{"id":1101889,"date":"2025-01-08T15:54:13","date_gmt":"2025-01-08T07:54:13","guid":{"rendered":""},"modified":"2025-01-08T15:54:16","modified_gmt":"2025-01-08T07:54:16","slug":"python%e5%a6%82%e4%bd%95%e8%af%86%e5%88%ab%e5%9b%be%e7%89%87%e4%b8%8a%e7%9a%84%e6%95%b0%e5%ad%97%e5%ad%97%e6%af%8d-2","status":"publish","type":"post","link":"https:\/\/docs.pingcode.com\/ask\/1101889.html","title":{"rendered":"python\u5982\u4f55\u8bc6\u522b\u56fe\u7247\u4e0a\u7684\u6570\u5b57\u5b57\u6bcd"},"content":{"rendered":"

\"python\u5982\u4f55\u8bc6\u522b\u56fe\u7247\u4e0a\u7684\u6570\u5b57\u5b57\u6bcd\"<\/p>\n

\u5728Python\u4e2d\u8bc6\u522b\u56fe\u7247\u4e0a\u7684\u6570\u5b57\u5b57\u6bcd\uff0c\u4e3b\u8981\u53ef\u4ee5\u901a\u8fc7\u4ee5\u4e0b\u51e0\u79cd\u65b9\u6cd5\uff1a\u4f7f\u7528OCR\u5e93\uff08\u5982Tesseract\uff09\u3001\u6df1\u5ea6\u5b66\u4e60\u6a21\u578b\uff08\u5982CNN\uff09\u3001\u9884\u5904\u7406\u56fe\u50cf\uff08\u7070\u5ea6\u5316\u3001\u4e8c\u503c\u5316\uff09<\/strong>\u3002\u5176\u4e2d\uff0c\u4f7f\u7528OCR\u5e93<\/strong>\u662f\u4e00\u79cd\u975e\u5e38\u65b9\u4fbf\u4e14\u5e38\u7528\u7684\u65b9\u6cd5\u3002Tesseract\u662f\u4e00\u4e2a\u5f00\u6e90\u7684OCR\uff08\u5149\u5b66\u5b57\u7b26\u8bc6\u522b\uff09\u5f15\u64ce\uff0c\u53ef\u4ee5\u901a\u8fc7Python\u5e93pytesseract\u8c03\u7528\u3002\u4e0b\u9762\u6211\u4eec\u5c06\u8be6\u7ec6\u4ecb\u7ecd\u5982\u4f55\u4f7f\u7528Tesseract\u6765\u8bc6\u522b\u56fe\u7247\u4e0a\u7684\u6570\u5b57\u548c\u5b57\u6bcd\u3002<\/p>\n<\/p>\n

\u4e00\u3001\u4f7f\u7528Tesseract\u8fdb\u884cOCR<\/h3>\n<\/p>\n

1. \u5b89\u88c5Tesseract<\/h4>\n<\/p>\n

\u5728\u4f7f\u7528Tesseract\u4e4b\u524d\uff0c\u9700\u8981\u5148\u5b89\u88c5Tesseract\u5f15\u64ce\u548cpytesseract\u5e93\u3002\u53ef\u4ee5\u901a\u8fc7\u4ee5\u4e0b\u547d\u4ee4\u5b89\u88c5\uff1a<\/p>\n<\/p>\n

sudo apt-get install tesseract-ocr<\/p>\n
pip install pytesseract<\/p>\n
pip install pillow<\/p>\n
<\/code><\/pre>\n<\/p>\n
2. \u8bfb\u53d6\u548c\u9884\u5904\u7406\u56fe\u50cf<\/h4>\n<\/p>\n
\u5728\u8bc6\u522b\u56fe\u50cf\u4e2d\u7684\u5b57\u7b26\u65f6\uff0c\u9884\u5904\u7406\u56fe\u50cf\u662f\u4e00\u4e2a\u91cd\u8981\u7684\u6b65\u9aa4\u3002\u901a\u5e38\uff0c\u6211\u4eec\u9700\u8981\u5c06\u56fe\u50cf\u8f6c\u6362\u4e3a\u7070\u5ea6\u56fe\u50cf\uff0c\u5e76\u8fdb\u884c\u4e8c\u503c\u5316\u5904\u7406\uff0c\u4ee5\u63d0\u9ad8\u8bc6\u522b\u7684\u51c6\u786e\u6027\u3002<\/p>\n<\/p>\n
