![](\"https:\/\/lh4.googleusercontent.com\/Xoi4m2wNRxwZEkAXlfAxvEBbsjYT2ntSAzoJqlN4lF4gl-iIF2vbTvi7GdEIb_8lq2x6qLVoEevxCid0nDxO_XtpC93RBlJiBX4meFMRtG436y9_CrBmS2MCKkewa0dBwsVRfMCa0emXO827Ce6TjwQQDABQde7oNWrI53udUJgzDXHn_iv54a6wSVFouw\")
<\/p>\n<\/b><\/p>\n
<\/p>\n
\u901a\u904e\u6a5f\u5668\u5b78\u7fd2\u6e1b\u5c11\u6578\u64da\u5dee\u7570<\/p>\n
Reduce data discrepancies via machine learning<\/p>\n
\u624b\u908a\u8d8a\u4f86\u8d8a\u591a\u6e2c\u91cf\u8272\u5f69\u7684\u5100\u5668\uff0c\u91cf\u6e2c\u51fa\u4f86\u7684\u6578\u503c\u90fd\u4e0d\u592a\u4e00\u6a23\uff0c\u9019\u500b"\u4e0d\u4e00\u6a23"\u662f\u53ef\u4ee5\u7406\u89e3\u7684\uff1a\u4e0d\u540c\u7684\u611f\u61c9\u5143\u4ef6\u3001\u4e0d\u540c\u7684\u7269\u7406\u7d50\u69cb\u3001\u4e0d\u540c\u7684\u6821\u6b63\u908f\u8f2f\u3001\u4e0d\u540c\u7684\u512a\u5316\u6f14\u7b97\uff0c\u5f97\u5230\u7684\u6578\u503c\u4e0d\u540c\u53ef\u4ee5\u7406\u89e3\u3002\u4f46\u5dee\u7570\u8981\u5728\u591a\u5c11\u4e4b\u5167\u624d\u7b97\u582a\u7528\uff1f\u5373\u4f7f\u662f\u81fa\u5e6320\u842c\u4ee5\u4e0a\u7684\u5100\u5668\uff0c\u9084\u662f\u6703\u6709\u4e00\u500bde00\u7684\u5dee\u8ddd\uff1b\u90a3\u9019\u4e9b\u7947\u6709\u5e7e\u5343\u584a\u7684\u5100\u5668\uff0c\u6211\u5011\u80fd\u6709\u591a\u5c11\u671f\u5f85\uff1f<\/p>\n
\u7576\u7136\u4e0d\u6703\u671f\u5f85\u9019\u4e9b\u4f4e\u968e\u7684\u5100\u5668\u62ff\u4f86\u505aFogra\u3001G7\u3001gmi\u9019\u985e\u7684\u8a8d\u8b49\uff1b\u4f46\u4f5c\u70ba\u4e00\u500b\u5c07\u8272\u5f69\u611f\u5b98\u5c0e\u5165\u70ba\u6578\u503c\u6982\u5ff5\u7684\u9019\u6a23\u4e00\u500b\u6572\u9580\u78da\u7684\u6027\u8cea\uff0c\u6211\u4e00\u6a23\u8a8d\u53ef\u9019\u4e9b\u4f4e\u968e\u7684\u8a2d\u5099\u3002<\/p>\n
\u5c0d\u65bc\u9019\u4e9b\u6572\u9580\u78da\u6027\u8cea\u3001\u4f4e\u968e\u5100\u5668\u7684\u7cbe\u51c6\u5ea6\uff0c\u6211\u8a8d\u70ba\u5169\u500bde00 \u662f\u582a\u7528\u7684\u3002<\/p>\n
\u5c0d\u65bc\u9019\u4e9b\u4f4e\u968e\u5100\u5668\u7684\u904b\u7528\uff0c\u66f4\u91cd\u8981\u7684\u662f\u60f3\u6cd5\u7684\u8f49\u8b8a\uff1a\u8272\u5f69\u9019\u4e00\u500b\u4e3b\u89c0\u7684\u611f\u53d7\uff0c\u7d93\u7531\u5100\u5668\uff0c\u8272\u5f69\u6703\u662f\u4e00\u500b\u6e05\u695a\u7684\u6578\u5b57\uff1b\u4e00\u65e6\u662f\u6578\u5b57\uff0c\u6211\u5011\u5728\u9a57\u6536\u6216\u662f\u64cd\u4f5c\u63a7\u5236\u4e0a\uff0c\u90fd\u80fd\u6709\u66f4\u597d\u7684\u638c\u63e1\u3002\u5c0d\u65bc\u9019\u4e9b\u4f4e\u968e\u7684\u5100\u5668\u7cbe\u6e96\u5ea6\uff0c\u5373\u4f7f\u53ea\u5e36\u5230\u5169\u500bde00\u7684\u4f4d\u7f6e\uff0c\u6211\u9084\u662f\u6301\u6b63\u9762\u7684\u614b\u5ea6\u3002\u518d\u5f37\u8abf\u4e00\u6b21\uff0c\u5c0d\u65bc\u9019\u4e9b\u6703\u62ff\u5100\u5668\u4f86\u770b\u5f85\u8272\u5f69\u7684\u4f7f\u7528\u8005\uff0c\u6709\u80fd\u529b\u6216\u8005\u6709\u610f\u9858\u5c07\u8272\u5f69\u7576\u6210\u6578\u5b57\u4f86\u8655\u7406\u7684\u4eba\uff0c\u601d\u8003\u6a21\u5f0f\u7684\u8f49\u8b8a\u3001\u624d\u662f\u6700\u91cd\u8981\u7684\u610f\u7fa9\u3002<\/p>\n
