Wolfram Language

Traitement d'images et de signaux

Vectorisez une icône de carte de bits

ImageMesh convertit les segments de premier plan d'une image raster dans des régions de maillage codées par des polygones. En partitionnant une image en couleur de segments, vous pouvez alors transcrire ces constituants en polygones et les accommoder comme une version Graphics de l'image originale. Ce processus est appelé vectorisation et constitue le processus inverse de Rasterize..

Une image d'icône pour être vectorisée.

In[1]:=
Click for copyable input
img = \!\(\* GraphicsBox[ TagBox[RasterBox[CompressedData[" 1:eJztnQd4FNXax0M6oZO+u9mWQgsdUaqA9N4JJQQIvVdBOgiI9JZCr0oV9epV uZbP67VcRS9NipQIKF0IhJIAIfO957wzZ86U3SSbQBRznpd5zk52l905v/2/ 5ZyZsQ0Y03mwu5ub2wRf2HTuP7nx+PH9X+tSGh50Gz1h2JDRgwa2Gj1x0JBB 418a4AE7vxT/kb5Q2ApbYStsha2wFbbCVtgKW2ErbIWtsBW2wlbYClthK2z5 2n4+9nPC6jVDBw3q1KFjjSpVbaYwW5i50P4yZgqLLle+S6dO/ePiVixf/sMP Pzx58uQZYHP40KE5s2a9WLOW1RhmMZqspjC72RJusRbaX85g4MBgBGEcwapX qfrqxInffP31UyLn0P8O9ejS1Wwwwv9Y4N+90PLdCEtUENq1afvF51/kIzlp aWmTJkywGIyFUvN3MEAIbFD8wMuXL+cdnl9TUl6u38CR5tgtVpvZYgkzm8NM ZrLNhVmeocF/B/BHWG2qzw974Cs84w/zJzErPSaOKAKP9kKNmj/99FNe4Pnm 628qREbp/i9moykgILCsf0CY0RRls1ewh5dDC1eaPUJp4cr+M7IqUeUqR5WD uNFOmZctzFzOZq8UEZnr97RpOtgXzc71tXvsaOWljmpP+fww+j7sg0n/u/RR y9vDw83W0JDQMgGBwSEhcGR0hQh+X/vf2e8aPHv37tUlxxBqAGyiI6OGte+0 bfLUH1YnXN259+6779/e9y6xPe/c3r3v9p79t3ftJbZzN1rqWztvv7UzdQfY 26Jtfyt121upW3ekbtmuY5u3ibYJO9vFziZu/6atqRu3OLdbG7ekb3vr9vrN DSqUdytSxFvZ3NzcGpSv8GjL9jvrNqaup7Z2g9hhD8GSN6QmrU9NZkZ3wh6w xHWpiWuJQT8hOTUhidpasl2VmLomKXV1IrGV2Kc7VySg3VqxhnTYnuVrbi1f k7oyAba3lq2Wje6/tWzVraUrxD2ks+rWctphO5UvIe+2YjX5ePCBVyeTz7Aq iX3a24nrTi9dfmDq9CX9BnSq1zDMaCwbGGjRUIQB9uqVq3ILzxeff66FxxIW