":{"rendered":"<p style=\"text-align:center;\" ><img decoding=\"async\" src=\"https:\/\/cdn-kb.worktile.com\/kb\/wp-content\/uploads\/2024\/04\/24182922\/1d47de92-b004-4285-ba4f-47841aacbbcc.webp\" alt=\"python\u5982\u4f55\u8f6c\u5316\u56fe\u7247\u4e2d\u6587\u5b57\" \/><\/p>\n<p><p> <strong>Python\u8f6c\u5316\u56fe\u7247\u4e2d\u6587\u5b57\u7684\u65b9\u6cd5\u6709\u591a\u79cd\uff0c\u5e38\u89c1\u7684\u6709\u4f7f\u7528Tesseract OCR\u3001Pytesseract\u5e93\u3001Pillow\u5e93\u4ee5\u53caOpenCV\u5e93\u3002<\/strong> \u5176\u4e2d\uff0c\u4f7f\u7528Tesseract OCR\u662f\u4e00\u79cd\u8f83\u4e3a\u5e38\u89c1\u4e14\u6548\u679c\u4e0d\u9519\u7684\u65b9\u6cd5\uff0c\u5177\u4f53\u6b65\u9aa4\u5305\u62ec\uff1a\u5b89\u88c5Tesseract OCR\u3001\u5b89\u88c5Pytesseract\u5e93\u3001\u8bfb\u53d6\u56fe\u7247\u3001\u8c03\u7528Pytesseract\u8fdb\u884c\u6587\u5b57\u8bc6\u522b\u7b49\u3002\u4e0b\u9762\u5c06\u8be6\u7ec6\u4ecb\u7ecd\u5176\u4e2d\u4e00\u79cd\u65b9\u6cd5\uff1a\u4f7f\u7528Tesseract OCR\u548cPytesseract\u5e93\u6765\u8f6c\u5316\u56fe\u7247\u4e2d\u6587\u5b57\u3002<\/p>\n<\/p>\n<p><p>\u4e00\u3001\u5b89\u88c5Tesseract OCR<\/p>\n<\/p>\n<p><p>\u8981\u4f7f\u7528Tesseract OCR\uff0c\u9996\u5148\u9700\u8981\u5728\u7cfb\u7edf\u4e0a\u5b89\u88c5Tesseract\u3002Tesseract OCR\u662f\u4e00\u4e2a\u5f00\u6e90\u7684\u5149\u5b66\u5b57\u7b26\u8bc6\u522b\u5f15\u64ce\uff0c\u53ef\u4ee5\u4ece\u56fe\u7247\u4e2d\u63d0\u53d6\u6587\u672c\u3002\u5b89\u88c5\u65b9\u6cd5\u56e0\u64cd\u4f5c\u7cfb\u7edf\u4e0d\u540c\u800c\u6709\u6240\u533a\u522b\uff0c\u4e0b\u9762\u4ecb\u7ecd\u51e0\u79cd\u5e38\u89c1\u7684\u5b89\u88c5\u65b9\u6cd5\uff1a<\/p>\n<\/p>\n<ul>\n<li><strong>Windows\u7cfb\u7edf<\/strong>\uff1a\u53ef\u4ee5\u4eceTesseract\u7684\u5b98\u65b9GitHub\u9875\u9762\u4e0b\u8f7dWindows\u5b89\u88c5\u5305\uff0c\u7136\u540e\u8fdb\u884c\u5b89\u88c5\u3002\u5728\u5b89\u88c5\u8fc7\u7a0b\u4e2d\u8bb0\u5f97\u52fe\u9009\u201cAdd Tesseract to the system PATH for current user\u201d\u9009\u9879\uff0c\u4ee5\u4fbf\u5728\u547d\u4ee4\u884c\u4e2d\u53ef\u4ee5\u76f4\u63a5\u4f7f\u7528<code>tesseract<\/code>\u547d\u4ee4\u3002<\/li>\n<li><strong>macOS\u7cfb\u7edf<\/strong>\uff1a\u53ef\u4ee5\u4f7f\u7528Homebrew\u6765\u5b89\u88c5Tesseract\u3002\u5728\u7ec8\u7aef\u4e2d\u8fd0\u884c\u4ee5\u4e0b\u547d\u4ee4\uff1a\n<pre><code class=\"language-bash\">brew install tesseract<\/p>\n<p><\/code><\/pre>\n<\/p>\n<\/li>\n<li><strong>Linux\u7cfb\u7edf<\/strong>\uff1a\u53ef\u4ee5\u4f7f\u7528\u5305\u7ba1\u7406\u5668\u5b89\u88c5Tesseract\u3002\u4f8b\u5982\uff0c\u5728Debian\u7cfb\u7684Linux\u53d1\u884c\u7248\uff08\u5982Ubuntu\uff09\u4e0a\uff0c\u53ef\u4ee5\u8fd0\u884c\u4ee5\u4e0b\u547d\u4ee4\uff1a\n<pre><code class=\"language-bash\">sudo apt-get install