\u4ee5\u4e0b\u6578\u64da\u662f\u4ee5 Fogra media wedge CMYK V3(72\u683c)\u53d6\u6a23\uff0c\u67d0\u4f4e\u968e\u5100\u5668\u76f8\u5c0d\u65bci1Pro3\u5dee\u7570\u70ba\u6700\u59271.79de00\uff0c\u5e73\u57470.91de00\u3002\u6211\u8a8d\u70ba\u9019\u6a23\u7684\u6572\u9580\u78da\u7b97\u662f\u582a\u7528\u4e86\u3002<\/p>\n
![](\"https:\/\/lh5.googleusercontent.com\/nBpxPwT827DnKwxz4-7Lcl2jhBifaxOCnw9VOfR-kOGEkjYfvGe9kVoMLtPynzMGZ5OU6vpHXEJ4R7ltUqaXC2udxSruE3yCgRuIlWX-QJCUeAXbb6QcEf00_HTdQUllMBgRqBLBdrsuYfmJSnMzsNgePxLO_hoORx_bCOf0lP9A936XHYaBFd1Q8iklIg\")
Fig. \u4ee5Fogra media wedge CMYK V3(72\u683c)\u53d6\u6a23\uff0c\u67d0\u4f4e\u968e\u5100\u5668\u76f8\u5c0d\u65bci1Pro3\u5dee\u7570\u70ba\uff1a\u6700\u59271.79de00\uff0c\u5e73\u57470.91de00\u3002<\/p>\n
\u9664\u4e86\u6e96\u5ea6\u5920\u4e0d\u5920\u597d\uff0c\u53e6\u4e00\u500b\u60f3\u6cd5\u662f\uff1a\u9019\u500b\u8272\u5f69\u73fe\u8c61\u73fe\u5728\u5df2\u7d93\u662f\u4e00\u500b\u6578\u5b57\uff0c\u4e00\u65e6\u662f\u6578\u5b57\uff0c\u6211\u5011\u5c31\u53ef\u4ee5\u900f\u904e\u4e00\u4e9b\u6578\u5b78(\u7d71\u8a08\u5b78)\u7a0b\u5e8f\u4f86\u512a\u5316\u9019\u4e9b\u6578\u5b57\uff0c\u8b93\u5100\u5668\u4e4b\u9593\u5448\u73fe\u7684\u6578\u503c\u53ef\u4ee5\u66f4\u63a5\u8fd1\u3002\u4e0d\u7ba1\u4f4e\u968e\u9ad8\u968e\uff0c\u5c0d\u6211\u4f86\u8aac\uff0c\u662f\u6709\u5fc5\u8981\u53bb\u627e\u4e00\u500b\u5de5\u4f5c\u7a0b\u5e8f\uff0c\u8b93\u5169\u500b\u5100\u5668\u7684\u6578\u5b57\u66f4\u63a5\u8fd1\u3002<\/p>\n
\u6700\u8fd1\u63a5\u89f8\u5230 Machine learning\uff0c\u5c31\u8a66\u8a66\u5176\u4e2d\u5e7e\u500b\u5de5\u5177\uff0c\u770b\u80fd\u4e0d\u80fd\u9054\u6210\u9810\u671f\u7684\u6548\u679c\u3002<\/p>\n
![](\"https:\/\/lh6.googleusercontent.com\/kOz_F1sUjn7S2aid7_L9q2EALhFMT5IU8bZUCcI-KTy0gOECLjHybKRw_6E_xGq8zBlAls4Z_Y4N3W4g0XDaBeHXgOg86Hdvk2amXT-ErbHKn1lYqhr2enMW-FCvgyUXyw7L0UtaquD3I8KLLC8FsBt7WJCIxxni03yUbEeHbC_M5O2ULRIDcx2_2JDbVQ\")
Fig. python machine learning \u5de5\u5177<\/p>\n
\u7a0d\u5fae\u77ad\u89e3\u4e00\u4e0b python machine learning \u7684\u529f\u80fd\uff0c\u770b\u4f86\u662f\u5f88\u591a\u7d71\u8a08\u5b78\u7684\u5de5\u5177\u5eab\uff0c\u4e3b\u8981\u662f\u5728\u8abf\u7528\u4e0a\u986f\u5f97\u65b9\u4fbf\u96f6\u6d3b\u8a31\u591a\u3002\u5167\u5bb9\u5f88\u591a\uff0c\u9700\u8981\u4e00\u4e9b\u6642\u9593\u53bb\u6d88\u5316\u7406\u89e3\u3002\u9019\u6b21\u5148\u8a66\u8a66 Multiple Regression \u770b\u770b\u80fd\u6709\u4ec0\u9ebc\u6210\u679c\u3002<\/p>\n
\u9996\u5148\uff0c\u8981\u600e\u9ebc\u9935\u8cc7\u6599\u5c31\u662f\u4e00\u9580\u5b78\u554f\uff1a\u8981\u591a\u5c11\u6a23\u672c\u6578\uff1f\u8272\u5f69\u6a23\u672c\u7684\u5206\u4f48\u60c5\u6cc1\uff1f\u6a23\u672c\u7684\u53c3\u8003\u9ede\u8981\u8a2d\u591a\u5c11\u500b\uff1f\u600e\u9ebc\u8a2d\uff1f<\/p>\n