VsY/oG616htenXL9nXeFf/9H+Ppb4f/+LRz4LOvjf2V99Amxjw9kffixbO// E+3Ju/8Qbf/7or3zHrG9+8Gy9r0L9mTPO7yJe3btwyeQzq69ZA+8ZM9+8nDn XtF2KY3s3EPsrd3CnneF9z7sXqceoOKpaUCUr7f32RUrBXjbHbvh+U/e2vXk 7d3MMre/Ldu2tzK3gm3P3KbcuWUHMfjT5u2Zm7YR27yD9DdsIduNW0XbBNst ZOf6zbJt2Czt2ZS5jrO1G2XDPckbMpPWky3tPAZLWEf2sJ3M8CXwhuvpm5PP s530YSd00LZsE7a/Lex+R9j/D9geW75qXOeuFlNYQGCQLkJbt2zNOTwpKSlR 9ghFnACyExJqMhgXDhry4L0PhS//I3x04NE//vnwvQ+IvfsPsIx972bs3S9u qaXv3pe+ay9axlu7M97alf72bmp7JNudsWMn20/6zLaDvU1s29vYSd/6VvrW HdLOt0BVyJ80Bs8Rbcv2h9veEt55b0zrtgiPl16DPyX1jwd+Hmzelo62cats 6zenb9gsb9HWbSK2ge3cJO5ZtzF9LbXkDdTWk23iOtmS1vMPHySsfbA68cGa 5HTorEmWLInrJz9YnUSeA9tVicRWJ4od9pDtWZlAbFWi+CbwkoS16QnwDgnk Y2zYkr6Wfir4eOTDb0kHZd60NQOOEoC0F35lHxxftSam8SulAwJUQgQIQZT7 3Xff5QSejIyMZo2bhCvFJzAwqGq58t8sXS58/mXW+x+KnIDteYcaBQY5ge3O PYAEse07ZSPDjYYDLT6kA71DfAh40HEnBuO4aSsx7OBOMNzJ22a+T59Mhp4c HGHP/uR+A4AQdwfwID/NK1cRduzKkDHYRA41M0IF7SAPaykSMDQAAxo+REsk RgcdjI7+qgTJpLHmd65Y82DlGrGD/ZVSn7dlq+T+8tXiw5XcX9FUz8F3XrUG /l+CDaAOn5Z8yHXyV9soHls4/kSL3vvwzfhBIcEhZpNJFQu9UKPGjRs3suVn 6eLFfLYVQeFpVLPWxc1bwU9lvPOeCM+edwgnu/fxRvbs4LFB6dhJwGD8MEJ0 DQCAn8aGLSIMPEX8Q4RE+hGJL9m4VexvJA+FnXu/nDqjiLs7OClH8ECDJxTz 8U1ZtESAl4vMSMAwQnRR4YxoCJUR8aHIT5KSH9SHBJmWFWvUD1esEWUEAFi6 kjAAHYQBO0tWEMM/geHDxcuJQYc9GZ/P3pDSm560jnw1hlAy5QeggoOGwr5j V+ae/cJH/9r96pTQUINZmXRDRjZ+7Djn8Fy9clWV5EKIXqNCxd8Bng8+IZAA NlRhZJFR6YzkX6grEcEmYHBuRYGBrDDbiPF/gj48GTlBZ0Fsi9rWbyGKsV76 ExUQYcv2G6sTTQGBIC+6msPvh/7avv0E+MBJ1MWI22yYEU10MQqDncLaDRng O1ZK2KDxoiHqxmpRLvihRwNIEA+tAS2LlqlN9Rx4OUMI/4uVCRShDWSbuE50 ZHjoqAqRUYPIYdde4eN/7ZgwMTAo2Ma5IeLFwsz/+9//nPAzc/p0PmyGgMpo MP538VLhw09EbCBKeYvGKtt3iuRI2qIvJqzDRxf86DPR0OLBQguJCse2UXQ6 tPN47UZh8/b21Wto4XF3J2cZxMXFjRw5kv2VuLDoysLWtzIcoUJDEYZKOtfn dCYRxeTh6kRh09Y3O3c5NmP2E3jaitUK58LrA0KC0qFiA3fqcvLmUmLa/SqD 