{"id":1938,"date":"2022-01-13T15:39:14","date_gmt":"2022-01-13T07:39:14","guid":{"rendered":"http:\/\/139.9.1.231\/?p=1938"},"modified":"2022-01-13T21:49:17","modified_gmt":"2022-01-13T13:49:17","slug":"fftandwatershed","status":"publish","type":"post","link":"http:\/\/139.9.1.231\/index.php\/2022\/01\/13\/fftandwatershed\/","title":{"rendered":"\u56fe\u50cf\u5085\u91cc\u53f6\u53d8\u6362 and\u56fe\u50cf\u5206\u5272\u7684\u7ecf\u5178\u7b97\u6cd5\uff1a\u5206\u6c34\u5cad\u7b97\u6cd5"},"content":{"rendered":"\n

\u6458\u81ea\u7f51\u7edc\uff0c\u4fb5\u6743\u7acb\u5220\u3002<\/p>\n\n\n\n

\u5085\u91cc\u53f6\u53d8\u6362\u7684\u4f5c\u7528<\/strong>\uff1a<\/p>\n\n\n\n

\u3000\u3000\u5bf9\u4e8e\u6570\u5b57\u56fe\u50cf\u8fd9\u79cd\u79bb\u6563\u7684\u4fe1\u53f7\uff0c\u9891\u7387\u5927\u5c0f\u8868\u793a\u4fe1\u53f7\u53d8\u6362\u7684\u5267\u70c8\u7a0b\u5ea6\u6216\u8005\u8bf4\u4fe1\u53f7\u53d8\u5316\u7684\u5feb\u6162\u3002\u9891\u7387\u8d8a\u5927\uff0c\u53d8\u6362\u8d8a\u5267\u70c8\uff0c\u9891\u7387\u8d8a\u5c0f\uff0c\u4fe1\u53f7\u8d8a\u5e73\u7f13\uff0c\u5bf9\u5e94\u5230\u7684\u56fe\u50cf\u4e2d\uff0c\u9ad8\u9891\u4fe1\u53f7\u5f80\u5f80\u662f\u56fe\u50cf\u4e2d\u7684\u8fb9\u7f18\u4fe1\u53f7\u548c\u566a\u58f0\u4fe1\u53f7\uff0c\u800c\u4f4e\u9891\u4fe1\u53f7\u5305\u542b\u56fe\u50cf\u53d8\u5316\u9891\u7e41\u7684\u56fe\u50cf\u8f6e\u5ed3\u53ca\u80cc\u666f\u706f\u4fe1\u53f7\u3002<\/p>\n\n\n\n

\u3000\u3000\u9700\u8981\u8bf4\u660e\u7684\u662f\uff1a\u5085\u91cc\u53f6\u53d8\u6362\u5f97\u5230\u7684\u9891\u8c31\u56fe\u4e0a\u7684\u70b9\u4e0e\u539f\u56fe\u50cf\u4e0a\u7684\u70b9\u4e4b\u95f4\u4e0d\u5b58\u5728\u4e00\u4e00\u5bf9\u5e94\u7684\u5173\u7cfb\u3002<\/p>\n\n\n\n

1\uff0c\u56fe\u50cf\u7ecf\u8fc7\u4e8c\u7ef4\u5085\u91cc\u53f6\u7684\u53d8\u6362\u540e\uff0c\u5176\u53d8\u6362\u7a00\u758f\u77e9\u9635\u5177\u6709\u5982\u4e0b\u6027\u8d28\uff1a\u82e5\u4ea4\u6362\u77e9\u9635\u539f\u70b9\u8bbe\u5728\u4e2d\u5fc3\uff0c\u5176\u9891\u8c31\u80fd\u91cf\u96c6\u4e2d\u5206\u5e03\u5728\u53d8\u6362\u7a00\u758f\u77e9\u9635\u7684\u4e2d\u5fc3\u9644\u8fd1\u3002\u82e5\u6240\u7528\u7684\u4e8c\u7ef4\u5085\u91cc\u53f6\u53d8\u6362\u77e9\u9635\u7684\u4e91\u5dc5\u8bbe\u5728\u5de6\u4e0a\u89d2\uff0c\u90a3\u4e48\u56fe\u50cf\u4fe1\u53f7\u80fd\u91cf\u5c06\u96c6\u4e2d\u5728\u7cfb\u6570\u77e9\u9635\u7684\u56db\u4e2a\u89d2\u4e0a\u3002\u8fd9\u662f\u7531\u4e8c\u7ef4\u5085\u91cc\u53f6\u53d8\u6362\u672c\u8eab\u6027\u8d28\u51b3\u5b9a\u7684\u3002\u540c\u65f6\u4e5f\u8868\u660e\u4e00\u80a1\u56fe\u50cf\u80fd\u91cf\u96c6\u4e2d\u4f4e\u9891\u533a\u57df\u3002<\/p>\n\n\n\n

2\uff0c\u56fe\u50cf\u7070\u5ea6\u53d8\u5316\u7f13\u6162\u7684\u533a\u57df\uff0c\u5bf9\u5e94\u5b83\u53d8\u6362\u540e\u7684\u4f4e\u9891\u5206\u91cf\u90e8\u5206\uff1b\u56fe\u50cf\u7070\u5ea6\u5448\u9636\u8dc3\u53d8\u5316\u7684\u533a\u57df\uff0c\u5bf9\u5e94\u53d8\u6362\u540e\u7684\u9ad8\u9891\u5206\u91cf\u90e8\u5206\u3002\u9664\u9897\u7c92\u566a\u97f3\u5916\uff0c\u56fe\u50cf\u7ec6\u8282\u7684\u8fb9\u7f18\uff0c\u8f6e\u5ed3\u5904\u90fd\u662f\u7070\u5ea6\u53d8\u6362\u7a81\u53d8\u533a\u57df\uff0c\u4ed6\u4eec\u90fd\u5177\u6709\u53d8\u6362\u540e\u7684\u9ad8\u9891\u5206\u91cf\u7279\u5f81\u3002<\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