\u4e00\u65b9\u9762\u8981\u589e\u52a0\u7cbe\u5ea6\u3001\u4e00\u65b9\u9762\u53c8\u8981\u8b93\u64cd\u4f5c\u5118\u91cf\u7cbe\u7c21\uff1b\u5404\u7a2e\u65b9\u9762\u7684\u8003\u616e\u9700\u8981\u4e0d\u65b7\u7684Trial and error\u53bb\u627e\u5230\u6700\u4f73\u7684\u7d44\u5408\u3002<\/p>\n
![](\"https:\/\/lh6.googleusercontent.com\/P7-k0Ym1y9lyjGMFB_p6yP_Fifh9H5YXTGSprD87ieVyVSzNM44qcJ7ZEfZSPu9CgnCY_rfsuN8DCzOatl7aWlkjaUOt2-OKovkHsYw-7Hd2G8TS5lCu8J0q32YhBwKk9LafgZGJJBGxwLQc0UKx9zhQ71BztC0vcdTRZYC-5EWGlvR7MF9Wq0CiBMLtEA\")
Fig. Fogra media wedge CMYK V3<\/p>\n
\u9019\u4e00\u6b21\u7528Fogra media wedge CMYK V3(72\u683c)\u53d6\u6a23\uff0c\u4ee5i1 Pro3 \u7684\u6578\u503c\u505a\u70ba\u76ee\u6a19\u7d44\uff0cCR30\u7684\u6578\u503c\u505a\u70ba\u5be6\u9a57\u7d44\uff0c\u5982\u4e0a\u5716\u6240\u793a\uff0c72\u500b\u6578\u64da\u4e0b\u4f86\uff0c\u6700\u5927\u5dee\u75701.79de00\uff0c\u5e73\u5747\u5dee\u75700.91de00\u3002<\/p>\n
\u8a66\u8457\u7528multiple regression \u505a\u5169\u7d44\u6578\u64da\u9593\u7684\u95dc\u806f\uff1b\u5206\u5225\u8a66\u4e86\u4e00\u6b21\u65b9\u7684\u95dc\u806f\u8207\u4e09\u6b21\u65b9\u7684\u95dc\u806f\uff0c\u7d50\u679c\u5982\u4e0b\uff1a<\/p>\n
72\u500b\u8a13\u7df4\u6a23\u672c\uff0c\u4e00\u6b21\u65b9(degree=1)\u3001\u4e09\u500b\u8b8a\u9805(L,a,b)\u95dc\u806f\uff0c<\/p>\n
\u7531machine learning \u5f97\u51fa\u4f86\u7684\u95dc\u806f\u5f0f\u5982\u4e0b\uff1a<\/p>\n
L(predict)=-0.6322502637360+1.02341127*L(measured)+0.01230391*a(measured)-0.00582456*b(measured)<\/p>\n
a(predict)=-0.5175238518122+0.00349049*L(measured)+1.00839398*a(measured)-0.00698096*b(measured)<\/p>\n
b(predict)=0.6629766575213-0.01703327*L(measured)-0.01942645*a(measured)+1.00670896*b(measured)<\/p>\n
\u7d93\u7531\u95dc\u806f\u5f0f\u4fee\u6b63CR30\u91cf\u6e2c\u503c\uff0c\u8207i1Pro3\u7684\u6700\u5927\u5dee\u8ddd\u75311.79 de00 \u964d\u5230 1.57 de00\uff0c\u5e73\u5747\u5dee\u8ddd\u75310.91 \u964d\u5230 0.45 de00\uff0c\u7b97\u662f\u6709\u52b9\u7684\u8a13\u7df4\u3002<\/p>\n
![](\"https:\/\/lh3.googleusercontent.com\/hcKD5RCVTq1xwO8mDfSVorL1OzEKZlBeFcDruHDyeF191YOAO_d3iMcHNU45I3fyR24vwSIXeAGaNCbxIZcRMk6ZlMV6FltMpqsa1wQ8i51BBAph7SiwzsAGSex04owOyHW4bSbNjVcNbwM56lUKL7Jf1qCHcQbSuWAvfxnZHKb08qLlOG9HX1E6ZVQYgQ\")
Fig. 72\u500b\u8a13\u7df4\u6a23\u672c\uff0c\u4e00\u6b21\u65b9(degree=1)\u3001\u4e09\u500b\u8b8a\u9805(L,a,b)\u95dc\u806f\uff0c\u6700\u5927\u5dee\u8ddd 1.576 de00\uff0c\u5e73\u5747\u5dee\u8ddd 0.445 de00<\/p>\n
\u8a66\u8457\u7528\u4e09\u6b21\u65b9(degree=3)\u4e09\u500b\u8b8a\u9805(L,a,b)\u95dc\u806f<\/p>\n
\u7531\u65bcmachine learning \u5f97\u51fa\u4f86\u7684\u95dc\u806f\u5f0f\u904e\u65bc\u8907\u96dc\uff0c\u7528\u7a0b\u5f0f\u4ee3\u78bc\u4f86\u4ee3\u8868\u4e00\u4e0b\u3002<\/p>\n