tesseract-ocr<\/p>\n<p><\/code><\/pre>\n<\/p>\n<\/li>\n<\/ul>\n<p><p>\u4e8c\u3001\u5b89\u88c5Pytesseract\u5e93<\/p>\n<\/p>\n<p><p>\u5b89\u88c5\u5b8cTesseract OCR\u540e\uff0c\u8fd8\u9700\u8981\u5b89\u88c5Pytesseract\u5e93\u3002Pytesseract\u662fTesseract\u7684Python\u5c01\u88c5\uff0c\u4f7f\u5f97\u6211\u4eec\u53ef\u4ee5\u5728Python\u4ee3\u7801\u4e2d\u8c03\u7528Tesseract\u8fdb\u884c\u6587\u5b57\u8bc6\u522b\u3002\u53ef\u4ee5\u4f7f\u7528pip\u6765\u5b89\u88c5Pytesseract\u5e93\uff0c\u5728\u7ec8\u7aef\u4e2d\u8fd0\u884c\u4ee5\u4e0b\u547d\u4ee4\uff1a<\/p>\n<\/p>\n<p><pre><code class=\"language-bash\">pip install pytesseract<\/p>\n<p><\/code><\/pre>\n<\/p>\n<p><p>\u4e09\u3001\u8bfb\u53d6\u56fe\u7247<\/p>\n<\/p>\n<p><p>\u8bfb\u53d6\u56fe\u7247\u53ef\u4ee5\u4f7f\u7528Pillow\u5e93\uff0c\u5b83\u662fPython Imaging Library (PIL) \u7684\u4e00\u4e2a\u53cb\u597d\u5206\u652f\uff0c\u53ef\u4ee5\u65b9\u4fbf\u5730\u8fdb\u884c\u56fe\u50cf\u5904\u7406\u3002\u9996\u5148\u9700\u8981\u5b89\u88c5Pillow\u5e93\uff1a<\/p>\n<\/p>\n<p><pre><code class=\"language-bash\">pip install pillow<\/p>\n<p><\/code><\/pre>\n<\/p>\n<p><p>\u7136\u540e\u53ef\u4ee5\u4f7f\u7528Pillow\u5e93\u6765\u8bfb\u53d6\u56fe\u7247\uff1a<\/p>\n<\/p>\n<p><pre><code class=\"language-python\">from PIL import Image<\/p>\n<p>image_path = &#39;path\/to\/your\/image.png&#39;<\/p>\n<p>image = Image.open(image_path)<\/p>\n<p><\/code><\/pre>\n<\/p>\n<p><p>\u56db\u3001\u8c03\u7528Pytesseract\u8fdb\u884c\u6587\u5b57\u8bc6\u522b<\/p>\n<\/p>\n<p><p>\u8bfb\u53d6\u56fe\u7247\u540e\uff0c\u5c31\u53ef\u4ee5\u4f7f\u7528Pytesseract\u5e93\u6765\u8fdb\u884c\u6587\u5b57\u8bc6\u522b\u4e86\u3002\u4ee5\u4e0b\u662f\u4e00\u4e2a\u5b8c\u6574\u7684\u793a\u4f8b\u4ee3\u7801\uff1a<\/p>\n<\/p>\n<p><pre><code class=\"language-python\">import pytesseract<\/p>\n<p>from PIL import Image<\/p>\n<h2><strong>\u6307\u5b9atesseract\u7684\u5b89\u88c5\u8def\u5f84\uff08\u5982\u679c\u6ca1\u6709\u6dfb\u52a0\u5230\u7cfb\u7edfPATH\u4e2d\uff0c\u9700\u8981\u6307\u5b9a\u5177\u4f53\u8def\u5f84\uff09<\/strong><\/h2>\n<p>pytesseract.pytesseract.tesseract_cmd = r&#39;C:\\Program Files\\Tesseract-OCR\\tesseract.exe&#39;<\/p>\n<h2><strong>\u8bfb\u53d6\u56fe\u7247<\/strong><\/h2>\n<p>image_path = &#39;path\/to\/your\/image.png&#39;<\/p>\n<p>image = Image.open(image_path)<\/p>\n<h2><strong>\u8fdb\u884c\u6587\u5b57\u8bc6\u522b<\/strong><\/h2>\n<p>text = pytesseract.image_to_string(image, lang=&#39;eng&#39;)<\/p>\n<h2><strong>\u8f93\u51fa\u8bc6\u522b\u7ed3\u679c<\/strong><\/h2>\n<p>print(text)<\/p>\n<p><\/code><\/pre>\n<\/p>\n<p><p>\u5728\u8fd9\u6bb5\u4ee3\u7801\u4e2d\uff0c\u9996\u5148\u6307\u5b9a\u4e86Tesseract\u7684\u5b89\u88c5\u8def\u5f84\uff08\u5982\u679c\u5df2\u7ecf\u6dfb\u52a0\u5230\u7cfb\u7edfPATH\u4e2d\uff0c\u53ef\u4ee5\u7701\u7565\u8fd9\u4e00\u6b65\uff09\uff0c\u7136\u540e\u8bfb\u53d6\u56fe\u7247\uff