\u53f3\u8fb9\u4e3a\u9891\u7387\u5206\u5e03\u56fe\u8c31\uff0c\u5176\u4e2d\u8d8a\u9760\u8fd1\u4e2d\u5fc3\u4f4d\u7f6e\u9891\u7387\u8d8a\u4f4e\uff0c\u8d8a\u4eae\uff08\u7070\u5ea6\u503c\u8d8a\u9ad8\uff09\u7684\u4f4d\u7f6e\u4ee3\u8868\u8be5\u9891\u7387\u7684\u4fe1\u606f\u632f\u5e45\u8d8a\u5927\u3002<\/p>\n\n\n\n

\u53f3\u8fb9\u56fe\u4e2d\uff0c\u6bcf\u4e00\u4e2a\u70b9<\/strong>\uff1a<\/p>\n\n\n\n

1\uff09\u5b83\u5230\u4e2d\u70b9\u7684\u8ddd\u79bb\u63cf\u8ff0\u7684\u662f\u9891\u7387<\/p>\n\n\n\n

2\uff09\u4e2d\u70b9\u5230\u5b83\u7684\u65b9\u5411\uff0c\u662f\u5e73\u9762\u6ce2\u7684\u65b9\u5411<\/p>\n\n\n\n

3\uff09\u90a3\u4e00\u70b9\u7684\u7070\u5ea6\u503c\u63cf\u8ff0\u7684\u662f\u5b83\u7684\u5e45\u503c<\/p>\n\n\n\n

OpenCV\u4e2d\u7684\u5085\u91cc\u53f6\u53d8\u6362<\/a><\/h3>\n\n\n\n

OpenCV\u4e3a\u6b64\u63d0\u4f9b\u4e86**cv.dft**()\u548c**cv.idft**()\u51fd\u6570\u3002\u5b83\u8fd4\u56de\u4e0e\u524d\u4e00\u4e2a\u76f8\u540c\u7684\u7ed3\u679c\uff0c\u4f46\u662f\u6709\u4e24\u4e2a\u901a\u9053\u3002\u7b2c\u4e00\u4e2a\u901a\u9053\u662f\u7ed3\u679c\u7684\u5b9e\u90e8\uff0c\u7b2c\u4e8c\u4e2a\u901a\u9053\u662f\u7ed3\u679c\u7684\u865a\u90e8\u3002\u8f93\u5165\u56fe\u50cf\u9996\u5148\u5e94\u8f6c\u6362\u4e3anp.float32<\/code>\u3002\u6211\u4eec\u6765\u770b\u770b\u600e\u4e48\u505a\u3002<\/p>\n\n\n\n

import<\/strong> numpy as<\/strong> np\nimport<\/strong> cv2 as<\/strong> cv\nfrom<\/strong> matplotlib import<\/strong> pyplot as<\/strong> plt\nimg = cv.imread('messi5.jpg',0)\ndft = cv.dft(np.float32(img),flags = cv.DFT_COMPLEX_OUTPUT)\ndft_shift = np.fft.fftshift(dft)\nmagnitude_spectrum = 20*np.log(cv.magnitude(dft_shift[:,:,0],dft_shift[:,:,1]))\nplt.subplot(121),plt.imshow(img, cmap = 'gray')\nplt.title('Input Image'), plt.xticks([]), plt.yticks([])\nplt.subplot(122),plt.imshow(magnitude_spectrum, cmap = 'gray')\nplt.title('Magnitude Spectrum'), plt.xticks([]), plt.yticks([])\nplt.show()\n\n\u73b0\u5728\u6211\u4eec\u8981\u505aDFT\u7684\u9006\u53d8\u6362\u3002\u5728\u4e0a\u4e00\u8282\u4e2d\uff0c\u6211\u4eec\u521b\u5efa\u4e86\u4e00\u4e2aHPF\uff0c\u8fd9\u6b21\u6211\u4eec\u5c06\u770b\u5230\u5982\u4f55\u5220\u9664\u56fe\u50cf\u4e2d\u7684\u9ad8\u9891\u5185\u5bb9\uff0c\u5373\u6211\u4eec\u5c06LPF\u5e94\u7528\u5230\u56fe\u50cf\u4e2d\u3002\u5b83\u5b9e\u9645\u4e0a\u6a21\u7cca\u4e86\u56fe\u50cf\u3002\u4e3a\u6b64\uff0c\u6211\u4eec\u9996\u5148\u521b\u5efa\u4e00\u4e2a\u9ad8\u503c(1)\u5728\u4f4e\u9891\u90e8\u5206\uff0c\u5373\u6211\u4eec\u8fc7\u6ee4\u4f4e\u9891\u5185\u5bb9\uff0c0\u5728\u9ad8\u9891\u533a\u3002\n\nrows, cols = img.shape\ncrow,ccol = rows\/2 , cols\/2\n# \u9996\u5148\u521b\u5efa\u4e00\u4e2a\u63a9\u7801\uff0c\u4e2d\u5fc3\u6b63\u65b9\u5f62\u4e3a1\uff0c\u5176\u4f59\u5168\u4e3a\u96f6\nmask = np.zeros((rows,cols,2),np.uint8)\nmask[crow-30:crow+30, ccol-30:ccol+30] = 1\n# \u5e94\u7528\u63a9\u7801\u548c\u9006DFT\nfshift = dft_shift*mask\nf_ishift = np.fft.ifftshift(fshift)\nimg_back = cv.idft(f_ishift)\nimg_back = cv.magnitude(img_back[:,:,0],img_back[:,:,1])\nplt.subplot(121),plt.imshow(img, cmap = 'gray')\nplt.title('Input Image'), plt.xticks([]), plt.yticks([])\nplt.subplot(122),plt.imshow(img_back, cmap = 'gray')\nplt.title('Magnitude Spectrum'), plt.xticks([]), plt.yticks([])\nplt.show()<\/code><\/pre>\n\n\n\n
\u5206\u6c34\u5cad\u7b97\u6cd5<\/h2>\n\n\n\n
\u56fe\u50cf\u7684\u7070\u5ea6\u7a7a\u95f4\u5f88\u50cf\u5730\u7403\u8868\u9762\u7684\u6574\u4e2a\u5730\u7406\u7ed3\u6784\uff0c\u6bcf\u4e2a\u50cf\u7d20\u7684\u7070\u5ea6\u503c\u4ee3\u8868\u9ad8\u5ea6\u3002\u5176\u4e2d\u7684\u7070\u5ea6\u503c\u8f83\u5927\u7684\u50cf\u7d20\u8fde\u6210\u7684\u7ebf\u53ef\u4ee5\u770b\u505a\u5c71\u810a\uff0c\u4e5f\u5c31\u662f\u5206\u6c34\u5cad\u3002\u5176\u4e2d\u7684\u6c34\u5c31\u662f\u7528\u4e8e\u4e8c\u503c\u5316\u7684gray threshold level\uff0c\u4e8c\u503c\u5316\u9608\u503c\u53ef\u4ee5\u7406\u89e3\u4e3a\u6c34\u5e73\u9762\uff0c\u6bd4\u6c34\u5e73\u9762\u4f4e\u7684\u533a\u57df\u4f1a\u88ab\u6df9\u6ca1\uff0c\u521a\u5f00\u59cb\u7528\u6c34\u586b\u5145\u6bcf\u4e2a\u5b64\u7acb\u7684\u5c71\u8c37(\u5c40\u90e8\u6700\u5c0f\u503c)\u3002<\/p>\n\n\n\n
\u5f53\u6c34\u5e73\u9762\u4e0a\u5347\u5230\u4e00\u5b9a\u9ad8\u5ea6\u65f6\uff0c\u6c34\u5c31\u4f1a\u6ea2\u51fa\u5f53\u524d\u5c71\u8c37\uff0c\u53ef\u4ee5\u901a\u8fc7\u5728\u5206\u6c34\u5cad\u4e0a\u4fee\u5927\u575d\uff0c\u4ece\u800c\u907f\u514d\u4e24\u4e2a\u5c71\u8c37\u7684\u6c34\u6c47\u96c6\uff0c\u8fd9\u6837\u56fe\u50cf\u5c31\u88ab\u5206\u62102\u4e2a\u50cf\u7d20\u96c6\uff0c\u4e00\u4e2a\u662f\u88ab\u6c34\u6df9\u6ca1\u7684\u5c71\u8c37\u50cf\u7d20\u96c6\uff0c\u4e00\u4e2a\u662f\u5206\u6c34\u5cad\u7ebf\u50cf\u7d20\u96c6\u3002\u6700\u7ec8\u8fd9\u4e9b\u5927\u575d\u5f62\u6210\u7684\u7ebf\u5c31\u5bf9\u6574\u4e2a\u56fe\u50cf\u8fdb\u884c\u4e86\u5206\u533a\uff0c\u5b9e\u73b0\u5bf9\u56fe\u50cf\u7684\u5206\u5272\u3002<\/p>\n\n\n\n
\"\"\/
\u56fe2<\/figcaption><\/figure>\n\n\n\n
\u5728\u8be5\u7b97\u6cd5\u4e2d\uff0c\u7a7a\u95f4\u4e0a\u76f8\u90bb\u5e76\u4e14\u7070\u5ea6\u503c\u76f8\u8fd1\u7684\u50cf\u7d20\u88ab\u5212\u5206\u4e3a\u4e00\u4e2a\u533a\u57df\u3002<\/p>\n\n\n\n
\u5206\u6c34\u5cad\u7b97\u6cd5\u7684\u6574\u4e2a\u8fc7\u7a0b\uff1a<\/p>\n\n\n\n
  1. \u628a\u68af\u5ea6\u56fe\u50cf\u4e2d\u7684\u6240\u6709\u50cf\u7d20\u6309\u7167\u7070\u5ea6\u503c\u8fdb\u884c\u5206\u7c7b\uff0c\u5e76\u8bbe\u5b9a\u4e00\u4e2a\u6d4b\u5730\u8ddd\u79bb\u9608\u503c\u3002<\/li>