\u4e09\u500b\u8b8a\u9805\u3001\u4e09\u6b21\u65b9\u95dc\u806f\u5e8f\uff1a<\/p>\n
‘l’, ‘a’, ‘b’, ‘l^2’, ‘l a’, ‘l b’, ‘a^2’, ‘a b’, ‘b^2’, ‘l^3’, ‘l^2 a’, ‘l^2 b’, ‘l a^2’, ‘l a b’, ‘l b^2’, ‘a^3’, ‘a^2 b’, ‘a b^2’, ‘b^3’<\/p>\n
\u517119\u9805\u3002<\/p>\n
\u7a0b\u5e8f\u78bc\u5982\u4e0b\uff1a<\/p>\n
Predict = intercept +regression_model.coef_[1]*mL+regression_model.coef_[2]*ma+regression_model.coef_[3]*mb+regression_model.coef_[4]*pow(mL,2)+regression_model.coef_[5]*mL*ma+regression_model.coef_[6]*mL*mb+regression_model.coef_[7]*pow(ma,2)+regression_model.coef_[8]*ma*mb+regression_model.coef_[9]*pow(mb,2)++regression_model.coef_[10]*pow(mL,3)++regression_model.coef_[11]*pow(mL,2)*ma+regression_model.coef_[12]*pow(mL,2)*mb++regression_model.coef_[13]*mL*pow(ma,2)+regression_model.coef_[14]*mL*ma*mb++regression_model.coef_[15]*mL*pow(mb,2)+regression_model.coef_[16]*pow(ma,3)+regression_model.coef_[17]*pow(ma,2)*mb+regression_model.coef_[18]*ma*pow(mb,2)+regression_model.coef_[19]*pow(mb,3)<\/p>\n
L intercept = -3.3311399027457753<\/p>\n
L coefficients = [1.17959558e+00,1.51891147e-02,1.67282780e-02,-2.49762450e-03,-1.09825293e-04,-1.09264761e-03,-1.50801917e-04,-2.44847912e-04,6.39178595e-04,1.17880166e-05,-5.01404941e-07,1.11556891e-05,2.25110985e-06,6.49420255e-06,-1.11279844e-05,1.39766751e-06,-6.58363756e-07,-3.61386136e-06,2.77494282e-06]<\/p>\n
a intercept = 0.3279017535550137<\/p>\n
a coefficients = [-3.31344905e-02,1.08827162e+00,1.32214608e-03,5.74573847e-04,-2.14549183e-03,-3.29938673e-04,-1.10366977e-03,-1.52137755e-04,4.83060199e-05,-3.00381953e-06,1.08584656e-05,2.18611803e-06,1.55290706e-05,3.73154456e-06,5.74063836e-07,1.41496091e-06,2.62797607e-06,-9.81271691e-07,1.41858407e-07]<\/p>\n
b intercept = 0.08716778562166638<\/p>\n
b coefficients = [5.57303728e-03,-6.06917904e-02,1.07554181e+00,-4.97617821e-04,2.19794160e-03,-7.81213339e-04,5.98329555e-04,-2.34457666e-04,7.51041096e-04,3.71288376e-06,-1.94526634e-05,-1.63889837e-06,-8.64799163e-06,-3.87728970e-06,-3.76102212e-06,-4.37681031e-06,-1.17723242e-05,-2.55467020e-06,-5.85244726e-06]<\/p>\n
\u4e09\u500b\u8b8a\u9805\uff0c\u4e09\u500b\u6b21\u65b9\u7684\u95dc\u806f\uff0c\u7522\u751f\u591a\u905419\u500b\u8ff4\u6b78\u4fc2\u6578\uff1b\u5982\u6b64\u9f90\u5927\u7684\u8a08\u7b97\u91cf\uff0c\u5728 python sklearn \u5957\u4ef6\u88cf\uff0c\u6392\u9664\u6389\u8cc7\u6599\u5ba3\u544a\uff0c\u53ea\u8981\u4e09\u500b\u654d\u8ff0\u5c31\u53ef\u5f97\u51fa\u9810\u6e2c\u95dc\u806f\u5f0f\u3002\u5957\u4ef6\u7684\u4f7f\u7528\u975e\u5e38\u65b9\u4fbf\uff0c\u91cd\u8981\u7684\u662f\u8981\u60f3\u6e05\u695a\u5de5\u4f5c\u908f\u8f2f\u3001\u6578\u64da\u95dc\u806f\u8207\u53d6\u6a23\u8a2d\u8a08\u3002<\/p>\n