1zKZQkOG1ySlr91EfDGvQngYkaLtOx/t3CP88+PXYnpBuqSSoAH9+juC586d OyrxKRsQuLBfPCiPRI7GN6lo4VUFOdm0VVSG9Vz8uZ7nZ4v8+VW2doP6T2Ls Sv2L+BwarCZJkQkNXIW3dy/v2VsbMyM8I0aMgC977Ngxcv6Jhwe6sKI+PhcX LhHWbVaSI3klkR+Ok5UJSsck/swzVqwRNmxZ1KUrOcFl7AQhab2+pEAH+1rp YMyo4GHYYIc3FVTQYUyy/4hpEXTgu6zdRKP3JJkiHBoYDurLssDXvLWrYbUa kDfxEgR28uRJXX52bN/BRz6gPC9FV34Ab7Vzr+SSuBhGRcumrTqehTkUWUA2 igEGDj0fporZwQaRDQLGeodPQGCSNSFKwlph07afZs7x9PRQhT0Iz4IFC/DL PnnyxG638xK0qmdvYcsOpapI4S50VkqBrhIY3gg86zYv7twV/eOe+MGEH9QZ rQ/Suh4GjJYQpd1fuOT+G4vBHixcQkyXKAYhvi2vRSvWiPE/IoS/EXjIfBl1 HMK+9w5MnxkUHKJwSUbTmwsX6vLTtXMXlfisHzZK2P8BgQeBYREsC3Gxz2IP ltuqZIQAo6FFZUlcUpMswqATwUqciD8cFp/Qfua6TU+SN1QzW1WeC+GZP38+ /32HDBnC89OwXHlh/aaM1VReVHHvKil6EZNlTdC7fHXG8tXC+i0iPB7kxLrV 3XoISRtkWthQOqJIFxXKiUOb/yYx7C9YRAz7lCsCmMrTsdgbPjNgA3hTR5xO +2KOKcVFDzduEXbtbVenXggnQaAwdWq/qF3sAWk777wg9q4cGXV97QbhrV2i b9qopy0IDPMmjiBJWi8LCOKBH9UJGypz/lfJIIaZ0bY98VyagHn48OGqr/zB Bx8wHwcuzMPT8/SceULSOp3kmkdIz9IpPCu79YA3hLfC/3FGi5ZC4no1JCyw UaECw41KImmLM2B07fWF9+e9qfsqWaMUAfYK8o3QI0OCj79B/lDTSGDnyNFl uaIi0SKz5cyZM6qD+d233/LOKzA4ZGCzlgKEPSySUTGjrZM4IceRjCSuk/WT S37Fb0H/ij8TRo4YwfKJj1R/E9Zv/t/0me4eHjiCPDxt27bVjffKli3LS9BS UIyN2+S0epVDYPhcKX3ZKmHtph1x/YjySE4TukPqNxDWbtTRGaXUiHLBg5Qr ZpjNmS92YD9q0bw35IcIEvvfmR5iJL9SKk/hsZV+sxDL/bZoWcXwcDNXUTQb jHv37lUdzA3r19u55/gHBW8ZMkzYtpNLcCTDuCWZi0acEMLrA9ZdxYxGDEqp s0jUSYGxJKuKRphbUT8/8SFNmetFRqo8Fzy0Wq1paWm6Lhu44vmpGxEpJKx/ CEeS5wRDUFUYLEXC6UtXCskbdvRVwIPv1qV6DQHYRp/FB7pICA4x73G0/Mxf RCRl7gKCAW55m0uZ0Rr+FeCBl8BD6DDeqEdT+zUWXROQ1rDDK0r62o3tX6wT HGqQXZjRNGfWbNWRnDNzNguT7BZrcEjod7PmCJt3OBMWNHyYsA6zWoUxxcCE V0xhEnWBcWZMDbAvRrZcXWVlgrBha1Kv3irPVaRIEdjz3//+VxceaGvWrGH8 FIGgxd392LRZQuJakRCOE9H4HGrx8vQlK3ThQX4aRZXPgqAaBohFuTwkjiSF EQUdZEbFBlLBbPY8Yto+PA0MHsIWISSdNxRaxChCr4o/EywcrRQpAn4mt20P /ogPgbRZ/LgxYxg/ljBzpM1+auEiYdNWOWjhAxh+EocXB0UHHFOS6GXEgkmi IiKVn5+ofLnGWL68ms7y4E4uGhES1l57Y0lAiZK8+JDpUS7h0m0pKSkQWhfh nM78dh3AGemX7FDtmfIsXi4krtOFB9+qmiksY9HSDBYDw8Axz4L9eUqTtWUB sblKDBAP3AOdWa/LzPB93thLmCFL85BV8jEeoCJh9IVahDiJirQKUoDEPn3L BgXz/HRs1151JOP79WdzXmajqVq5cpeWrHiyfpOYFgEAietYJUQx1quklITX Cjbuq6QIbRW3Z6UklTjds5Kr22vzYngmxiF8NKKcJwLxGfpyY63nql69uhN4 sMFzeBf2gtUmgGioZIfV9KSZAhGe2DhdePCtwvz978xf+IjxM1+ChDkU2RMt kDtMXhgDjB/gBFGBhzPmkP7MuaQP21nzqL0uw4N/ZS/h8RN9nBRyM0O/hiIp ZYvCyoT1feJ4fmymsKaNm6gO44C4/lzyZaoSVe7iwiWZkIzIU35JLMRSVFxl bKT4RBzfNTIh8swOFvC5Giz7q2oGkC/y42+BPZMvhUHik7T+p8nTSNrDhc3o uQ4ePJgtP+PHj+ddWBEP9yOTpwrwRbT8SCFuxpvLhDVrt/Xp6wgefKuivr43 5i54vHi5LD4EoTeo0T7Sgh6KkaPSEAQAYUDDh9PnEITg4Qxq02YTIzvpQ/LM 18kTZtA9/Fuxd2ZhFd9hDm6e6EmFpSvW9eqTO37CIHkHfhZnQiTAg8GEhXmc larJHWWSy88Layd3xKB0tT4//JQQmxlkc4jSy0EoQAdaRUdrxUebsOu2L7/8 kq9UQ39e23bwnrLaKMt64I+E1clLOnZyAg80nPE/O2Va1uJlHDxvyirEVIXX HBQZxglsAQmGCqNlxhxx//Q5916bCeTcmzoLjDwkndkiTsymz6avnUOeCQ9n KkFSeUnoIOdkz3xh0bJ1PXrniR+UmpVrNJagHncc35zwwxv7kzZqZX3Vn6T5 HSFx/YdDhpMVGnQmgjFQunTpnCw2gHbv3r3g4GDehdW0WAUgU6+gBzsh1lrS oROr8zjhByToxzHjBfBfoua8IQ4T/t75EIWJjEpqOE5wzz0EgErNvakzAZ57 r81AeDh+6E5i0hOmTMe/EkPAUJfEd35d/QHgU6F7nTlXWLB4XbeYsiG54sdc OTLq4oI3MxOSpbpZgloW+PSW8cOPMjMVGNpcWPUzZ2VSLtPRiWaXrHi4bJWw KrFeeLhWfFSlZuetffv2Mj8eHp5eHidemy6AMCrheQQxT8LaxR066sKD4Tp7 6EEfHhg4hPDDnBQf0/I+hfdN6Hc0/CA5BA+kgkECHeSH7QdaiFFsEJ7JxEQH B28iPnMWytE9pHG63BfDqgUgkvOFeW+u69LdFX7mv5kJysM7DjZlvJxzOipC +GyF1VpVD/nn8FmkqkjLnszPEHHvAF7m7bgBqnlSeBgQEHDnzp2c87N+/XrV co5lHTuTEIiD5/GSlbBnctOmuvDg/EiNGjVU/OyP7SdAaoPBBgs/RGZel0MU xg8yg0aZEQed+ikcdBEMZIOJDDx8dRox1gGbPE3mZwp7Mv/ymQqTIAR/R+iC j/f6QmHBorWdu5XlpjByxE9E1MV5Cwk/WnfDHAq/fomfreOVRPUnvnSvO+mj 2snoYlVcqZYLavBk6cqqJlMexQfauXPngEA+i68fHgFZWLo0Wfl40XJ4OK5R