  2. \u627e\u5230\u7070\u5ea6\u503c\u6700\u5c0f\u7684\u50cf\u7d20\u70b9\uff08\u9ed8\u8ba4\u6807\u8bb0\u4e3a\u7070\u5ea6\u503c\u6700\u4f4e\u70b9\uff09\uff0c\u8ba9threshold\u4ece\u6700\u5c0f\u503c\u5f00\u59cb\u589e\u957f\uff0c\u8fd9\u4e9b\u70b9\u4e3a\u8d77\u59cb\u70b9\u3002<\/li>
  3. \u6c34\u5e73\u9762\u5728\u589e\u957f\u7684\u8fc7\u7a0b\u4e2d\uff0c\u4f1a\u78b0\u5230\u5468\u56f4\u7684\u90bb\u57df\u50cf\u7d20\uff0c\u6d4b\u91cf\u8fd9\u4e9b\u50cf\u7d20\u5230\u8d77\u59cb\u70b9\uff08\u7070\u5ea6\u503c\u6700\u4f4e\u70b9\uff09\u7684\u6d4b\u5730\u8ddd\u79bb\uff0c\u5982\u679c\u5c0f\u4e8e\u8bbe\u5b9a\u9608\u503c\uff0c\u5219\u5c06\u8fd9\u4e9b\u50cf\u7d20\u6df9\u6ca1\uff0c\u5426\u5219\u5728\u8fd9\u4e9b\u50cf\u7d20\u4e0a\u8bbe\u7f6e\u5927\u575d\uff0c\u8fd9\u6837\u5c31\u5bf9\u8fd9\u4e9b\u90bb\u57df\u50cf\u7d20\u8fdb\u884c\u4e86\u5206\u7c7b\u3002<\/li><\/ol>\n\n\n\n
    \"\"\/
    \u56fe3<\/figcaption><\/figure>\n\n\n\n
    4. \u968f\u7740\u6c34\u5e73\u9762\u8d8a\u6765\u8d8a\u9ad8\uff0c\u4f1a\u8bbe\u7f6e\u66f4\u591a\u66f4\u9ad8\u7684\u5927\u575d\uff0c\u76f4\u5230\u7070\u5ea6\u503c\u7684\u6700\u5927\u503c\uff0c\u6240\u6709\u533a\u57df\u90fd\u5728\u5206\u6c34\u5cad\u7ebf\u4e0a\u76f8\u9047\uff0c\u8fd9\u4e9b\u5927\u575d\u5c31\u5bf9\u6574\u4e2a\u56fe\u50cf\u50cf\u7d20\u7684\u8fdb\u884c\u4e86\u5206\u533a\u3002<\/p>\n\n\n\n
    \u7528\u4e0a\u9762\u7684\u7b97\u6cd5\u5bf9\u56fe\u50cf\u8fdb\u884c\u5206\u6c34\u5cad\u8fd0\u7b97\uff0c\u7531\u4e8e\u566a\u58f0\u70b9\u6216\u5176\u5b83\u56e0\u7d20\u7684\u5e72\u6270\uff0c\u53ef\u80fd\u4f1a\u5f97\u5230\u5bc6\u5bc6\u9ebb\u9ebb\u7684\u5c0f\u533a\u57df\uff0c\u5373\u56fe\u50cf\u88ab\u5206\u5f97\u592a\u7ec6\uff08over-segmented\uff0c\u8fc7\u5ea6\u5206\u5272\uff09\uff0c\u8fd9\u56e0\u4e3a\u56fe\u50cf\u4e2d\u6709\u975e\u5e38\u591a\u7684\u5c40\u90e8\u6781\u5c0f\u503c\u70b9\uff0c\u6bcf\u4e2a\u70b9\u90fd\u4f1a\u81ea\u6210\u4e00\u4e2a\u5c0f\u533a\u57df\u3002<\/p>\n\n\n\n
    \u5176\u4e2d\u7684\u89e3\u51b3\u65b9\u6cd5\uff1a<\/p>\n\n\n\n
    1. \u5bf9\u56fe\u50cf\u8fdb\u884c\u9ad8\u65af\u5e73\u6ed1\u64cd\u4f5c\uff0c\u62b9\u9664\u5f88\u591a\u5c0f\u7684\u6700\u5c0f\u503c\uff0c\u8fd9\u4e9b\u5c0f\u5206\u533a\u5c31\u4f1a\u5408\u5e76\u3002<\/li>
    2. \u4e0d\u4ece\u6700\u5c0f\u503c\u5f00\u59cb\u589e\u957f\uff0c\u53ef\u4ee5\u5c06\u76f8\u5bf9\u8f83\u9ad8\u7684\u7070\u5ea6\u503c\u50cf\u7d20\u4f5c\u4e3a\u8d77\u59cb\u70b9\uff08\u9700\u8981\u7528\u6237\u624b\u52a8\u6807\u8bb0\uff09\uff0c\u4ece\u6807\u8bb0\u5904\u5f00\u59cb\u8fdb\u884c\u6df9\u6ca1\uff0c\u5219\u5f88\u591a\u5c0f\u533a\u57df\u90fd\u4f1a\u88ab\u5408\u5e76\u4e3a\u4e00\u4e2a\u533a\u57df\uff0c\u8fd9\u88ab\u79f0\u4e3a\u57fa\u4e8e\u56fe\u50cf\u6807\u8bb0(mark)\u7684\u5206\u6c34\u5cad\u7b97\u6cd5<\/strong>\u3002<\/li><\/ol>\n\n\n\n
      OpenCV\u4e2d\u5206\u6c34\u5cad\u7b97\u6cd5<\/strong><\/h2>\n\n\n\n