![](\"https:\/\/lh3.googleusercontent.com\/IXV1oplCPV4bml-9iyoZ7efc7Q7Q8RvYpxg-ZBT94Xn2dRe_pi9fX7_56aqf8Gn7FB4sP60xtgnzrWOdKKT9wv_krWSmOHgBCnlKlD9t5XAxynAdJ-TU0dskVzIyeDBmkQ52bQYjW9lKdNrsvl_0tKhDUSJ8Ral0G8Ju8M_9ojMTVPraD5uY_064U66okA\")
Fig. python sklearn \u4e09\u500b\u654d\u8ff0\u5373\u53ef\u53d6\u5f97\u753119\u500b\u8ff4\u6b78\u4fc2\u6578\u7d44\u5408\u6210\u7684\u9810\u6e2c\u503c\u3002<\/p>\n
\u7d93\u7531\u4e09\u6b21\u65b9\u8ff4\u6b78\u9810\u6e2c\uff0cCR30\u8207i1Pro3\u7684\u6700\u5927\u5dee\u8ddd\u75311.79 de00 \u964d\u5230 0.63 de00\uff0c\u5e73\u5747\u5dee\u8ddd\u75310.91 \u964d\u5230 0.22 de00\uff0c\u662f\u975e\u5e38\u6709\u52b9\u7684\u8a13\u7df4\u3002<\/p>\n
![](\"https:\/\/lh6.googleusercontent.com\/ZKbTryptS8-TA2C9PTzY3M8FMhrDHd7hEbO_WEw0bF5mRYQIbjA2jLvQe__jjRd27AszJfcudEK-G5K1jKERjTq87m1v3rtbP8jM5one0ER0Cuowppa8fGFO-XfV59PVYr-2OrxNVxqSDsIf_g_AAOjelG3IxR4y9NoET4PQ9ArFVqXnR6I-ARquWYA0Iw\")
Fig. 72\u500b\u8a13\u7df4\u6a23\u672c\uff0c\u4e09\u6b21\u65b9(degree=3)\u3001\u4e09\u9805\u6b21(L,a,b)\u95dc\u806f\uff0c\u6700\u5927\u5dee\u8ddd 0.63 de00\uff0c\u5e73\u5747\u5dee\u8ddd 0.22 de00<\/p>\n
![](\"https:\/\/lh4.googleusercontent.com\/HSofSZNbUjFmTeypMTVAkmcs5JQjNiW4MUy39c97AZ53lOM067Ytu-B17qqIt0-rGnCVWJGOmbJbpMzBTuGlM3rvtDWAwaWG_d32VJPz3OD9ngNVl9ODbhcayuvg3oyyJTEq-HiudVvjC-2U3YIlaRSXQj_wam3-Vty4MyJwNV-pEqXfcgx2l-GUNmFOBA\")
Fig. \u4e09\u6b21\u65b9(degree=3)\u3001\u4e09\u9805\u6b21(L,a,b)\u95dc\u806f\uff0cL\u503c\u7684\u76ee\u6a19\u8207\u9810\u6e2c<\/p>\n
![](\"https:\/\/lh5.googleusercontent.com\/HWXPBCiCZtZRS4WiOGGHCV4Ad9svofoPlNoqNb4IAGSWDcWxoaaGygmEXuemdmmHAmzhtH4WsrjLQVvNU_WhbFoeXUPdXFtG6eJi8aAD5LS4yNPFXoYb9xsWMqdJUkCLwwsCD2Yo07qOBE22yqT1AIZqa-r8XgrMW1bfhSszThubMCbxFsNL1lAg2DBSeg\")
Fig. \u4e09\u6b21\u65b9(degree=3)\u3001\u4e09\u9805\u6b21(L,a,b)\u95dc\u806f\uff0ca\u503c\u7684\u76ee\u6a19\u8207\u9810\u6e2c<\/p>\n
![](\"https:\/\/lh5.googleusercontent.com\/LpXDHZ2G0CZzCqn8U8EOEtJYNzgiCf3TxwoDcBHe0UVIjcmMjj9TGKDJDAj7-WI-PAiPxopgkGdAFjnze_xHpnghrH_eF0wJK3mC6w2g7KAg66C-xMSGiB5epPEuMBIpyW7DzfRdEw_9pIDwUFhZjKzP8-8BEFyhA7yB1tjnPRmRFCaEwU_PmtciVtfdDw\")
Fig. \u4e09\u6b21\u65b9(degree=3)\u3001\u4e09\u9805\u6b21(L,a,b)\u95dc\u806f\uff0cb\u503c\u7684\u76ee\u6a19\u8207\u9810\u6e2c<\/p>\n
\u5982\u6b64\u6709\u6548\u7684\u8a13\u7df4\uff0c\u6700\u5927\u5dee\u8ddd 0.63 de00\uff0c\u5e73\u5747\u5dee\u8ddd 0.22 de00\uff0c\u9019\u6a23\u7684\u6578\u503c\u5dee\u8ddd\u582a\u6bd420\u842c\u7d1a\u8ddd\u7684\u8a2d\u5099\u4e86\uff0c\u4f46\u9019\u53ea\u662f\u8a13\u7df4\u7d44\u88cf\u7684\u6a23\u672c\uff0c\u5176\u5b83\u7684\u91cf\u6e2c\u503c\u4e00\u6a23\u80fd\u9054\u5230\u9019\u6a23\u7684\u6210\u679c\u55ce\uff1f<\/p>\n