from PIL import Image<\/p>\n
import pytesseract<\/p>\n
import cv2<\/p>\n
\u8bfb\u53d6\u56fe\u50cf<\/strong><\/h2>\n
image = cv2.imread('image.png')<\/p>\n
\u5c06\u56fe\u50cf\u8f6c\u6362\u4e3a\u7070\u5ea6\u56fe\u50cf<\/strong><\/h2>\n
gray_image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)<\/p>\n
\u4e8c\u503c\u5316\u5904\u7406<\/strong><\/h2>\n
_, binary_image = cv2.threshold(gray_image, 128, 255, cv2.THRESH_BINARY)<\/p>\n
\u4fdd\u5b58\u9884\u5904\u7406\u540e\u7684\u56fe\u50cf<\/strong><\/h2>\n
cv2.imwrite('processed_image.png', binary_image)<\/p>\n
<\/code><\/pre>\n<\/p>\n
3. \u4f7f\u7528Tesseract\u8bc6\u522b\u5b57\u7b26<\/h4>\n<\/p>\n
\u4f7f\u7528pytesseract\u8c03\u7528Tesseract\u5f15\u64ce\u8fdb\u884c\u5b57\u7b26\u8bc6\u522b\uff1a<\/p>\n<\/p>\n
# \u8bfb\u53d6\u9884\u5904\u7406\u540e\u7684\u56fe\u50cf<\/p>\n
processed_image = Image.open('processed_image.png')<\/p>\n
\u4f7f\u7528Tesseract\u8bc6\u522b\u5b57\u7b26<\/strong><\/h2>\n
text = pytesseract.image_to_string(processed_image)<\/p>\n
print("\u8bc6\u522b\u7ed3\u679c:", text)<\/p>\n
<\/code><\/pre>\n<\/p>\n
\u4e8c\u3001\u4f7f\u7528\u6df1\u5ea6\u5b66\u4e60\u6a21\u578b\uff08\u5982CNN\uff09<\/h3>\n<\/p>\n
1. \u6570\u636e\u96c6\u51c6\u5907<\/h4>\n<\/p>\n
\u5728\u4f7f\u7528\u6df1\u5ea6\u5b66\u4e60\u6a21\u578b\u8fdb\u884c\u5b57\u7b26\u8bc6\u522b\u65f6\uff0c\u9996\u5148\u9700\u8981\u51c6\u5907\u4e00\u4e2a\u5305\u542b\u5927\u91cf\u6807\u6ce8\u6570\u636e\u7684\u8bad\u7ec3\u96c6\u3002\u53ef\u4ee5\u4f7f\u7528\u73b0\u6210\u7684\u5b57\u7b26\u6570\u636e\u96c6\uff08\u5982MNIST\uff09\u6216\u81ea\u5df1\u5236\u4f5c\u6570\u636e\u96c6\u3002<\/p>\n<\/p>\n
2. \u6784\u5efaCNN\u6a21\u578b<\/h4>\n<\/p>\n
\u5229\u7528Keras\u6216TensorFlow\u7b49\u6df1\u5ea6\u5b66\u4e60\u6846\u67b6\u6784\u5efa\u4e00\u4e2a\u5377\u79ef\u795e\u7ecf\u7f51\u7edc\uff08CNN\uff09\u6a21\u578b\uff0c\u7528\u4e8e\u5b57\u7b26\u8bc6\u522b\u3002<\/p>\n<\/p>\n
from tensorflow.keras.models import Sequential<\/p>\n
from tensorflow.keras.layers import Conv2D, MaxPooling2D, Flatten, Dense<\/p>\n
model = Sequential([<\/p>\n
    Conv2D(32, (3, 3), activation='relu', input_shape=(28, 28, 1)),<\/p>\n