      \u5728OpenCV\u4e2d\uff0c\u6211\u4eec\u9700\u8981\u7ed9\u4e0d\u540c\u533a\u57df\u8d34\u4e0a\u4e0d\u540c\u7684\u6807\u7b7e\u3002\u7528\u5927\u4e8e1\u7684\u6574\u6570\u8868\u793a\u6211\u4eec\u786e\u5b9a\u4e3a\u524d\u666f\u6216\u5bf9\u8c61\u7684\u533a\u57df\uff0c\u75281\u8868\u793a\u6211\u4eec\u786e\u5b9a\u4e3a\u80cc\u666f\u6216\u975e\u5bf9\u8c61\u7684\u533a\u57df\uff0c\u6700\u540e\u75280\u8868\u793a\u6211\u4eec\u65e0\u6cd5\u786e\u5b9a\u7684\u533a\u57df\u3002\u7136\u540e\u5e94\u7528\u5206\u6c34\u5cad\u7b97\u6cd5\uff0c\u6211\u4eec\u7684\u6807\u8bb0\u56fe\u50cf\u5c06\u88ab\u66f4\u65b0\uff0c\u66f4\u65b0\u540e\u7684\u6807\u8bb0\u56fe\u50cf\u7684\u8fb9\u754c\u50cf\u7d20\u503c\u4e3a-1\u3002<\/p>\n\n\n\n
      \u4e0b\u9762\u5bf9\u76f8\u4e92\u63a5\u89e6\u7684\u786c\u5e01\u5e94\u7528\u8ddd\u79bb\u53d8\u6362\u548c\u5206\u6c34\u5cad\u5206\u5272\u3002<\/p>\n\n\n\n
      \"\"\/
      \u56fe6<\/figcaption><\/figure>\n\n\n\n
      \u5148\u4f7f\u7528 Otsu’s \u4e8c\u503c\u5316\u5bf9\u56fe\u50cf\u8fdb\u884c\u4e8c\u503c\u5316\u3002<\/p>\n\n\n\n
      import cv2\nimport numpy as np\n\nimg = cv2.imread('coins.png')\ngray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)\nret, thresh = cv2.threshold(gray, 0, 255, cv2.THRESH_BINARY_INV+cv2.THRESH_OTSU)<\/code><\/pre>\n\n\n\n
      \"\"\/
      \u56fe7<\/figcaption><\/figure>\n\n\n\n
      \u5148\u4f7f\u7528\u5f00\u8fd0\u7b97\u53bb\u9664\u56fe\u50cf\u4e2d\u7684\u7ec6\u5c0f\u767d\u8272\u566a\u70b9\uff0c\u7136\u540e\u901a\u8fc7\u8150\u8680\u8fd0\u7b97\u79fb\u9664\u8fb9\u754c\u50cf\u7d20\uff0c\u5f97\u5230\u7684\u56fe\u50cf\u4e2d\u7684\u767d\u8272\u533a\u57df\u80af\u5b9a\u662f\u771f\u5b9e\u524d\u666f\uff0c\u5373\u9760\u8fd1\u786c\u5e01\u4e2d\u5fc3\u7684\u533a\u57df\uff08\u4e0b\u9762\u5de6\u8fb9\u7684\u56fe\uff09\uff1b\u81a8\u80c0\u8fd0\u7b97\u4f7f\u5f97\u4e00\u90e8\u5206\u80cc\u666f\u6210\u4e3a\u4e86\u7269\u4f53\u5230\u7684\u8fb9\u754c\uff0c\u5f97\u5230\u7684\u56fe\u50cf\u4e2d\u7684\u9ed1\u8272\u533a\u57df\u80af\u5b9a\u662f\u771f\u5b9e\u80cc\u666f\uff0c\u5373\u8fdc\u79bb\u786c\u5e01\u7684\u533a\u57df\uff08\u4e0b\u9762\u4e2d\u95f4\u7684\u56fe\uff09\u3002<\/p>\n\n\n\n
      \u5269\u4e0b\u7684\u533a\u57df\uff08\u786c\u5e01\u7684\u8fb9\u754c\u9644\u8fd1\uff09\u8fd8\u4e0d\u80fd\u786e\u5b9a\u662f\u524d\u666f\u8fd8\u662f\u80cc\u666f\u3002\u53ef\u901a\u8fc7\u81a8\u80c0\u56fe\u51cf\u53bb\u8150\u8680\u56fe\u5f97\u5230\uff0c\u4e0b\u56fe\u4e2d\u7684\u767d\u8272\u90e8\u5206\u4e3a\u4e0d\u786e\u5b9a\u533a\u57df\uff08\u4e0b\u9762\u53f3\u8fb9\u7684\u56fe\uff09\u3002<\/p>\n\n\n\n
      # noise removal<\/em>\nkernel = cv2.getStructuringElement(cv2.MORPH_RECT, (3,3))\nopening = cv2.morphologyEx(thresh, cv2.MORPH_OPEN, kernel, iterations=2)\n\nsure_bg = cv2.dilate(opening, kernel, iterations=2)  # sure background area<\/em>\nsure_fg = cv2.erode(opening, kernel, iterations=2)  # sure foreground area<\/em>\nunknown = cv2.subtract(sure_bg, sure_fg)  # unknown area<\/em><\/code><\/pre>\n\n\n\n
      \"\"\/
      \u56fe8<\/figcaption><\/figure>\n\n\n\n