\u518d\u6e2c\u4e86\u4e09\u500b\u4e0d\u5728\u8a13\u7df4\u7d44\u7684\u6e2c\u8a66\u9ede\uff0c\u5206\u5225\u70baLab (50.69,65.07,47.67)\u3001 (86.81,1.55,11.29)\u53ca(71.69,32.55,-0.15)\uff1bCR30\u4fee\u6b63\u524d\u5dee\u8ddd\u70bade00 2.14\u30012.42\u30011.61\uff0c\u4e00\u6b21\u65b9\u4fee\u6b63\u7d50\u679c\u70ba1.10\u30011.83\u30010.49\uff0c\u6709\u9054\u5230\u4fee\u6b63\u6548\u679c\u3002\u4e09\u6b21\u65b9\u4fee\u6b63\u7d50\u679c\u70ba0.52\u30011.87\u30010.55\uff0c\u9664\u7b2c\u4e8c\u500b\u6a23\u672c(86.81,1.55,11.29)\u4fee\u6b63\u6709\u9650\uff0c\u5176\u5b83\u5169\u500b\u6a23\u672c\u53ef\u4ee5\u5f9e2\u500bde00\u4fee\u6b630.5\u5de6\u53f3\uff0c\u7b97\u662f\u5f88\u6709\u52b9\u7684\u4fee\u6b63\u3002<\/p>\n
![](\"https:\/\/lh5.googleusercontent.com\/7uSRBytWTqqZf6hCBY4QkjwL60c8-HJtLpd7Lak21OaKIn2qvsWnvc2bT-uXaf1aI8HAcaxyom0dYnEOi3TKtcDzDG-jBwGxhejIPFZZzGQjdoX453JGVtrEY_AVJbrNhRxmislKI_846qllVwQu63kICAz9b5rBrQE7XCLMrTT7O72B7wHVmMg4VnPXbg\")
Fig. Fogra media wedge \u8a13\u7df4\u7d44\u4e4b\u5916\u7684\u4e09\u500b\u6a23\u672c\u3002<\/p>\n
![](\"https:\/\/lh3.googleusercontent.com\/P-EPaHQjU-V4hLStuljPCo8wBh02Dqxw0uH2X9K6CcFY2jImKDou7xeTlwRGwC3iS-p80brGJF7mhuQ8meYVzSdbz9Aijmtuq17k0qj-31905onhStipaD4TLBaJVPoJHWB1x0LufNoglO-RrFZc6pMCo3QVg_hIIgqqRmm3SZ0F23WCy8EW0vdBpzMJ2g\")
Fig. \u8a13\u7df4\u7d44\u5916\u4e09\u500b\u6a23\u672c\u7684\u4fee\u6b63\u503c\uff0c\u5206\u4e00\u6b21\u65b9\u4fee\u6b63\u53ca\u4e09\u6b21\u65b9\u4fee\u6b63\u3002<\/p>\n
\u525b\u958b\u59cb\u63a5\u89f8 Machine learning\uff0c\u9019\u6a23\u7684\u6210\u679c\u7b97\u662f\u6709\u9054\u5230\u6211\u7684\u9810\u671f\u3002<\/p>\n
\u4e4b\u6240\u4ee5\u6703\u5e36\u51faMachine learning\u9019\u500b\u984c\u76ee\uff0c\u4e00\u958b\u59cb\uff0c\u5f88\u55ae\u7d14\u7684\uff0c\u624b\u908a\u597d\u5e7e\u500b\u91cf\u6e2c\u5de5\u5177\uff0c\u53ea\u662f\u5e0c\u671b\u5b83\u5011\u7684\u6578\u64da\u80fd\u5920\u66f4\u52a0\u4e00\u81f4\u3002 <\/p>\n
\u628a\u9019\u500b\u984c\u76ee\u7684\u4f7f\u7528\u5834\u666f\u7e7c\u7e8c\u64f4\u5927\uff0c\u8b93\u4efb\u610f\u5169\u500b\u91cf\u6e2c\u5100\u5668\u7d93\u7531\u4e00\u5957\u5b78\u7fd2\u6a21\u5f0f\uff0c\u6578\u503c\u80fd\u5920\u5f7c\u6b64\u63a5\u8fd1\u3002<\/p>\n
\u4e00\u500b\u5834\u666f\u53ef\u4ee5\u662f\u5ba2\u6236\u8207\u751f\u6210\u7aef\u7684\u5100\u5668\u4e92\u76f8\u5b78\u7fd2\uff0c\u9019\u6a23\u5728\u5ba2\u6236\u8207\u751f\u7522\u7aef\u5404\u81ea\u4f7f\u7528\u5100\u5668\u6642\uff0c\u80fd\u66f4\u52a0\u76f8\u4e92\u4fe1\u4efb\uff0c\u6e1b\u5c11\u722d\u8b70\u3002\u9019\u6a23\u7684\u61c9\u7528\u5834\u666f\uff0c\u4e5f\u53ef\u4ee5\u662f\u751f\u7522\u55ae\u4f4d\u7d81\u5b9a\u5ba2\u6236\u7684\u4e00\u7a2e\u6280\u8853\u624b\u6bb5\uff0c\u9019\u500b\u5834\u666f\u7684\u8a2d\u5b9a\u662f\uff0c\u5ba2\u6236\u7528\u8f03\u4f4e\u968e\u7684\u5100\u5668\u5411\u751f\u7522\u55ae\u4f4d\u9ad8\u968e\u7684\u5100\u5668\u5b78\u7fd2\uff0c\u5ba2\u6236\u7684\u9019\u500b\u5b78\u7fd2\u6210\u679c\uff0c\u8879\u80fd\u91dd\u5c0d\u4ed6\u5b78\u7fd2\u5c0d\u8c61\u7684\u55ae\u4f4d\uff0c\u5982\u6b64\uff0c\u751f\u7522\u55ae\u4f4d\u5373\u4ee5\u6b64\u6578\u64da(\u5c0d\u61c9\u8868)\u4f86\u7d81\u5b9a\u5ba2\u6236\u7684\u5fe0\u8aa0\u5ea6\u3002<\/p>\n