    MaxPooling2D((2, 2)),<\/p>\n
    Conv2D(64, (3, 3), activation='relu'),<\/p>\n
    MaxPooling2D((2, 2)),<\/p>\n
    Flatten(),<\/p>\n
    Dense(128, activation='relu'),<\/p>\n
    Dense(10, activation='softmax')<\/p>\n
])<\/p>\n
model.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])<\/p>\n
<\/code><\/pre>\n<\/p>\n
3. \u8bad\u7ec3\u6a21\u578b<\/h4>\n<\/p>\n
\u4f7f\u7528\u51c6\u5907\u597d\u7684\u6570\u636e\u96c6\u5bf9CNN\u6a21\u578b\u8fdb\u884c\u8bad\u7ec3\u3002<\/p>\n<\/p>\n
from tensorflow.keras.datasets import mnist<\/p>\n
\u52a0\u8f7d\u6570\u636e\u96c6<\/strong><\/h2>\n
(trAI<\/a>n_images, train_labels), (test_images, test_labels) = mnist.load_data()<\/p>\n
\u6570\u636e\u9884\u5904\u7406<\/strong><\/h2>\n
train_images = train_images.reshape((60000, 28, 28, 1)) \/ 255.0<\/p>\n
test_images = test_images.reshape((10000, 28, 28, 1)) \/ 255.0<\/p>\n
\u8bad\u7ec3\u6a21\u578b<\/strong><\/h2>\n
model.fit(train_images, train_labels, epochs=10, validation_data=(test_images, test_labels))<\/p>\n
<\/code><\/pre>\n<\/p>\n
4. \u8bc6\u522b\u56fe\u50cf\u4e2d\u7684\u5b57\u7b26<\/h4>\n<\/p>\n
\u4f7f\u7528\u8bad\u7ec3\u597d\u7684\u6a21\u578b\u8bc6\u522b\u56fe\u50cf\u4e2d\u7684\u5b57\u7b26\u3002<\/p>\n<\/p>\n
import numpy as np<\/p>\n
\u8bfb\u53d6\u5e76\u9884\u5904\u7406\u56fe\u50cf<\/strong><\/h2>\n
image = cv2.imread('image.png', cv2.IMREAD_GRAYSCALE)<\/p>\n
image = cv2.resize(image, (28, 28))<\/p>\n
image = image.reshape((1, 28, 28, 1)) \/ 255.0<\/p>\n
\u4f7f\u7528\u6a21\u578b\u9884\u6d4b<\/strong><\/h2>\n
prediction = model.predict(image)<\/p>\n
predicted_label = np.argmax(prediction)<\/p>\n
print("\u8bc6\u522b\u7ed3\u679c:", predicted_label)<\/p>\n
<\/code><\/pre>\n<\/p>\n
\u4e09\u3001\u56fe\u50cf\u9884\u5904\u7406<\/h3>\n<\/p>\n
1. \u7070\u5ea6\u5316<\/h4>\n<\/p>\n
\u7070\u5ea6\u5316\u662f\u5c06\u5f69\u8272\u56fe\u50cf\u8f6c\u6362\u4e3a\u7070\u5ea6\u56fe\u50cf\u7684\u8fc7\u7a0b\uff0c\u8fd9\u53ef\u4ee5\u51cf\u5c11\u8ba1\u7b97\u590d\u6742\u6027\uff0c\u5e76\u4e14\u5728\u5927\u591a\u6570\u60c5\u51b5\u4e0b\uff0c\u7070\u5ea6\u56fe\u50cf\u8db3\u4ee5\u8fdb\u884c\u5b57\u7b26\u8bc6\u522b\u3002<\/p>\n<\/p>\n
gray_image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)<\/p>\n
<\/code><\/pre>\n<\/p>\n