      \u5269\u4e0b\u7684\u533a\u57df\u4e0d\u786e\u5b9a\u662f\u786c\u5e01\u8fd8\u662f\u80cc\u666f\uff0c\u8fd9\u4e9b\u533a\u57df\u901a\u5e38\u5728\u524d\u666f\u548c\u80cc\u666f\u63a5\u89e6\u7684\u533a\u57df(\u6216\u8005\u4e24\u4e2a\u4e0d\u540c\u786c\u5e01\u63a5\u89e6\u7684\u533a\u57df)\uff0c\u6211\u4eec\u79f0\u4e4b\u4e3a\u8fb9\u754c\u3002\u901a\u8fc7\u5206\u6c34\u5cad\u7b97\u6cd5\u5e94\u8be5\u80fd\u627e\u5230\u786e\u5b9a\u7684\u8fb9\u754c\u3002<\/p>\n\n\n\n
      \u7531\u4e8e\u786c\u5e01\u4e4b\u95f4\u5f7c\u6b64\u63a5\u89e6\uff0c\u6211\u4eec\u4f7f\u7528\u53e6\u4e00\u4e2a\u786e\u5b9a\u524d\u666f\u7684\u65b9\u6cd5\uff0c\u5c31\u662f\u5e26\u9608\u503c\u7684\u8ddd\u79bb\u53d8\u6362<\/strong>\u3002<\/p>\n\n\n\n
      \u4e0b\u9762\u5de6\u8fb9\u7684\u56fe\u4e3a\u5f97\u5230\u7684\u8ddd\u79bb\u8f6c\u6362\u56fe\u50cf<\/strong>\uff0c\u5176\u4e2d\u6bcf\u4e2a\u50cf\u7d20\u7684\u503c\u4e3a\u5176\u5230\u6700\u8fd1\u7684\u80cc\u666f\u50cf\u7d20\uff08\u7070\u5ea6\u503c\u4e3a0\uff09\u7684\u8ddd\u79bb\uff0c\u53ef\u4ee5\u770b\u5230\u786c\u5e01\u7684\u4e2d\u5fc3\u50cf\u7d20\u503c\u6700\u5927\uff08\u4e2d\u5fc3\u79bb\u80cc\u666f\u50cf\u7d20\u6700\u8fdc\uff09\u3002\u5bf9\u5176\u8fdb\u884c\u4e8c\u503c\u5904\u7406\u5c31\u5f97\u5230\u4e86\u5206\u79bb\u7684\u524d\u666f\u56fe\uff08\u4e0b\u9762\u4e2d\u95f4\u7684\u56fe\uff09\uff0c\u767d\u8272\u533a\u57df\u80af\u5b9a\u662f\u786c\u5e01\u533a\u57df\uff0c\u800c\u4e14\u8fd8\u76f8\u4e92\u5206\u79bb\uff0c\u4e0b\u9762\u53f3\u8fb9\u7684\u56fe\u4e3a\u4e4b\u524d\u7684\u81a8\u80c0\u56fe\u51cf\u53bb\u4e2d\u95f4\u8fd9\u4e2a\u8868\u793a\u524d\u666f\u7684\u56fe\u3002<\/p>\n\n\n\n
      # Perform the distance transform algorithm<\/em>\ndist_transform = cv2.distanceTransform(opening, cv2.DIST_L2, 5)\n# Normalize the distance image for range = {0.0, 1.0}<\/em>\ncv2.normalize(dist_transform, dist_transform, 0, 1.0, cv2.NORM_MINMAX)\n\n# Finding sure foreground area<\/em>\nret, sure_fg = cv2.threshold(dist_transform, 0.5*dist_transform.max(), 255, 0)\n\n# Finding unknown region<\/em>\nsure_fg = np.uint8(sure_fg)\nunknown = cv2.subtract(sure_bg,sure_fg)<\/code><\/pre>\n\n\n\n
      \"\"\/
      \u56fe9<\/figcaption><\/figure>\n\n\n\n
      \u73b0\u5728\u6211\u4eec\u53ef\u4ee5\u786e\u5b9a\u54ea\u4e9b\u662f\u786c\u5e01\u533a\u57df\uff0c\u54ea\u4e9b\u662f\u80cc\u666f\u533a\u57df\u3002\u7136\u540e\u521b\u5efa\u6807\u8bb0(marker\uff0c\u5b83\u662f\u4e00\u4e2a\u4e0e\u539f\u59cb\u56fe\u50cf\u5927\u5c0f\u76f8\u540c\u7684\u77e9\u9635\uff0cint32\u6570\u636e\u7c7b\u578b)\uff0c\u8868\u793a\u5176\u4e2d\u7684\u6bcf\u4e2a\u533a\u57df\u3002\u5206\u6c34\u5cad\u7b97\u6cd5\u5c06\u6807\u8bb0\u76840\u7684\u533a\u57df\u89c6\u4e3a\u4e0d\u786e\u5b9a\u533a\u57df\uff0c\u5c06\u6807\u8bb0\u4e3a1\u7684\u533a\u57df\u89c6\u4e3a\u80cc\u666f\u533a\u57df\uff0c\u5c06\u6807\u8bb0\u5927\u4e8e1\u7684\u6b63\u6574\u6570\u8868\u793a\u6211\u4eec\u60f3\u5f97\u5230\u7684\u524d\u666f\u3002<\/p>\n\n\n\n
