{"id":1006142,"date":"2024-12-27T10:41:44","date_gmt":"2024-12-27T02:41:44","guid":{"rendered":"https:\/\/docs.pingcode.com\/ask\/ask-ask\/1006142.html"},"modified":"2024-12-27T10:41:46","modified_gmt":"2024-12-27T02:41:46","slug":"python%e5%a6%82%e4%bd%95%e7%bc%96%e7%a8%8b%e6%8a%a0%e5%9f%ba%e7%ba%bf","status":"publish","type":"post","link":"https:\/\/docs.pingcode.com\/ask\/1006142.html","title":{"rendered":"python\u5982\u4f55\u7f16\u7a0b\u62a0\u57fa\u7ebf"},"content":{"rendered":"

\"python\u5982\u4f55\u7f16\u7a0b\u62a0\u57fa\u7ebf\"<\/p>\n

\u5728Python\u4e2d\u8fdb\u884c\u57fa\u7ebf\u62a0\u56fe\u4e3b\u8981\u901a\u8fc7\u56fe\u50cf\u5904\u7406\u5e93\u6765\u5b9e\u73b0\uff0c\u5982OpenCV\u3001Pillow\u3001Scikit-Image\u7b49\u3002\u62a0\u57fa\u7ebf\u7684\u6b65\u9aa4\u901a\u5e38\u5305\u62ec\u56fe\u50cf\u8bfb\u53d6\u3001\u9884\u5904\u7406\uff08\u5982\u7070\u5ea6\u5316\u3001\u4e8c\u503c\u5316\uff09\u3001\u57fa\u7ebf\u68c0\u6d4b\u548c\u5206\u5272\u3001\u540e\u5904\u7406\u7b49\u3002\u53ef\u4ee5\u5229\u7528OpenCV\u8fdb\u884c\u56fe\u50cf\u8bfb\u53d6\u4e0e\u9884\u5904\u7406\u3001\u901a\u8fc7Scikit-Image\u8fdb\u884c\u57fa\u7ebf\u68c0\u6d4b\u548c\u5206\u5272\u3002<\/strong><\/p>\n<\/p>\n

\u5728\u672c\u6587\u4e2d\uff0c\u6211\u4eec\u5c06\u8be6\u7ec6\u63a2\u8ba8\u5982\u4f55\u5728Python\u4e2d\u7f16\u7a0b\u5b9e\u73b0\u57fa\u7ebf\u62a0\u56fe\u3002\u57fa\u7ebf\u62a0\u56fe\u662f\u4e00\u79cd\u5e38\u89c1\u7684\u56fe\u50cf\u5904\u7406\u6280\u672f\uff0c\u7528\u4e8e\u4ece\u56fe\u50cf\u4e2d\u63d0\u53d6\u7279\u5b9a\u7684\u57fa\u7ebf\u4fe1\u606f\u3002\u901a\u8fc7\u4ee5\u4e0b\u51e0\u4e2a\u6b65\u9aa4\uff0c\u6211\u4eec\u53ef\u4ee5\u9ad8\u6548\u5730\u5b8c\u6210\u8fd9\u4e00\u4efb\u52a1\u3002<\/p>\n<\/p>\n

\u4e00\u3001\u56fe\u50cf\u8bfb\u53d6\u4e0e\u9884\u5904\u7406<\/h2>\n<\/p>\n

\u5728\u8fdb\u884c\u57fa\u7ebf\u62a0\u56fe\u4e4b\u524d\uff0c\u6211\u4eec\u9996\u5148\u9700\u8981\u8bfb\u53d6\u56fe\u50cf\u5e76\u8fdb\u884c\u9884\u5904\u7406\u3002\u9884\u5904\u7406\u7684\u6b65\u9aa4\u5305\u62ec\u56fe\u50cf\u7070\u5ea6\u5316\u3001\u53bb\u566a\u3001\u589e\u5f3a\u5bf9\u6bd4\u5ea6\u7b49\u3002<\/p>\n<\/p>\n

1. \u56fe\u50cf\u8bfb\u53d6<\/h3>\n<\/p>\n

\u4f7f\u7528OpenCV\u6216Pillow\u5e93\u53ef\u4ee5\u8f7b\u677e\u8bfb\u53d6\u56fe\u50cf\u3002OpenCV\u63d0\u4f9b\u4e86\u66f4\u591a\u7684\u56fe\u50cf\u5904\u7406\u529f\u80fd\uff0c\u800cPillow\u5219\u66f4\u7b80\u5355\u6613\u7528\u3002<\/p>\n<\/p>\n

import cv2<\/p>\n
import numpy as np<\/p>\n
\u8bfb\u53d6\u56fe\u50cf<\/strong><\/h2>\n