Y+oo9eGZMmXKyZMnGcye9H3WdeoqvLmMY2OurCeSx5G1Zdos1AQxnhEHeoY0 rDMoBqKYEENIJk0VbcIUsoU9E18jfdgykHiWyHaGTJQoVtI7i2jNFD/AtNmu +6/X38gEz6Ust+pIh2rcddei6M4jS8VY1TQQDpm4X/c96ROElYlv9YmjBUOF +AQGBuZKfLBVrVpVdmFFivj5+v46beaTxcvuLlgEATDEkH1r1tZdB4uJHq4J OX/+vIqf5W07wGvlMJgzkR9Gixy0SIZD+SrlBIcbtkgIGgKj4gewQX7QoI9/ xc7kaWr88M3xf0Hk2P9F/zthxty1XXv451Z/gJ858zMBHl5MVGsDNNPEPEhi hRNngaXJQbF0Js0OKxYhSMbv1O2ApS9c8ujNpVWMOuIzffr03MIDbeLEiSoX tql7DKSuwqKl6Qve7Fq1uhN45s2bh2/C64/4YZo0lfmZocOPRI4Uz0yZoaYI FYaNOKMFOeEfjp9MDJjBDm8IEm55ulRPYB32nuNeFSZOW9uuY675Af81e14m yI4WGKYbrCbPDzq3kEBRpceqAj/jo2vixB+dR16wWHwhLlHg1ioIKxP2aiIf GE1fX9+LFy+6wM+//vUv1dC3r1RZWJF4Z9b8JhFR2gl9Bg/vKy9cuODh4cHK UPDXcfXrCwuXSvxI/kvWHArPVCn8mMw8iOS5GDAoDjwwDCE27uNevTd2krzV GqNlzERi0tPu8g9xy72hMGlacpsO/qF54IebtZHn/rBiiR1VWX6+VB6f96Zc 98CJYMWfFqnnEIktEosP7L9gM8gSh4CosHglhCha8YmNjXUBHoGe4+/v78+7 MGPpMsfHTaoVZtYugvWSVlCrAq1r166VKlWKLwX0q1FTeH3hg2kSM7DlZWeq FNWwmETlWRgnTBYYM0xwwHDEERIAgPEzZiKyAdu7oycQSGA7esJdye7RneSv I8fdHTWe9OkWn4l9YdLUte06uKA/F2bMfvwmq33haL4hLglghmMtl1Xf4IBR lsexz3biln8hK8yyGWf2f4n/tVh/E5Ys/xxrPpz4YBD7/fffu8YPtNatW7Oh h/f19vYuW6y4E3g2bNigeoerV6+WLFmS56dnlarCnAX35eRaSrQnc+kSMsO8 FY8NMsPDw3slST3u8tKB/CAVlAfZABI01cOR8JDbOULaP3zs3RFjhXGvJrdu l3v9ibzw2szH85dw4/imGh5WDZvHleL5/VrTzvJoJ5T5yT5VdX3uggfzgJ8V bSvpFJxr1arlMjzQEhMTFdOvtBbkCJ6NGzdq30HLT7OIqCczX39ASi5S+swn 4NpkStdJsegFmEE8OJERxYQ6IDUzo1WcABtj04aPkfAYkzZsDGzvDif70chf KTbwpzTyhNHCmInJLdv6c+s3csrP1JmPFyxRrjnhSFBVv9mUinYi2NEaA9VL +Fdp1ydIO4U3Fh8dM4FMdnHig7W7zZs354WfU6dO8XPxug3h2bRpk+47AD8l SpTg+WkSHpk1e366WCvmKjDosJjP4gNjFTy84PDwMFqooyGjz+vMCESCaosE D/JAbPgYuU8NmEkbNpb0h45CuztkJG6FUeOSm7d2lZ9FMjBaKlSlVDZ9oyqU qaYCVfODOfyrZBCODq9TVys+pUuXvn37dl74gVahQgVdh8X+FwDsww8/dPRy