0c\u5e76\u4f7f\u7528Pytesseract\u7684<code>image_to_string<\/code>\u65b9\u6cd5\u8fdb\u884c\u6587\u5b57\u8bc6\u522b\u3002<code>lang=&#39;eng&#39;<\/code>\u53c2\u6570\u6307\u5b9a\u4e86\u4f7f\u7528\u82f1\u8bed\u5b57\u5178\u8fdb\u884c\u8bc6\u522b\uff0c\u53ef\u4ee5\u6839\u636e\u9700\u8981\u9009\u62e9\u5176\u4ed6\u8bed\u8a00\u5b57\u5178\u3002\u6700\u540e\uff0c\u5c06\u8bc6\u522b\u7ed3\u679c\u8f93\u51fa\u5230\u63a7\u5236\u53f0\u3002<\/p>\n<\/p>\n<p><p>\u4e94\u3001\u63d0\u9ad8\u8bc6\u522b\u51c6\u786e\u6027<\/p>\n<\/p>\n<p><p>\u4e3a\u4e86\u63d0\u9ad8\u8bc6\u522b\u7684\u51c6\u786e\u6027\uff0c\u53ef\u4ee5\u5bf9\u56fe\u7247\u8fdb\u884c\u9884\u5904\u7406\uff0c\u4f8b\u5982\u7070\u5ea6\u5316\u3001\u4e8c\u503c\u5316\u3001\u964d\u566a\u7b49\u3002\u4ee5\u4e0b\u662f\u4e00\u4e9b\u5e38\u89c1\u7684\u9884\u5904\u7406\u65b9\u6cd5\uff1a<\/p>\n<\/p>\n<ol>\n<li>\n<p><strong>\u7070\u5ea6\u5316<\/strong>\uff1a\u5c06\u5f69\u8272\u56fe\u7247\u8f6c\u6362\u4e3a\u7070\u5ea6\u56fe\u7247\uff0c\u53ef\u4ee5\u4f7f\u7528Pillow\u5e93\u4e2d\u7684<code>convert<\/code>\u65b9\u6cd5\uff1a<\/p>\n<\/p>\n<p><pre><code class=\"language-python\">gray_image = image.convert(&#39;L&#39;)<\/p>\n<p><\/code><\/pre>\n<\/p>\n<\/li>\n<li>\n<p><strong>\u4e8c\u503c\u5316<\/strong>\uff1a\u5c06\u7070\u5ea6\u56fe\u7247\u8f6c\u6362\u4e3a\u4e8c\u503c\u56fe\u7247\uff0c\u53ef\u4ee5\u4f7f\u7528OpenCV\u5e93\u4e2d\u7684<code>threshold<\/code>\u65b9\u6cd5\u3002\u9996\u5148\u9700\u8981\u5b89\u88c5OpenCV\u5e93\uff1a<\/p>\n<\/p>\n<p><pre><code class=\"language-bash\">pip install opencv-python<\/p>\n<p><\/code><\/pre>\n<\/p>\n<p><p>\u7136\u540e\u53ef\u4ee5\u4f7f\u7528\u4ee5\u4e0b\u4ee3\u7801\u8fdb\u884c\u4e8c\u503c\u5316\u5904\u7406\uff1a<\/p>\n<\/p>\n<p><pre><code class=\"language-python\">import cv2<\/p>\n<p>import numpy as np<\/p>\n<p>image = cv2.imread(image_path)<\/p>\n<p>gray_image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)<\/p>\n<p>_, binary_image = cv2.threshold(gray_image, 127, 255, cv2.THRESH_BINARY)<\/p>\n<p><\/code><\/pre>\n<\/p>\n<\/li>\n<li>\n<p><strong>\u964d\u566a<\/strong>\uff1a\u53ef\u4ee5\u4f7f\u7528OpenCV\u5e93\u4e2d\u7684<code>medianBlur<\/code>\u65b9\u6cd5\u8fdb\u884c\u4e2d\u503c\u6ee4\u6ce2\u964d\u566a\uff1a<\/p>\n<\/p>\n<p><pre><code class=\"language-python\">denoised_image = cv2.medianBlur(binary_image, 3)<\/p>\n<p><\/code><\/pre>\n<\/p>\n<\/li>\n<li>\n<p><strong>\u81a8\u80c0\u548c\u8150\u8680<\/strong>\uff1a\u53ef\u4ee5\u4f7f\u7528OpenCV\u5e93\u4e2d\u7684<code>dilate<\/code>\u548c<code>erode<\/code>\u65b9\u6cd5\u8fdb\u884c\u81a8\u80c0\u548c\u8150\u8680\u64cd\u4f5c\uff0c\u4ee5\u589e\u5f3a\u6587\u5b57\u533a\u57df\uff1a<\/p>\n<\/p>\n<p><pre><code class=\"language-python\">kernel = np.ones((1, 1), np.uint8)<\/p>\n<p>dilated_image = cv2.dilate(denoised_image, kernel, iterations=1)<\/p>\n<p>eroded_image = cv2.erode(dilated_image, kernel, iterations=1)<\/p>\n<p><\/code><\/pre>\n<\/p>\n<\/li>\n<\/ol>\n<p><p>\u4ee5\u4e0b\u662f\u4e00\u4e2a\u5b8c\u6574\u7684\u9884\u5904\u7406\u548c\u6587\u5b57\u8bc6\u522b\u793a\u4f8b\u4ee3\u7801\uff1a<\/p>\n<\/p>\n<p><pre><code class=\"language-python\">import pytesseract<\/p>\n<p>from PIL import Image<\/p>\n<p>import cv2<\/p>\n<p>import numpy as np<\/p>\n<h2><strong>\u6307\u5b9atesseract\u7684\u5b89\u88c5\u8def\u5f84\uff08\u5982\u679c\u6ca1\u6709\u6dfb\u52a0\u5230\u7cfb\u7edfPATH\u4e2d\uff0c\u9700\u8981\u6307\u5b9a\u5177\u4f53\u8def\u5f84\uff09<\/strong><\/h2>\n<p>pytesseract.pytesseract.tesseract_cmd = r&#39;C:\\Program Files\\Tesseract-OCR\\tesseract.exe&#39;<\/p>\n<h2><strong>\u8bfb\u53d6\u56fe\u7247<\/strong><\/h2>\n<p>image_path = &#39;path\/to\/your\/image.png&#39;<\/p>\n<p>image = cv2.imread(image_path)<\/p>\n<h2><strong>\u7070\u5ea6\u5316<\/strong><\/h2>\n<p>gray_image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)<\/p>\n<h2><strong>\u4e8c\u503c\u5316<\/strong><\/h2>\n<p>_, binary_image = cv2.threshold(gray_image, 127, 255, cv2.THRESH_BINARY)<\/p>\n<h2><strong>\u964d\u566a<\/strong><\/h2>\n<p>denoised_image = cv2.medianBlur(binary_image, 3)<\/p>\n<h2><strong>\u81a8\u80c0\u548c\u8150\u8680<\/strong><\/h2>\n<p>kernel = np.ones((1, 1), np.uint8)<\/p>\n<p>dilated_image = cv2.dilate(denoised_image, kernel, iterations=1)<\/p>\n<p>eroded_image = cv2.erode(dilated_image, kernel, iterations=1)<\/p>\n<h2><strong>\u4fdd\u5b58\u9884\u5904\u7406\u540e\u7684\u56fe\u7247\uff08\u53ef\u9009\uff09<\/strong><\/h2>\n<p>cv2.imwrite(&#39;preprocessed_image.png&#39;, eroded_image)<\/p>\n<h2><strong>\u8fdb\u884c\u6587\u5b57\u8bc6\u522b<\/strong><\/h2>\n<p>text = pytesseract.image_to_string(eroded_image, lang=&#39;eng&#39;)<\/p>\n<h2><strong>\u8f93\u51fa\u8bc6\u522b\u7ed3\u679c<\/strong><\/h2>\n<p>print(text)<\/p>\n<p><\/code><\/pre>\n<\/p>\n<p><p>\u516d\u3001\u5904\u7406\u591a\u9875PDF\u6587\u4ef6<\/p>\n<\/p>\n<p><p>\u5728\u67d0\u4e9b\u60c5\u51b5\u4e0b\uff0c\u53ef\u80fd\u9700\u8981\u5904\u7406\u591a\u9875PDF\u6587\u4ef6\u3002\u53ef\u4ee5\u4f7f\u7528PyMuPDF\u5e93\uff08\u53c8\u540dfitz\uff09\u6765\u5c06PDF\u6587\u4ef6\u4e2d\u7684\u6bcf\u4e00\u9875\u8f6c\u6362\u4e3a\u56fe\u7247\uff0c\u7136\u540e\u8fdb\u884c\u6587\u5b57\u8bc6\u522b\u3002\u9996\u5148\u9700\u8981\u5b89\u88c5PyMuPDF\u5e93\uff1a<\/p>\n<\/p>\n<p><pre><code class=\"language-bash\">pip install pymupdf<\/p>\n<p><\/code><\/pre>\n<\/p>\n<p><p>\u7136\u540e\u53ef\u4ee5\u4f7f\u7528\u4ee5\u4e0b\u4ee3\u7801\u5904\u7406\u591a\u9875PDF\u6587\u4ef6\uff1a<\/p>\n<\/p>\n<p><pre><code class=\"language-python\">import pytesseract<\/p>\n<p>from PIL import Image<\/p>\n<p>import fitz  # PyMuPDF<\/p>\n<p>import cv2<\/p>\n<p>import numpy as np<\/p>\n<h2><strong>\u6307\u5b9atesseract\u7684\u5b89\u88c5\u8def\u5f84\uff08\u5982\u679c\u6ca1\u6709\u6dfb\u52a0\u5230\u7cfb\u7edfPATH\u4e2d\uff0c\u9700\u8981\u6307\u5b9a\u5177\u4f53\u8def\u5f84\uff09<\/strong><\/h2>\n<p>pytesseract.pytesseract.tesseract_cmd = r&#39;C:\\Program Files\\Tesseract-OCR\\tesseract.exe&#39;<\/p>\n<h2><strong>\u8bfb\u53d6PDF\u6587\u4ef6<\/strong><\/h2>\n<p>pdf_path = &#39;path\/to\/your\/document.pdf&#39;<\/p>\n<p>pdf_document = fitz.open(pdf_path)<\/p>\n<h2><strong>\u904d\u5386\u6bcf\u4e00\u9875<\/strong><\/h2>\n<p>for page_num in range(pdf_document.page_count):<\/p>\n<p>    # \u83b7\u53d6\u9875\u9762<\/p>\n<p>    page = pdf_document.load_page(page_num)<\/p>\n<p>    # \u5c06\u9875\u9762\u8f6c\u6362\u4e3a\u56fe\u7247<\/p>\n<p>    pix = page.get_pixmap()<\/p>\n<p>    image = Image.frombytes(&#39;RGB&#39;, [pix.width, pix.height], pix.samples)<\/p>\n<p>    # \u5c06\u56fe\u7247\u4fdd\u5b58\u5230\u6587\u4ef6\uff08\u53ef\u9009\uff09<\/p>\n<p>    image_path = f&#39;page_{page_num + 1}.png&#39;<\/p>\n<p>    image.save(image_path)<\/p>\n<p>    # \u8f6c\u6362\u4e3aOpenCV\u683c\u5f0f<\/p>\n<p>    image_cv = cv2.cvtColor(np.array(image), cv2.COLOR_RGB2BGR)<\/p>\n<p>    # \u7070\u5ea6\u5316<\/p>\n<p>    gray_image = cv2.cvtColor(image_cv, cv2.COLOR_BGR2GRAY)<\/p>\n<p>    # \u4e8c\u503c\u5316<\/p>\n<p>    _, binary_image = cv2.threshold(gray_image, 127, 255, cv2.THRESH_BINARY)<\/p>\n<p>    # \u964d\u566a<\/p>\n<p>    denoised_image = cv2.medianBlur(binary_image, 3)<\/p>\n<p>    # \u81a8\u80c0\u548c\u8150\u8680<\/p>\n<p>    kernel = np.ones((1, 1), np.uint8)<\/p>\n<p>    dilated_image = cv2.dilate(denoised_image, kernel, iterations=1)<\/p>\n<p>    eroded_image = cv2.erode(dilated_image, kernel, iterations=1)<\/p>\n<p>    # \u8fdb\u884c\u6587\u5b57\u8bc6\u522b<\/p>\n<p>    text = pytesseract.image_to_string(eroded_image, lang=&#39;eng&#39;)<\/p>\n<p>    # \u8f93\u51fa\u8bc6\u522b\u7ed3\u679c<\/p>\n<p>    print(f&#39;Page {page_num + 1}:\\n{text}\\n&#39;)<\/p>\n<h2><strong>\u5173\u95edPDF\u6587\u4ef6<\/strong><\/h2>\n<p>pdf_document.close()<\/p>\n<p><\/code><\/pre>\n<\/p>\n<p><p>\u8fd9\u6bb5\u4ee3\u7801\u9996\u5148\u4f7f\u7528PyMuPDF\u5e93\u8bfb\u53d6PDF\u6587\u4ef6\uff0c\u5e76\u5c06\u6bcf\u4e00\u9875\u8f6c\u6362\u4e3a\u56fe\u7247\u3002\u7136\u540e\u5bf9\u6bcf\u4e00\u9875\u56fe\u7247\u8fdb\u884c\u9884\u5904\u7406\u548c\u6587\u5b57\u8bc6\u522b\uff0c\u6700\u540e\u8f93\u51fa\u8bc6\u522b\u7ed3\u679c\u3002\u8fd9\u6837\u53ef\u4ee5\u65b9\u4fbf\u5730\u5904\u7406\u591a\u9875PDF\u6587\u4ef6\u4e2d\u7684\u6587\u5b57\u3002<\/p>\n<\/p>\n<p><p>\u4e03\u3001\u5904\u7406\u591a\u79cd\u8bed\u8a00<\/p>\n<\/p>\n<p><p>\u5728\u5b9e\u9645\u5e94\u7528\u4e2d\uff0c\u53ef\u80fd