image = cv2.imread('path\/to\/image.jpg')<\/p>\n
<\/code><\/pre>\n<\/p>\n
2. \u56fe\u50cf\u7070\u5ea6\u5316<\/h3>\n<\/p>\n
\u7070\u5ea6\u5316\u662f\u5c06\u5f69\u8272\u56fe\u50cf\u8f6c\u6362\u4e3a\u7070\u5ea6\u56fe\u50cf\u7684\u8fc7\u7a0b\uff0c\u8fd9\u6709\u52a9\u4e8e\u51cf\u5c11\u8ba1\u7b97\u590d\u6742\u5ea6\u3002<\/p>\n<\/p>\n
# \u7070\u5ea6\u5316<\/p>\n
gray_image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)<\/p>\n
<\/code><\/pre>\n<\/p>\n
3. \u53bb\u566a\u4e0e\u589e\u5f3a\u5bf9\u6bd4\u5ea6<\/h3>\n<\/p>\n
\u4e3a\u4e86\u63d0\u9ad8\u57fa\u7ebf\u68c0\u6d4b\u7684\u51c6\u786e\u6027\uff0c\u6211\u4eec\u901a\u5e38\u9700\u8981\u5bf9\u56fe\u50cf\u8fdb\u884c\u53bb\u566a\u548c\u589e\u5f3a\u5bf9\u6bd4\u5ea6\u7684\u5904\u7406\u3002<\/p>\n<\/p>\n
# \u9ad8\u65af\u6a21\u7cca\u53bb\u566a<\/p>\n
blurred_image = cv2.GaussianBlur(gray_image, (5, 5), 0)<\/p>\n
\u81ea\u9002\u5e94\u76f4\u65b9\u56fe\u5747\u8861\u5316\u589e\u5f3a\u5bf9\u6bd4\u5ea6<\/strong><\/h2>\n
clahe = cv2.createCLAHE(clipLimit=2.0, tileGridSize=(8, 8))<\/p>\n
enhanced_image = clahe.apply(blurred_image)<\/p>\n
<\/code><\/pre>\n<\/p>\n
\u4e8c\u3001\u57fa\u7ebf\u68c0\u6d4b<\/h2>\n<\/p>\n
\u57fa\u7ebf\u68c0\u6d4b\u662f\u4ece\u56fe\u50cf\u4e2d\u63d0\u53d6\u51fa\u57fa\u7ebf\u7684\u8fc7\u7a0b\uff0c\u901a\u5e38\u901a\u8fc7\u8fb9\u7f18\u68c0\u6d4b\u548c\u5f62\u6001\u5b66\u53d8\u6362\u7b49\u6280\u672f\u5b9e\u73b0\u3002<\/p>\n<\/p>\n
1. \u8fb9\u7f18\u68c0\u6d4b<\/h3>\n<\/p>\n
\u4f7f\u7528Canny\u8fb9\u7f18\u68c0\u6d4b\u53ef\u4ee5\u6709\u6548\u5730\u8bc6\u522b\u56fe\u50cf\u4e2d\u7684\u8fb9\u7f18\u3002<\/p>\n<\/p>\n
# Canny\u8fb9\u7f18\u68c0\u6d4b<\/p>\n
edges = cv2.Canny(enhanced_image, 100, 200)<\/p>\n
<\/code><\/pre>\n<\/p>\n
2. \u5f62\u6001\u5b66\u53d8\u6362<\/h3>\n<\/p>\n
\u901a\u8fc7\u5f62\u6001\u5b66\u53d8\u6362\uff0c\u6211\u4eec\u53ef\u4ee5\u8fde\u63a5\u65ad\u88c2\u7684\u8fb9\u7f18\u5e76\u53bb\u9664\u566a\u58f0\u3002<\/p>\n<\/p>\n
# \u5f62\u6001\u5b66\u53d8\u6362<\/p>\n
kernel = np.ones((5, 5), np.uint8)<\/p>\n
dilated_edges = cv2.dilate(edges, kernel, iterations=1)<\/p>\n
<\/code><\/pre>\n<\/p>\n
\u4e09\u3001\u57fa\u7ebf\u5206\u5272<\/h2>\n<\/p>\n
\u57fa\u7ebf\u5206\u5272\u662f\u5c06\u68c0\u6d4b\u5230\u7684\u57fa\u7ebf\u4ece\u56fe\u50cf\u4e2d\u5206\u5272\u51fa\u6765\u7684\u8fc7\u7a0b\u3002<\/p>\n<\/p>\n
1. \u8f6e\u5ed3\u68c0\u6d4b<\/h3>\n<\/p>\n
\u901a\u8fc7\u8f6e\u5ed3\u68c0\u6d4b\uff0c\u53ef\u4ee5\u8bc6\u522b\u56fe\u50cf\u4e2d\u7684\u95ed\u5408\u533a\u57df\u3002<\/p>\n<\/p>\n