2. \u4e8c\u503c\u5316<\/h4>\n<\/p>\n
\u4e8c\u503c\u5316\u662f\u5c06\u7070\u5ea6\u56fe\u50cf\u8f6c\u6362\u4e3a\u53ea\u6709\u9ed1\u767d\u4e24\u8272\u7684\u56fe\u50cf\uff0c\u8fd9\u53ef\u4ee5\u8fdb\u4e00\u6b65\u51cf\u5c11\u8ba1\u7b97\u590d\u6742\u6027\uff0c\u5e76\u6709\u52a9\u4e8e\u63d0\u53d6\u5b57\u7b26\u8f6e\u5ed3\u3002<\/p>\n<\/p>\n
_, binary_image = cv2.threshold(gray_image, 128, 255, cv2.THRESH_BINARY)<\/p>\n
<\/code><\/pre>\n<\/p>\n
3. \u53bb\u566a<\/h4>\n<\/p>\n
\u53bb\u566a\u662f\u53bb\u9664\u56fe\u50cf\u4e2d\u7684\u566a\u58f0\uff0c\u8fd9\u53ef\u4ee5\u63d0\u9ad8\u5b57\u7b26\u8bc6\u522b\u7684\u51c6\u786e\u6027\u3002\u5e38\u7528\u7684\u53bb\u566a\u65b9\u6cd5\u5305\u62ec\u9ad8\u65af\u6a21\u7cca\u3001\u4e2d\u503c\u6ee4\u6ce2\u7b49\u3002<\/p>\n<\/p>\n
denoised_image = cv2.GaussianBlur(binary_image, (5, 5), 0)<\/p>\n
<\/code><\/pre>\n<\/p>\n
\u56db\u3001\u7efc\u5408\u5e94\u7528<\/h3>\n<\/p>\n
\u5728\u5b9e\u9645\u5e94\u7528\u4e2d\uff0c\u53ef\u4ee5\u7ed3\u5408\u4e0a\u8ff0\u65b9\u6cd5\u6765\u8bc6\u522b\u56fe\u50cf\u4e2d\u7684\u5b57\u7b26\u3002\u4f8b\u5982\uff0c\u5148\u4f7f\u7528\u56fe\u50cf\u9884\u5904\u7406\u65b9\u6cd5\u5bf9\u56fe\u50cf\u8fdb\u884c\u5904\u7406\uff0c\u7136\u540e\u4f7f\u7528Tesseract\u8fdb\u884cOCR\u8bc6\u522b\uff0c\u6216\u8005\u4f7f\u7528\u8bad\u7ec3\u597d\u7684\u6df1\u5ea6\u5b66\u4e60\u6a21\u578b\u8fdb\u884c\u8bc6\u522b\u3002<\/p>\n<\/p>\n
# \u8bfb\u53d6\u56fe\u50cf<\/p>\n
image = cv2.imread('image.png')<\/p>\n
\u56fe\u50cf\u9884\u5904\u7406<\/strong><\/h2>\n
gray_image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)<\/p>\n
_, binary_image = cv2.threshold(gray_image, 128, 255, cv2.THRESH_BINARY)<\/p>\n
denoised_image = cv2.GaussianBlur(binary_image, (5, 5), 0)<\/p>\n
\u4fdd\u5b58\u9884\u5904\u7406\u540e\u7684\u56fe\u50cf<\/strong><\/h2>\n
cv2.imwrite('processed_image.png', denoised_image)<\/p>\n
\u4f7f\u7528Tesseract\u8bc6\u522b\u5b57\u7b26<\/strong><\/h2>\n
processed_image = Image.open('processed_image.png')<\/p>\n
text = pytesseract.image_to_string(processed_image)<\/p>\n
print("\u8bc6\u522b\u7ed3\u679c:", text)<\/p>\n
<\/code><\/pre>\n<\/p>\n