iH/KlCnD81M5JDRj+tx0rP7xmsNXZhAePjNifUfMqMgZPhZFQ8JG2iMpDBGT YWN4PBgkdwePAMM/YV80+NOg4XcHDhOGjElu1soVfqbMeAzxiaO5YNV+spxg nmJemJ8HZFOBrK+a7mFTPFr28D2pPZk9/8qkqWWLF9cu1ho4cGAe4YHGnxSm hQek6aOPPnLy8gcPHthsNgU/wSHpU2enE5+FIoOmLNrw7onFwFIkIwa3XBgs YoNhjCg11DERJRmjtNEiHkNHUhtFkFDZwGF344fJHWbwwgFDwVzn59Xpj19/ Q5rzVc7IzOT2TJ+tLqjS5FRbMRNfgmX56cqZwRlzNXvmiOsQpkszfTPmCguX rGzTXjXEHvRkh6/z48J9/Ip6FTxFixb95JNPnL8c+FGcE1TEze4fcH/KjHTw U6o0ipHD6jB8oo3Y8KjwzMiaI8YqYqxLfNBoohsqAxJkWoYrIAEDSPoP0TH4 E+0Ig0YmN22Za37CIy9MmPIYfv7TpUl/biLvvjQLzBXBZvEFeWV9bJbMj1QA kSfv4Ani6gIFQvd4CClFGTNfz5w9v7pezbBKlSp5h0egK+r5BJy9P0Q1ObkS V3p6ejg3mQvhd8miRW+Pn/KQFYf5mjCtzqmwUfgplgTxeAxHfzSaoUI6IDXg jxg8vLwwT4RGqBiqi01av0G83e03mHTiBgnxQ5JeaeGK/oyf/Gj6XC0hcgTI rRxQoyKNvr4QcfywrWJuiCOTPV94/c1P+8XD6Hh4earGd9GiRfnCD7SWLVvy /IDPCgoKcn7SN2t3794NCwvj+Snt55c2ccpDnHdgTgoMWeIzbi7vVvMjmcTP WDGYYRQhG0xwUGcGDpMRcsBPGoXkbv/BWn7S+g4k1meA0G9wUuNm/gaX+JnB +NHMzjhEaCYryIv80JlBcdnA1JnS+gF+4YFyHQJDlHvCfeBnweJYzSnt+PC3 337LL36WL1+umsjo27dvDl975coVRf7u4VGqqN8fYyY9RP/FuyrcKmNjkZ9R E5iTShOZEeERE6XhY8T8CINeBga6qoFKP0VjGNKBPdjvR2i5OwA0Z7CIShw1 oAWxiR1AjMKT1quf0HdgUqOm/sZc83N+/CSZH746qpqj0a1mYEFMXIsyQ547 niJ15ClCB1jiO0t0Zc58/dKEyaX8/Nzc1ZFz69at8wseaCdOnMBQmb1/+fLl c3htbXX9x8PDx9vr5OChWfCV+ckmVUqF1RtWKx6BhTsSFTOdEaMdKadWBDYs kkHNIWwMRVNKCvqjeEoI1RbsAzlASx9qEjPEevdP6xWX1rOvEDsg6eVXcs9P xPkxEx5Nny0OPcsXFEsCpssTfMwmc8sA8E9TNO/AmGHrCnSNQ1GYuyChVTvV PBSWffbt25eP/AjKLB6D85SUFFf58T45eJjID5ui0vLDQmXYjkC1oZW9keNE P0XDGwU8YKpImCA0TIpnBivgAVTiaAc5ATbAgBbsgOHOnnHiQ+gAQj375oGf yPOjxj3SVrdUa054MNgerKOqiOLfQfW2rB7CpoG4VXNgD6bOFGbMrWu2aJ1X SEhIRkZG/vIzduxYeSKDIrpjx4488DOc8KOaX+CxARsplfvEio3ED60Dy2qD wTBxUsNEP6VFyBE/GM/EDhQRYiARTuKIxcSm9YglzMC2e5+0mL7EuscKPeOS GjT2Nxpd4Qcddw4Nf2IIA/Zz+ELG0uTpaqNCJMyce2jICDdycoSHih8Y6/yF