# \u8f6e\u5ed3\u68c0\u6d4b<\/p>\n
contours, _ = cv2.findContours(dilated_edges, cv2.RETR_TREE, cv2.CHAI<\/a>N_APPROX_SIMPLE)<\/p>\n
<\/code><\/pre>\n<\/p>\n
2. \u5206\u5272\u57fa\u7ebf<\/h3>\n<\/p>\n
\u901a\u8fc7\u5206\u6790\u8f6e\u5ed3\uff0c\u6211\u4eec\u53ef\u4ee5\u5206\u5272\u51fa\u57fa\u7ebf\u3002<\/p>\n<\/p>\n
# \u5206\u5272\u57fa\u7ebf<\/p>\n
for contour in contours:<\/p>\n
    if cv2.contourArea(contour) > 100:  # \u8fc7\u6ee4\u6389\u5c0f\u9762\u79ef\u7684\u8f6e\u5ed3<\/p>\n
        x, y, w, h = cv2.boundingRect(contour)<\/p>\n
        # \u63d0\u53d6\u57fa\u7ebf\u533a\u57df<\/p>\n
        baseline_region = image[y:y+h, x:x+w]<\/p>\n
        # \u4fdd\u5b58\u6216\u5904\u7406\u57fa\u7ebf\u533a\u57df<\/p>\n
<\/code><\/pre>\n<\/p>\n
\u56db\u3001\u540e\u5904\u7406<\/h2>\n<\/p>\n
\u5728\u5b8c\u6210\u57fa\u7ebf\u5206\u5272\u4e4b\u540e\uff0c\u6211\u4eec\u53ef\u80fd\u8fd8\u9700\u8981\u8fdb\u884c\u4e00\u4e9b\u540e\u5904\u7406\u64cd\u4f5c\uff0c\u6bd4\u5982\u53bb\u9664\u80cc\u666f\u3001\u8c03\u6574\u989c\u8272\u7b49\u3002<\/p>\n<\/p>\n
1. \u53bb\u9664\u80cc\u666f<\/h3>\n<\/p>\n
\u901a\u8fc7\u63a9\u819c\u64cd\u4f5c\uff0c\u6211\u4eec\u53ef\u4ee5\u53bb\u9664\u57fa\u7ebf\u533a\u57df\u7684\u80cc\u666f\u3002<\/p>\n<\/p>\n
# \u521b\u5efa\u63a9\u819c<\/p>\n
mask = np.zeros_like(image)<\/p>\n
cv2.drawContours(mask, contours, -1, (255, 255, 255), -1)<\/p>\n
\u5e94\u7528\u63a9\u819c<\/strong><\/h2>\n
baseline_only = cv2.bitwise_and(image, mask)<\/p>\n
<\/code><\/pre>\n<\/p>\n
2. \u8c03\u6574\u989c\u8272<\/h3>\n<\/p>\n
\u5982\u679c\u9700\u8981\u8c03\u6574\u57fa\u7ebf\u533a\u57df\u7684\u989c\u8272\uff0c\u53ef\u4ee5\u4f7f\u7528\u8272\u5f69\u53d8\u6362\u3002<\/p>\n<\/p>\n
# \u5c06\u57fa\u7ebf\u533a\u57df\u8f6c\u6362\u4e3a\u7070\u5ea6<\/p>\n
baseline_gray = cv2.cvtColor(baseline_only, cv2.COLOR_BGR2GRAY)<\/p>\n
\u5e94\u7528\u4f3d\u9a6c\u6821\u6b63<\/strong><\/h2>\n
gamma = 1.2<\/p>\n
look_up_table = np.array([((i \/ 255.0)  gamma) * 255 for i in range(256)], dtype="uint8")<\/p>\n
baseline_corrected = cv2.LUT(baseline_gray, look_up_table)<\/p>\n
<\/code><\/pre>\n<\/p>\n
\u4e94\u3001\u603b\u7ed3\u4e0e\u5e94\u7528<\/h2>\n<\/p>\n
\u57fa\u7ebf\u62a0\u56fe\u662f\u4e00\u9879\u590d\u6742\u4f46\u975e\u5e38\u6709\u7528\u7684\u6280\u672f\uff0c\u5e7f\u6cdb\u5e94\u7528\u4e8e\u6587\u6863\u5206\u6790\u3001\u533b\u5b66\u5f71\u50cf\u5904\u7406\u7b49\u9886\u57df\u3002\u901a\u8fc7\u4e0a\u8ff0\u6b65\u9aa4\uff0c\u6211\u4eec\u53ef\u4ee5\u5728Python\u4e2d\u5b9e\u73b0\u9ad8\u6548\u7684\u57fa\u7ebf\u62a0\u56fe\u3002\u6839\u636e\u5177\u4f53\u7684\u5e94\u7528\u573a\u666f\uff0c\u53ef\u4ee5\u5bf9\u6bcf\u4e2a\u6b65\u9aa4\u8fdb\u884c\u8c03\u6574\u548c\u4f18\u5316\uff0c\u4ee5\u8fbe\u5230\u6700\u4f73\u6548\u679c\u3002<\/p>\n<\/p>\n