\u901a\u8fc7\u4e0a\u8ff0\u6b65\u9aa4\uff0c\u6211\u4eec\u53ef\u4ee5\u4f7f\u7528Python\u8bc6\u522b\u56fe\u50cf\u4e2d\u7684\u6570\u5b57\u548c\u5b57\u6bcd\u3002\u65e0\u8bba\u662f\u4f7f\u7528OCR\u5e93\u8fd8\u662f\u6df1\u5ea6\u5b66\u4e60\u6a21\u578b\uff0c\u90fd\u9700\u8981\u5bf9\u56fe\u50cf\u8fdb\u884c\u9002\u5f53\u7684\u9884\u5904\u7406\uff0c\u4ee5\u63d0\u9ad8\u8bc6\u522b\u7684\u51c6\u786e\u6027\u3002\u5728\u5b9e\u9645\u5e94\u7528\u4e2d\uff0c\u53ef\u4ee5\u6839\u636e\u5177\u4f53\u9700\u6c42\u9009\u62e9\u5408\u9002\u7684\u65b9\u6cd5\uff0c\u5e76\u4e0d\u65ad\u4f18\u5316\u9884\u5904\u7406\u548c\u6a21\u578b\u8bad\u7ec3\u8fc7\u7a0b\uff0c\u4ee5\u83b7\u5f97\u66f4\u597d\u7684\u8bc6\u522b\u6548\u679c\u3002<\/p>\n<\/p>\n
\u76f8\u5173\u95ee\u7b54FAQs\uff1a<\/strong><\/h2>\n
 \u5982\u4f55\u4f7f\u7528Python\u8bc6\u522b\u56fe\u7247\u4e2d\u7684\u6570\u5b57\u548c\u5b57\u6bcd\uff1f<\/strong>
\u4f7f\u7528Python\u8bc6\u522b\u56fe\u7247\u4e2d\u7684\u6570\u5b57\u548c\u5b57\u6bcd\u901a\u5e38\u6d89\u53ca\u5230\u8ba1\u7b97\u673a\u89c6\u89c9\u548c\u56fe\u50cf\u5904\u7406\u6280\u672f\u3002\u53ef\u4ee5\u4f7f\u7528\u5e93\u5982OpenCV\u548cPytesseract\u6765\u5b8c\u6210\u8fd9\u4e2a\u4efb\u52a1\u3002OpenCV\u7528\u4e8e\u56fe\u50cf\u5904\u7406\uff0c\u800cPytesseract\u662f\u4e00\u4e2aOCR\uff08\u5149\u5b66\u5b57\u7b26\u8bc6\u522b\uff09\u5de5\u5177\uff0c\u53ef\u4ee5\u5c06\u56fe\u7247\u4e2d\u7684\u6587\u672c\u63d0\u53d6\u51fa\u6765\u3002\u9996\u5148\uff0c\u60a8\u9700\u8981\u5b89\u88c5\u8fd9\u4e24\u4e2a\u5e93\uff0c\u5e76\u51c6\u5907\u597d\u5f85\u5904\u7406\u7684\u56fe\u7247\uff0c\u7136\u540e\u901a\u8fc7\u7f16\u5199\u4ee3\u7801\u52a0\u8f7d\u56fe\u7247\u5e76\u8fdb\u884c\u5904\u7406\uff0c\u6700\u540e\u63d0\u53d6\u51fa\u56fe\u50cf\u4e2d\u7684\u6570\u5b57\u548c\u5b57\u6bcd\u3002<\/p>\n
\u8bc6\u522b\u56fe\u7247\u4e2d\u6587\u5b57\u65f6\uff0c\u9700\u6ce8\u610f\u54ea\u4e9b\u56e0\u7d20\uff1f<\/strong>
\u5728\u4f7f\u7528Python\u8fdb\u884c\u6587\u5b57\u8bc6\u522b\u65f6\uff0c\u9700\u8981\u8003\u8651\u591a\u4e2a\u56e0\u7d20\uff0c\u4f8b\u5982\u56fe\u7247\u7684\u8d28\u91cf\u3001\u6587\u5b57\u7684\u6e05\u6670\u5ea6\u548c\u5b57\u4f53\u7684\u6837\u5f0f\u3002\u9ad8\u8d28\u91cf\u7684\u56fe\u7247\u548c\u6e05\u6670\u7684\u6587\u5b57\u53ef\u4ee5\u663e\u8457\u63d0\u9ad8\u8bc6\u522b\u7684\u51c6\u786e\u6027\u3002\u6b64\u5916\uff0c\u6587\u5b57\u7684\u5927\u5c0f\u3001\u989c\u8272\u5bf9\u6bd4\u5ea6\u4ee5\u53ca\u80cc\u666f\u7684\u590d\u6742\u7a0b\u5ea6\u4e5f\u4f1a\u5f71\u54cdOCR\u7684\u6548\u679c\u3002\u4e3a\u4e86\u63d0\u9ad8\u8bc6\u522b\u7387\uff0c\u53ef\u4ee5\u5c1d\u8bd5\u5bf9\u56fe\u7247\u8fdb\u884c\u9884\u5904\u7406\uff0c\u5982\u8c03\u6574\u5bf9\u6bd4\u5ea6\u3001\u53bb\u566a\u58f0\u3001\u4e8c\u503c\u5316\u7b49\u3002<\/p>\n