\u4e5f\u53ef\u4ee5\u662f\u540c\u4e00\u55ae\u4f4d\u5404\u500b\u90e8\u9580\u7684\u5100\u5668\uff0c\u7d71\u4e00\u5b78\u7fd2\u4e00\u500b\u76ee\u6a19\uff1b\u6bd4\u5982\u696d\u52d9\u90e8\u9580\u8f03\u4f4e\u968e\u7684\u5100\u5668\u53bb\u5b78\u7fd2\u751f\u7522\u90e8\u9580\u8f03\u9ad8\u968e\u7684\u5100\u5668\u3002\u9019\u6a23\u904b\u7528\u7684\u60f3\u50cf\uff0c\u53ef\u4ee5\u6492\u51fa\u66f4\u591a\u4f4e\u968e\u7684\u5100\u5668\u5230\u5404\u500b\u90e8\u9580\uff0c\u6709\u52a9\u65bc\u628a\u8272\u5f69\u5373\u662f\u6578\u5b57\u7684\u6982\u5ff5\u71df\u9020\u8d77\u4f86\u7684\u540c\u6642\uff0c\u80cc\u5f8c\u9084\u80fd\u6709\u4e00\u500b\u7cbe\u51c6\u5ea6\u7684\u652f\u6490\u3002<\/p>\n
\u4ee5\u9019\u4e00\u6b21\u7684\u5b78\u7fd2\u6848\u4f8b\u5df2\u7d93\u80fd\u770b\u5230\u4e00\u4e9b\u6210\u6548\u3002\u4f46\u5c31\u5982\u540c\u524d\u9762\u6240\u8ff0\uff0c\u64cd\u4f5c\u7684\u65b9\u4fbf\u6027\u8207\u7cbe\u51c6\u5ea6\u7684\u8981\u6c42\u9084\u662f\u5fc5\u9808\u627e\u51fa\u4e00\u500b\u5e73\u8861\u9ede\u3002\u5176\u5be6\u9019\u4e00\u6b21\u6e2c\u8a66\uff0c\u4e00\u6b21\u65b9\u7684\u5b78\u7fd2\uff0c\u4e5f\u80fd\u5e36\u4f86\u76f8\u7576\u7684\u6548\u679c\uff1b\u4e09\u6b21\u65b9\u7684\u5b78\u96d6\u7136\u5e36\u4f86\u4e86\u66f4\u597d\u7684\u7d50\u679c\uff0c\u4f46\u5728\u90e8\u7f72\u4e0a\u76f8\u5c0d\u8907\u96dc\u5f88\u591a\u3002<\/p>\n
\u985e\u4f3c\u9019\u6a23\u5b50\u5e73\u8861\u9ede\u7684\u53d6\u6368\uff0c\u8af8\u5982\u6a23\u672c\u8a13\u7df4\u6578\u91cf\uff0c\u6a23\u672c\u5206\u4f48\u65b9\u5f0f\uff0c\u6a5f\u5668\u5b78\u7fd2\u7684\u5de5\u5177\u904b\u7528\u2026\u2026\u90fd\u9084\u5728\u5617\u8a66\u7576\u4e2d\u3002<\/p>\n
\u53e6\u5916\u4e00\u7a2e\u60f3\u6cd5\u662f\uff0c\u7528icc profile\u7684\u64cd\u4f5c\u6982\u5ff5\uff1a\u5c0d\u8272\u5f69\u8a2d\u5099\u53d6\u6a23\u8a13\u7df4\uff0c\u900f\u904eicc profile\uff0c\u8a2d\u5099\u9593\u7684\u8272\u5f69\u5c31\u53ef\u4ee5\u4e00\u81f4\u3002<\/p>\n
\u540c\u6a23\u7684\uff0c\u5c0d\u5169\u500b\u5100\u5668\u9593\u505a\u53d6\u6a23\u8a13\u7df4\uff0c\u5efa\u7acb\u5169\u500b\u5100\u5668\u9593\u7684profile(or look up table)\uff0c\u900f\u904eicc profile\u7684\u904b\u4f5c\u65b9\u5f0f\uff0c\u5169\u500b\u5100\u5668\u9593\u7684\u6578\u5b57\u5c31\u53ef\u4ee5\u76f8\u4e92\u63a5\u8fd1\u3002<\/p>\n