\u5728\u5b9e\u9645\u5e94\u7528\u4e2d\uff0c\u8fd8\u53ef\u4ee5\u7ed3\u5408\u673a\u5668\u5b66\u4e60<\/a>\u6280\u672f\uff0c\u5bf9\u57fa\u7ebf\u8fdb\u884c\u66f4\u667a\u80fd\u7684\u8bc6\u522b\u4e0e\u5904\u7406\u3002\u4f8b\u5982\uff0c\u4f7f\u7528\u5377\u79ef\u795e\u7ecf\u7f51\u7edc\uff08CNN\uff09\u8fdb\u884c\u57fa\u7ebf\u68c0\u6d4b\uff0c\u53ef\u4ee5\u63d0\u9ad8\u68c0\u6d4b\u7684\u51c6\u786e\u6027\u548c\u9c81\u68d2\u6027\u3002\u901a\u8fc7\u4e0d\u65ad\u7684\u5b9e\u9a8c\u548c\u4f18\u5316\uff0c\u53ef\u4ee5\u8fdb\u4e00\u6b65\u63d0\u5347\u57fa\u7ebf\u62a0\u56fe\u7684\u6548\u679c\u548c\u6548\u7387\u3002<\/p>\n<\/p>\n
\u76f8\u5173\u95ee\u7b54FAQs\uff1a<\/strong><\/h2>\n
 \u62a0\u57fa\u7ebf\u5728Python\u7f16\u7a0b\u4e2d\u6709\u4ec0\u4e48\u5177\u4f53\u5e94\u7528\uff1f<\/strong>
\u62a0\u57fa\u7ebf\u901a\u5e38\u7528\u4e8e\u56fe\u50cf\u5904\u7406\u548c\u6570\u636e\u5206\u6790\u9886\u57df\u3002\u5728\u56fe\u50cf\u5904\u7406\u4e2d\uff0c\u62a0\u57fa\u7ebf\u5e2e\u52a9\u63d0\u53d6\u56fe\u50cf\u7684\u4e3b\u8981\u8f6e\u5ed3\u6216\u80cc\u666f\uff0c\u4ee5\u4fbf\u8fdb\u884c\u8fdb\u4e00\u6b65\u7684\u5206\u6790\uff0c\u4f8b\u5982\u8bc6\u522b\u548c\u5206\u5272\u5bf9\u8c61\u3002\u5728\u6570\u636e\u5206\u6790\u4e2d\uff0c\u62a0\u57fa\u7ebf\u53ef\u4ee5\u7528\u4e8e\u5e73\u6ed1\u66f2\u7ebf\u3001\u53bb\u9664\u566a\u58f0\uff0c\u5e76\u5e2e\u52a9\u8bc6\u522b\u6570\u636e\u7684\u8d8b\u52bf\u3002\u5728Python\u4e2d\uff0c\u53ef\u4ee5\u4f7f\u7528\u50cfOpenCV\u548cNumPy\u8fd9\u6837\u7684\u5e93\u6765\u5b9e\u73b0\u8fd9\u4e9b\u529f\u80fd\u3002<\/p>\n
\u5982\u4f55\u5728Python\u4e2d\u4f7f\u7528\u5e93\u5b9e\u73b0\u62a0\u57fa\u7ebf\u7684\u529f\u80fd\uff1f<\/strong>
\u5728Python\u4e2d\uff0cOpenCV\u662f\u4e00\u4e2a\u5f3a\u5927\u7684\u56fe\u50cf\u5904\u7406\u5e93\uff0c\u53ef\u4ee5\u901a\u8fc7\u4e00\u7cfb\u5217\u56fe\u50cf\u5904\u7406\u51fd\u6570\u6765\u5b9e\u73b0\u62a0\u57fa\u7ebf\u3002\u4f8b\u5982\uff0c\u53ef\u4ee5\u4f7f\u7528\u8fb9\u7f18\u68c0\u6d4b\u7b97\u6cd5\uff08\u5982Canny\u8fb9\u7f18\u68c0\u6d4b\uff09\u6765\u8bc6\u522b\u56fe\u50cf\u7684\u57fa\u7ebf\u3002\u6b64\u5916\uff0cSciPy\u548cMatplotlib\u7b49\u5e93\u4e5f\u53ef\u4ee5\u7528\u4e8e\u5904\u7406\u6570\u636e\u96c6\u7684\u57fa\u7ebf\uff0c\u901a\u8fc7\u62df\u5408\u66f2\u7ebf\u6216\u4f7f\u7528\u6ee4\u6ce2\u5668\u6765\u63d0\u53d6\u57fa\u7ebf\u3002<\/p>\n