\u9664\u4e86Pytesseract\uff0c\u8fd8\u6709\u5176\u4ed6\u54ea\u4e9b\u5e93\u53ef\u4ee5\u7528\u4e8e\u6587\u5b57\u8bc6\u522b\uff1f<\/strong>
\u9664\u4e86Pytesseract\uff0cPython\u4e2d\u8fd8\u6709\u5176\u4ed6\u4e00\u4e9b\u5e93\u53ef\u4ee5\u7528\u4e8e\u6587\u5b57\u8bc6\u522b\u3002\u4f8b\u5982\uff0cEasyOCR\u662f\u4e00\u4e2a\u652f\u6301\u591a\u79cd\u8bed\u8a00\u7684OCR\u5e93\uff0c\u5177\u6709\u8f83\u597d\u7684\u8bc6\u522b\u7387\u548c\u901f\u5ea6\u3002\u6b64\u5916\uff0c\u8fd8\u6709Keras-OCR\uff0c\u5b83\u7ed3\u5408\u4e86\u6df1\u5ea6\u5b66\u4e60\u6280\u672f\uff0c\u80fd\u591f\u5904\u7406\u66f4\u590d\u6742\u7684\u573a\u666f\u6587\u672c\u8bc6\u522b\u3002\u9009\u62e9\u5408\u9002\u7684\u5e93\u65f6\uff0c\u53ef\u4ee5\u6839\u636e\u81ea\u5df1\u7684\u9700\u6c42\u548c\u9879\u76ee\u590d\u6742\u5ea6\u8fdb\u884c\u6bd4\u8f83\u548c\u9009\u62e9\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"\u5728Python\u4e2d\u8bc6\u522b\u56fe\u7247\u4e0a\u7684\u6570\u5b57\u5b57\u6bcd\uff0c\u4e3b\u8981\u53ef\u4ee5\u901a\u8fc7\u4ee5\u4e0b\u51e0\u79cd\u65b9\u6cd5\uff1a\u4f7f\u7528OCR\u5e93\uff08\u5982Tesseract\uff09\u3001\u6df1\u5ea6\u5b66\u4e60 […]","protected":false},"author":3,"featured_media":1101896,"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\/1101889"}],"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=1101889"}],"version-history":[{"count":"1","href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts\/1101889\/revisions"}],"predecessor-version":[{"id":1101897,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts\/1101889\/revisions\/1101897"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/media\/1101896"}],"wp:attachment":[{"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/media?parent=1101889"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/categories?post=1101889"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/tags?post=1101889"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}