\u9700\u8981\u5904\u7406\u591a\u79cd\u8bed\u8a00\u7684\u6587\u5b57\u8bc6\u522b\u3002Tesseract\u652f\u6301\u591a\u79cd\u8bed\u8a00\u5b57\u5178\uff0c\u53ef\u4ee5\u901a\u8fc7\u5b89\u88c5\u76f8\u5e94\u7684\u8bed\u8a00\u6570\u636e\u5305\u6765\u5b9e\u73b0\u591a\u8bed\u8a00\u8bc6\u522b\u3002\u4ee5\u4e0b\u662f\u4e00\u4e9b\u5e38\u89c1\u8bed\u8a00\u6570\u636e\u5305\u7684\u5b89\u88c5\u65b9\u6cd5\uff1a<\/p>\n<\/p>\n<ul>\n<li><strong>Windows\u7cfb\u7edf<\/strong>\uff1a\u5728Tesseract\u7684\u5b89\u88c5\u76ee\u5f55\u4e0b\u7684<code>tessdata<\/code>\u6587\u4ef6\u5939\u4e2d\u653e\u7f6e\u76f8\u5e94\u7684\u8bed\u8a00\u6570\u636e\u5305\uff08\u4f8b\u5982<code>chi_sim.tr<a href=\"https:\/\/docs.pingcode.com\/blog\/59162.html\" target=\"_blank\">AI<\/a>neddata<\/code>\u7528\u4e8e\u7b80\u4f53\u4e2d\u6587\uff09\u3002<\/li>\n<li><strong>macOS\u548cLinux\u7cfb\u7edf<\/strong>\uff1a\u53ef\u4ee5\u4eceTesseract\u7684\u5b98\u65b9GitHub\u9875\u9762\u4e0b\u8f7d\u76f8\u5e94\u7684\u8bed\u8a00\u6570\u636e\u5305\uff0c\u5e76\u5c06\u5176\u653e\u7f6e\u5728Tesseract\u7684<code>tessdata<\/code>\u76ee\u5f55\u4e2d\u3002<\/li>\n<\/ul>\n<p><p>\u5728\u8fdb\u884c\u6587\u5b57\u8bc6\u522b\u65f6\uff0c\u53ef\u4ee5\u901a\u8fc7<code>lang<\/code>\u53c2\u6570\u6307\u5b9a\u4f7f\u7528\u7684\u8bed\u8a00\u5b57\u5178\u3002\u4f8b\u5982\uff0c\u8bc6\u522b\u7b80\u4f53\u4e2d\u6587\u53ef\u4ee5\u4f7f\u7528<code>lang=&#39;chi_sim&#39;<\/code>\u53c2\u6570\uff1a<\/p>\n<\/p>\n<p><pre><code class=\"language-python\">text = pytesseract.image_to_string(eroded_image, lang=&#39;chi_sim&#39;)<\/p>\n<p><\/code><\/pre>\n<\/p>\n<p><p>\u5982\u679c\u9700\u8981\u540c\u65f6\u8bc6\u522b\u591a\u79cd\u8bed\u8a00\uff0c\u53ef\u4ee5\u901a\u8fc7\u9017\u53f7\u5206\u9694\u7684\u65b9\u5f0f\u6307\u5b9a\u591a\u4e2a\u8bed\u8a00\u5b57\u5178\uff1a<\/p>\n<\/p>\n<p><pre><code class=\"language-python\">text = pytesseract.image_to_string(eroded_image, lang=&#39;eng+chi_sim&#39;)<\/p>\n<p><\/code><\/pre>\n<\/p>\n<p><p>\u516b\u3001\u5904\u7406\u590d\u6742\u6392\u7248<\/p>\n<\/p>\n<p><p>\u5728\u5904\u7406\u590d\u6742\u6392\u7248\u7684\u56fe\u7247\u65f6\uff0c\u53ef\u80fd\u9700\u8981\u4f7f\u7528Tesseract\u7684\u5e03\u5c40\u5206\u6790\u529f\u80fd\u3002Tesseract\u63d0\u4f9b\u4e86\u51e0\u79cd\u4e0d\u540c\u7684\u9875\u9762\u5e03\u5c40\u6a21\u5f0f\uff0c\u53ef\u4ee5\u901a\u8fc7<code>config<\/code>\u53c2\u6570\u8fdb\u884c\u914d\u7f6e\u3002\u4ee5\u4e0b\u662f\u4e00\u4e9b\u5e38\u89c1\u7684\u9875\u9762\u5e03\u5c40\u6a21\u5f0f\uff1a<\/p>\n<\/p>\n<ul>\n<li><code>--psm 1<\/code>\uff1a\u81ea\u52a8\u68c0\u6d4b\u9875\u9762\u5e03\u5c40\uff08\u9ed8\u8ba4\uff09\u3002<\/li>\n<li><code>--psm 3<\/code>\uff1a\u5b8c\u5168\u81ea\u52a8\u7684\u9875\u9762\u5206\u5272\uff0c\u4f46\u6ca1\u6709OCR\u3002<\/li>\n<li><code>--psm 