      \u6211\u4eec\u53ef\u4ee5\u4f7f\u7528 cv2.connectedComponents()<\/a> \u6765\u5b9e\u73b0\u8fd9\u4e2a\u529f\u80fd\uff0c\u5b83\u662f\u75280\u6807\u8bb0\u56fe\u50cf\u7684\u80cc\u666f\uff0c\u7528\u5927\u4e8e0\u7684\u6574\u6570\u6807\u8bb0\u5176\u4ed6\u5bf9\u8c61\u3002\u6240\u4ee5\u6211\u4eec\u9700\u8981\u5bf9\u5176\u8fdb\u884c\u52a0\u4e00\uff0c\u75281\u6765\u6807\u8bb0\u56fe\u50cf\u7684\u80cc\u666f\u3002<\/p>\n\n\n\n
      cv2.connectedComponents() \u5c06\u4f20\u5165\u56fe\u50cf\u4e2d\u7684\u767d\u8272\u533a\u57df\u89c6\u4e3a\u7ec4\u4ef6\uff08\u524d\u666f\uff09\u3002<\/p>\n\n\n\n
      # Marker labelling<\/em>\nret, markers = cv2.connectedComponents(sure_fg)\n# Add one to all labels so that sure background is not 0, but 1<\/em>\nmarkers = markers+1\n# Now, mark the region of unknown with zero<\/em>\nmarkers[unknown==255] = 0<\/code><\/pre>\n\n\n\n
      \u6ce8\u610f\uff1a\u5f97\u5230\u7684markers\u77e9\u9635\u7684\u5143\u7d20\u7c7b\u578b\u4e3a int32\uff0c\u8981\u4f7f\u7528 imshow() \u8fdb\u884c\u663e\u793a\uff0c\u9700\u8981\u5c06\u5176\u8f6c\u6362\u4e3a uint8 \u7c7b\u578b\uff08 markers=np.uint8(markers) \uff09\u3002<\/p>\n\n\n\n
      \u6211\u4eec\u5bf9\u5f97\u5230\u7684markers\u8fdb\u884c\u663e\u793a\uff1a<\/p>\n\n\n\n
      markers_copy = markers.copy()\nmarkers_copy[markers==0] = 150  # \u7070\u8272\u8868\u793a\u80cc\u666f<\/em>\nmarkers_copy[markers==1] = 0    # \u9ed1\u8272\u8868\u793a\u80cc\u666f<\/em>\nmarkers_copy[markers>1] = 255   # \u767d\u8272\u8868\u793a\u524d\u666f<\/em>\n\nmarkers_copy = np.uint8(markers_copy)<\/code><\/pre>\n\n\n\n
      \"\"\/
      \u56fe10<\/figcaption><\/figure>\n\n\n\n
      \u6807\u8bb0\u56fe\u50cf\u5df2\u7ecf\u5b8c\u6210\u4e86\uff0c\u6700\u540e\u5e94\u7528\u5206\u6c34\u5cad\u7b97\u6cd5\u3002\u7136\u540e\u6807\u8bb0\u56fe\u50cf\u5c06\u88ab\u4fee\u6539\uff0c\u8fb9\u754c\u533a\u57df\u5c06\u88ab\u6807\u8bb0\u4e3a-1\u3002<\/p>\n\n\n\n
      # \u4f7f\u7528\u5206\u6c34\u5cad\u7b97\u6cd5\u6267\u884c\u57fa\u4e8e\u6807\u8bb0\u7684\u56fe\u50cf\u5206\u5272\uff0c\u5c06\u56fe\u50cf\u4e2d\u7684\u5bf9\u8c61\u4e0e\u80cc\u666f\u5206\u79bb<\/em>\nmarkers = cv2.watershed(img, markers)\nimg[markers==-1] = [0,0,255]  # \u5c06\u8fb9\u754c\u6807\u8bb0\u4e3a\u7ea2\u8272<\/em><\/code><\/pre>\n\n\n\n
      \u7ecf\u8fc7\u5206\u6c34\u5cad\u7b97\u6cd5\u5f97\u5230\u7684\u65b0\u7684\u6807\u8bb0\u56fe\u50cf\u548c\u5206\u5272\u540e\u7684\u56fe\u50cf\u5982\u4e0b\u56fe\u6240\u793a\uff1a<\/p>\n\n\n\n
      \"\"\/
      \u56fe11<\/figcaption><\/figure>\n\n\n\n
      \u4efb\u4f55\u4e24\u4e2a\u76f8\u90bb\u8fde\u63a5\u7684\u7ec4\u4ef6\u4e0d\u4e00\u5b9a\u88ab\u5206\u6c34\u5cad\u8fb9\u754c(-1\u7684\u50cf\u7d20)\u5206\u5f00\uff1b\u4f8b\u5982\u5728\u4f20\u9012\u7ed9 watershed \u51fd\u6570\u7684\u521d\u59cb\u6807\u8bb0\u56fe\u50cf\u4e2d\u7684\u7269\u4f53\u76f8\u4e92\u63a5\u89e6\u3002<\/p>\n\n\n\n