\u9084\u6709\u5f88\u591amachine learning\u7684\u5de5\u5177\u9084\u5728\u6478\u7d22\uff0c\u9019\u8879\u662f\u500b\u958b\u982d\uff0c\u9664\u4e86\u5728\u5100\u5668\u9593\u6578\u64da\u4e00\u81f4\u6027\u7684\u61c9\u7528\u4e4b\u5916\uff0c\u6211\u4e5f\u5728\u60f3\u8457\u80fd\u5728\u9019\u500b\u7522\u696d\u7684\u4ec0\u9ebc\u5730\u65b9\u4e5f\u80fd\u7528\u5f97\u4e0a\uff1f\u6bd4\u5982\u5370\u5237\u6a5f\u7684\u5404\u7a2e\u8b8a\u6578\u600e\u9ebc\u6a23\u95dc\u806f\u5230\u5370\u5237\u6a5f\u7684\u8abf\u6821\u9031\u671f\uff1f\u751f\u7522\u6642\u7684\u5373\u6642\u6578\u64da\u600e\u9ebc\u6a23\u95dc\u806f\u5230\u5370\u524d\u90e8\u9580\u3001\u7ba1\u7406\u90e8\u9580\u53ca\u54c1\u7ba1\u90e8\u9580\u7684\u64cd\u4f5c\u7b56\u7565\uff1f\u9019\u4e9b\u6578\u64da\u53c8\u600e\u9ebc\u95dc\u806f\u5230\u5ba2\u6236\u9a57\u6536\u7684\u559c\u597d\u5ea6\uff1f<\/p>\n
<\/b> <\/p>\n
\u64fa\u5728\u773c\u524d\u7684\u662f\u8d8a\u4f86\u8d8a\u4fbf\u5b9c\u7684\u8a08\u7b97\u80fd\u529b\uff0c\u8d8a\u4f86\u8d8a\u5f37\u5927\u7684\u61c9\u7528\u6a21\u7d44\u3002\u4f46\u524d\u63d0\u662f\u8981\u6578\u64da\uff0c\u5373\u4f7f\u662f\u9019\u985e\u4f4e\u968e\u7684\u5100\u5668\uff0c\u4e0d\u7ba1\u518d\u4f4e\u968e\uff0c\u5b83\u7e3d\u662f\u6578\u64da\u7684\u958b\u7aef\u3002\u6216\u8a31\u4e5f\u662f\u500b\u6a5f\u6703\uff0c\u7d93\u7531\u9019\u4e9b\u4f4e\u968e\u7684\u8272\u5f69\u5100\u5668\uff0c\u8b93\u66f4\u591a\u4eba\u5efa\u7acb\u8d77\u8272\u5f69\u5373\u662f\u6578\u5b57\u7684\u6982\u5ff5\uff0c\u8b93\u9019\u500b\u7522\u696d\u7684\u904b\u4f5c\u53ef\u4ee5\u518d\u5f80\u524d\u524d\u9032\u4e00\u6b65\u3002<\/b><\/p>\n","protected":false},"excerpt":{"rendered":"
\u901a\u904e\u6a5f\u5668\u5b78\u7fd2\u6e1b\u5c11\u6578\u64da\u5dee\u7570 Reduce data discrepancies via machine lear […]<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[],"class_list":["post-3561","post","type-post","status-publish","format-standard","hentry","category-uncategorized"],"_links":{"self":[{"href":"https:\/\/fredkuo.idv.tw\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/3561"}],"collection":[{"href":"https:\/\/fredkuo.idv.tw\/wordpress\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/fredkuo.idv.tw\/wordpress\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/fredkuo.idv.tw\/wordpress\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/fredkuo.idv.tw\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=3561"}],"version-history":[{"count":1,"href":"https:\/\/fredkuo.idv.tw\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/3561\/revisions"}],"predecessor-version":[{"id":3562,"href":"https:\/\/fredkuo.idv.tw\/wordpress\/index.php?rest_route=\/wp\/v2\/posts\/3561\/revisions\/3562"}],"wp:attachment":[{"href":"https:\/\/fredkuo.idv.tw\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=3561"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/fredkuo.idv.tw\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=3561"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/fredkuo.idv.tw\/wordpress\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=3561"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}