6<\/code>\uff1a\u5047\u8bbe\u4e00\u4e2a\u5355\u72ec\u7684\u7edf\u4e00\u7684\u5757\uff08\u5355\u5217\uff09\u3002<\/li>\n<\/ul>\n<p><p>\u53ef\u4ee5\u901a\u8fc7\u4ee5\u4e0b\u4ee3\u7801\u6307\u5b9a\u9875\u9762\u5e03\u5c40\u6a21\u5f0f\uff1a<\/p>\n<\/p>\n<p><pre><code class=\"language-python\">config = &#39;--psm 6&#39;<\/p>\n<p>text = pytesseract.image_to_string(eroded_image, lang=&#39;eng&#39;, config=config)<\/p>\n<p><\/code><\/pre>\n<\/p>\n<p><p>\u4e5d\u3001\u4fdd\u5b58\u8bc6\u522b\u7ed3\u679c<\/p>\n<\/p>\n<p><p>\u4e3a\u4e86\u65b9\u4fbf\u540e\u7eed\u5904\u7406\uff0c\u53ef\u4ee5\u5c06\u8bc6\u522b\u7ed3\u679c\u4fdd\u5b58\u5230\u6587\u4ef6\u4e2d\u3002\u4ee5\u4e0b\u662f\u4e00\u4e2a\u7b80\u5355\u7684\u793a\u4f8b\uff0c\u5c06\u8bc6\u522b\u7ed3\u679c\u4fdd\u5b58\u5230\u6587\u672c\u6587\u4ef6\uff1a<\/p>\n<\/p>\n<p><pre><code class=\"language-python\">with open(&#39;recognized_text.txt&#39;, &#39;w&#39;, encoding=&#39;utf-8&#39;) as file:<\/p>\n<p>    file.write(text)<\/p>\n<p><\/code><\/pre>\n<\/p>\n<p><p>\u5341\u3001\u603b\u7ed3<\/p>\n<\/p>\n<p><p>\u672c\u6587\u8be6\u7ec6\u4ecb\u7ecd\u4e86Python\u8f6c\u5316\u56fe\u7247\u4e2d\u6587\u5b57\u7684\u591a\u79cd\u65b9\u6cd5\uff0c\u91cd\u70b9\u4ecb\u7ecd\u4e86\u4f7f\u7528Tesseract OCR\u548cPytesseract\u5e93\u8fdb\u884c\u6587\u5b57\u8bc6\u522b\u7684\u8fc7\u7a0b\u3002\u901a\u8fc7\u5b89\u88c5Tesseract OCR\u3001\u5b89\u88c5Pytesseract\u5e93\u3001\u8bfb\u53d6\u56fe\u7247\u3001\u8c03\u7528Pytesseract\u8fdb\u884c\u6587\u5b57\u8bc6\u522b\uff0c\u53ef\u4ee5\u65b9\u4fbf\u5730\u5b9e\u73b0\u56fe\u7247\u4e2d\u6587\u5b57\u7684\u8f6c\u5316\u3002\u4e3a\u4e86\u63d0\u9ad8\u8bc6\u522b\u7684\u51c6\u786e\u6027\uff0c\u53ef\u4ee5\u5bf9\u56fe\u7247\u8fdb\u884c\u9884\u5904\u7406\uff0c\u4f8b\u5982\u7070\u5ea6\u5316\u3001\u4e8c\u503c\u5316\u3001\u964d\u566a\u7b49\u3002\u6b64\u5916\uff0c\u8fd8\u4ecb\u7ecd\u4e86\u5904\u7406\u591a\u9875PDF\u6587\u4ef6\u3001\u591a\u79cd\u8bed\u8a00\u548c\u590d\u6742\u6392\u7248\u7684\u65b9\u6cd5\uff0c\u4ee5\u53ca\u4fdd\u5b58\u8bc6\u522b\u7ed3\u679c\u7684\u793a\u4f8b\u4ee3\u7801\u3002\u5e0c\u671b\u8fd9\u4e9b\u5185\u5bb9\u5bf9\u60a8\u6709\u6240\u5e2e\u52a9\u3002<\/p>\n<\/p>\n<h2><strong>\u76f8\u5173\u95ee\u7b54FAQs\uff1a<\/strong><\/h2>\n<p> <strong>\u5982\u4f55\u4f7f\u7528Python\u63d0\u53d6\u56fe\u7247\u4e2d\u7684\u6587\u5b57\uff1f<\/strong><br \/>\u5728Python\u4e2d\uff0c\u53ef\u4ee5\u5229\u7528OCR\uff08\u5149\u5b66\u5b57\u7b26\u8bc6\u522b\uff09\u6280\u672f\u6765\u63d0\u53d6\u56fe\u7247\u4e2d\u7684\u6587\u5b57\u3002\u4e00\u4e2a\u5e38\u7528\u7684\u5e93\u662fTesseract\u3002\u9996\u5148\uff0c\u786e\u4fdd\u5b89\u88c5\u4e86Tesseract\u548c\u5176Python\u5c01\u88c5\u5e93Pytesseract\u3002\u901a\u8fc7\u52a0\u8f7d\u56fe\u7247\u5e76\u8c03\u7528Pytesseract\u7684image_to_string()\u65b9\u6cd5\uff0c\u5c31\u53ef\u4ee5\u83b7\u5f97\u56fe\u7247\u4e2d\u7684\u6587\u5b57\u5185\u5bb9\u3002<\/p>\n<p><strong>Python\u4e2d\u6709\u54ea\u4e9b\u5e93\u53ef\u4ee5\u5b9e\u73b0\u56fe\u7247\u6587\u5b57\u8f6c\u5316\uff1f<\/strong><br \/>\u5b9e\u73b0\u56fe\u7247\u6587\u5b57\u8f6c\u5316\u7684\u5e38\u7528\u5e93\u5305\u62ecPytesseract\u3001OpenCV\u548cPillow\u3002Pytesseract\u4e3b\u8981\u7528\u4e8eOCR\u5904\u7406\uff0cOpenCV\u53ef\u4ee5\u7528\u4e8e\u56fe\u50cf\u5904\u7406\u548c\u9884\u5904\u7406\uff0c\u4ee5\u63d0\u9ad8OCR\u7684\u8bc6\u522b\u51c6\u786e\u6027\uff0c\u800cPillow\u5219\u7528\u4e8e\u6253\u5f00\u548c\u64cd\u4f5c\u56fe\u7247\u6587\u4ef6\uff0c\u8fd9\u4e9b\u5e93\u7684\u7ed3\u5408\u53ef\u4ee5\u5b9e\u73b0\u9ad8\u6548\u7684\u6587\u5b57\u63d0\u53d6\u3002<\/p>\n<p><strong>\u5982\u4f55\u63d0\u9ad8Python\u63d0\u53d6\u56fe\u7247\u4e2d\u6587\u5b57\u7684\u51c6\u786e\u6027\uff1f<\/strong><br \/>\u63d0\u9ad8\u6587\u5b57\u63d0\u53d6\u51c6\u786e\u6027\u7684\u65b9\u5f0f\u5305\u62ec\u5bf9\u56fe\u7247\u8fdb\u884c\u9884\u5904\u7406\uff0c\u4f8b\u5982\u8c03\u6574\u5bf9\u6bd4\u5ea6\u3001\u53bb\u566a\u58f0\u3001\u6539\u53d8\u56fe\u7247\u5927\u5c0f\u6216\u989c\u8272\u6a21\u5f0f\u7b49\u3002\u8fd9\u4e9b\u5904\u7406\u53ef\u4ee5\u4f7f\u7528OpenCV\u5e93\u5b8c\u6210\u3002\u6b64\u5916\uff0c\u786e\u4fdd\u4f7f\u7528\u6700\u65b0\u7248\u672c\u7684Tesseract\uff0c\u5e76\u6839\u636e\u5177\u4f53\u7684\u6587\u5b57\u7279\u5f81\uff08\u5982\u5b57\u4f53\u3001\u5927\u5c0f\u3001\u8bed\u8a00\u7b49\uff09\u8c03\u6574\u5176\u8bc6\u522b\u53c2\u6570\uff0c\u4e5f\u80fd\u663e\u8457\u63d0\u5347\u8bc6\u522b\u6548\u679c\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"Python\u8f6c\u5316\u56fe\u7247\u4e2d\u6587\u5b57\u7684\u65b9\u6cd5\u6709\u591a\u79cd\uff0c\u5e38\u89c1\u7684\u6709\u4f7f\u7528Tesseract OCR\u3001Pytesseract\u5e93\u3001Pi [&hellip;]","protected":false},"author":3,"featured_media":1080335,"comment_status":"closed","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[37],"tags":[],"acf":[],"_links":{"self":[{"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts\/1080326"}],"collection":[{"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/comments?post=1080326"}],"version-history":[{"count":"1","href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts\/1080326\/revisions"}],"predecessor-version":[{"id":1080336,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts\/1080326\/revisions\/1080336"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/media\/1080335"}],"wp:attachment":[{"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/media?parent=1080326"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/categories?post=1080326"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/tags?post=1080326"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}