      \u603b\u7ed3<\/h2>\n\n\n\n
      \u6211\u4eec\u901a\u8fc7\u4e00\u4e2a\u4f8b\u5b50\u4ecb\u7ecd\u4e86\u5206\u6c34\u5cad\u7b97\u6cd5\u7684\u6574\u4e2a\u8fc7\u7a0b\uff0c\u4e3b\u8981\u5206\u4e3a\u4ee5\u4e0b\u51e0\u6b65\uff1a<\/p>\n\n\n\n
      1. \u5bf9\u56fe\u8fdb\u884c\u7070\u5ea6\u5316\u548c\u4e8c\u503c\u5316\u5f97\u5230\u4e8c\u503c\u56fe\u50cf<\/li>
      2. \u901a\u8fc7\u81a8\u80c0\u5f97\u5230\u786e\u5b9a\u7684\u80cc\u666f\u533a\u57df\uff0c\u901a\u8fc7\u8ddd\u79bb\u8f6c\u6362\u5f97\u5230\u786e\u5b9a\u7684\u524d\u666f\u533a\u57df\uff0c\u5269\u4f59\u90e8\u5206\u4e3a\u4e0d\u786e\u5b9a\u533a\u57df<\/li>
      3. \u5bf9\u786e\u5b9a\u7684\u524d\u666f\u56fe\u50cf\u8fdb\u884c\u8fde\u63a5\u7ec4\u4ef6\u5904\u7406\uff0c\u5f97\u5230\u6807\u8bb0\u56fe\u50cf<\/li>
      4. \u6839\u636e\u6807\u8bb0\u56fe\u50cf\u5bf9\u539f\u56fe\u50cf\u5e94\u7528\u5206\u6c34\u5cad\u7b97\u6cd5\uff0c\u66f4\u65b0\u6807\u8bb0\u56fe\u50cf<\/li><\/ol>\n\n\n\n
        \u53c2\u8003\uff1a<\/p>\n\n\n\n
        OpenCV Watershed Algorithm<\/a><\/p>\n\n\n\n
        IMAGE SEGMENTATION AND MATHEMATICAL MORPHOLOGY<\/a><\/p>\n\n\n\n
        Classic Watershed<\/a><\/p>\n\n\n\n
        https:\/\/zhuanlan.zhihu.com\/p\/67741538<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"
        \u6458\u81ea\u7f51\u7edc\uff0c\u4fb5\u6743\u7acb\u5220\u3002 \u5085\u91cc\u53f6\u53d8\u6362\u7684\u4f5c\u7528\uff1a \u3000\u3000\u5bf9\u4e8e\u6570\u5b57\u56fe\u50cf\u8fd9\u79cd\u79bb\u6563\u7684\u4fe1\u53f7\uff0c\u9891\u7387\u5927\u5c0f\u8868\u793a\u4fe1\u53f7\u53d8\u6362\u7684\u5267\u70c8\u7a0b\u5ea6\u6216\u8005\u8bf4 … \u7ee7\u7eed\u9605\u8bfb\u56fe\u50cf\u5085\u91cc\u53f6\u53d8\u6362 and\u56fe\u50cf\u5206\u5272\u7684\u7ecf\u5178\u7b97\u6cd5\uff1a\u5206\u6c34\u5cad\u7b97\u6cd5<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":[],"categories":[15],"tags":[],"_links":{"self":[{"href":"http:\/\/139.9.1.231\/index.php\/wp-json\/wp\/v2\/posts\/1938"}],"collection":[{"href":"http:\/\/139.9.1.231\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/139.9.1.231\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/139.9.1.231\/index.php\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"http:\/\/139.9.1.231\/index.php\/wp-json\/wp\/v2\/comments?post=1938"}],"version-history":[{"count":14,"href":"http:\/\/139.9.1.231\/index.php\/wp-json\/wp\/v2\/posts\/1938\/revisions"}],"predecessor-version":[{"id":1955,"href":"http:\/\/139.9.1.231\/index.php\/wp-json\/wp\/v2\/posts\/1938\/revisions\/1955"}],"wp:attachment":[{"href":"http:\/\/139.9.1.231\/index.php\/wp-json\/wp\/v2\/media?parent=1938"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/139.9.1.231\/index.php\/wp-json\/wp\/v2\/categories?post=1938"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/139.9.1.231\/index.php\/wp-json\/wp\/v2\/tags?post=1938"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}