\u62a0\u57fa\u7ebf\u7684\u5e38\u89c1\u7b97\u6cd5\u6709\u54ea\u4e9b\uff1f<\/strong>
\u5728Python\u7f16\u7a0b\u4e2d\uff0c\u6709\u591a\u79cd\u7b97\u6cd5\u53ef\u4ee5\u7528\u4e8e\u62a0\u57fa\u7ebf\u3002\u5e38\u89c1\u7684\u7b97\u6cd5\u5305\u62ec\u6700\u5c0f\u4e8c\u4e58\u6cd5\u62df\u5408\u3001Savitzky-Golay\u6ee4\u6ce2\u5668\u548c\u5c0f\u6ce2\u53d8\u6362\u7b49\u3002\u8fd9\u4e9b\u7b97\u6cd5\u5404\u6709\u4f18\u7f3a\u70b9\uff0c\u9002\u7528\u4e8e\u4e0d\u540c\u7684\u6570\u636e\u7279\u5f81\u548c\u5904\u7406\u9700\u6c42\u3002\u9009\u62e9\u5408\u9002\u7684\u7b97\u6cd5\u53ef\u4ee5\u663e\u8457\u63d0\u9ad8\u62a0\u57fa\u7ebf\u7684\u6548\u679c\u548c\u51c6\u786e\u6027\u3002<\/p>\n
\u4f7f\u7528Python\u62a0\u57fa\u7ebf\u65f6\u9700\u8981\u6ce8\u610f\u54ea\u4e9b\u95ee\u9898\uff1f<\/strong>
\u5728\u8fdb\u884c\u62a0\u57fa\u7ebf\u5904\u7406\u65f6\uff0c\u6570\u636e\u7684\u9884\u5904\u7406\u975e\u5e38\u5173\u952e\u3002\u566a\u58f0\u3001\u5f02\u5e38\u503c\u548c\u6570\u636e\u7684\u5206\u5e03\u7279\u5f81\u90fd\u4f1a\u5f71\u54cd\u62a0\u57fa\u7ebf\u7684\u7ed3\u679c\u3002\u6b64\u5916\uff0c\u9009\u62e9\u5408\u9002\u7684\u53c2\u6570\u548c\u7b97\u6cd5\u4e5f\u662f\u6210\u529f\u7684\u5173\u952e\uff0c\u8fc7\u5ea6\u62df\u5408\u6216\u6b20\u62df\u5408\u90fd\u53ef\u80fd\u5bfc\u81f4\u7ed3\u679c\u4e0d\u7406\u60f3\u3002\u5efa\u8bae\u5728\u5904\u7406\u524d\u8fdb\u884c\u6570\u636e\u53ef\u89c6\u5316\uff0c\u4ee5\u4fbf\u66f4\u597d\u5730\u7406\u89e3\u6570\u636e\u7684\u7279\u5f81\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"\u5728Python\u4e2d\u8fdb\u884c\u57fa\u7ebf\u62a0\u56fe\u4e3b\u8981\u901a\u8fc7\u56fe\u50cf\u5904\u7406\u5e93\u6765\u5b9e\u73b0\uff0c\u5982OpenCV\u3001Pillow\u3001Scikit-Image\u7b49 […]","protected":false},"author":3,"featured_media":1006148,"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\/1006142"}],"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=1006142"}],"version-history":[{"count":"1","href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts\/1006142\/revisions"}],"predecessor-version":[{"id":1006150,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts\/1006142\/revisions\/1006150"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/media\/1006148"}],"wp:attachment":[{"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/media?parent=1006142"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/categories?post=1006142"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/tags?post=1006142"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}