B9qBAwdUc/E5XIqvnj8l/PicGjQsC74azw/LzUeMkzRnHFfxGy1rDrHRYmys SMOHiVGNmhyMh5WxjUjOAAU5THCoyCjg6dYb7U7XXmldewkxfZPqN/I3mlzk h18uojWc6Ed4MAnVfQ4+TbuHn+5R9ZlMTZkhzF4wvWEjlfjg8okjR47kOz/3 7t1TlZEbNmyYkxeePXvWTbqmkBddQu/h6fl9bH9h0jQ5POa9FcY5dCpKLONQ kFitWE6m+ASKr9twmiN6KDEejpfJQSNRTZy4BU5IX9IcrQE8XXqiCT37JdV3 SX9GjHkEg8gyBX4BCfPmjBnd1Sa6+1VP4I0VRpBJdF5TZmRMnl5OcyU6eNig QYN8hwdbly5deH5Kly4NUGX7KtX6MbIE2t39sx69RX5U8HBTD1g6lvVHFSRj Js6nVDTj5qp8g+UECix2oJhA8ZqDUoNRDRCC7gldFagNbCXlYeTc6QwWI3SP Tarnkv4MG/VYVVHnK126+3Wfo13C5Igf1Tpe6tSEqbP+r2dftyLq5X85D0tc aBs3bmQkYAj9ww8/ZPuq06dPa/n5vEcfAX4I0hob9doJPbcluiqy5XwWV/HD jFuu27CUiuVQxD2xCEdyUqKr6quQGuQHrGtvAgyS0ymG9GHbKUbo2tt1fmAo cWUaDwz9KenULriwENcgiX32cm7NmzMVUi6tFGbOG/XCC88mcmbt4sWLqhX1 69aty/ZV3333nQ4/3XoJ46dIyyekejJXGFQFPHKoDKacaNCJh1mfhTR8Ao7w ICcxfeXYmAlO9z6izpBtDAWmB9l26EasYw/YCp1ikurUzzU/4ZHnh458DF9c XF7CqS6biOGlmJ+sYcbWubEqB8MMIXTi3eA5414Vpsw8ET+0tJ+f9pI+U6dO fTrsiK1evXq8Cxs6dGi2L/nkk090+OkaI4x9lSzFGSkVdpQlZdlhMWwGaQIe vSkGViJW51O94qSUSgJGNBISqw2xYfx07E7Iad+VmdCpe9JL9Vzg59fBwx+N mSRN8krLA0ZqDH9QbNnAiLE6z6HwyIUyaW3kXalcr5r6QaUSJk29NnJc+YAg beTsla81Z902d+5cnp/GjRtn+xKH/IybLK6swNhYtYJCNdfJ5sEpPGks1GH8 MPFhpRs0ICcGA5tYJTbECCooNaAwICydeohS00mSHVFzusvwtOtyp01noUNX F/kZOOzRyPH8j0XBzwhuPz9N44gfUuVQLqHkCyBKA34eTpiSOX5yI4tVu6AC 9sTHxz8dauT2888/s2QKQ+i7d+86fwlGTexzcvy8KgvO0NGKiSq+pIzKg3Ey V0NWKA+fjKOfYtazL8ue+BycpOHYYbENQIL8MD+FtHCyc6dtZ2KtOwntXeGn SnhEyoDBD7F+JWaUYyTJHSNXRDWm4E1Bl/QOasDGq/eMGn9/9ATh1WmDqlXX woPrnI8fP/7UwBFbVlZWxYoVeQkqV67cV1995eQlqvNYRX669BDGUH4k5RHn zVWLuMRJz+FcDXCIDjkKP9VPpKhnXxkepjkkJJbcE82kqOxQqQF+eD9FdUbk hzJDDDptSEdo2zmpdl1/U274MVN++g16yFXOxS87mH7ZIVIhVDUjww4Lx5IS sDFs4Yq4wAkXsbDwkkqcMHFqwivN3fROuYKdXbt2fZrgyO3VV19VzaVCmzZt mqPnL1682AE/k/isXD5EbPYTp67ihytqOyI/gxV1YzbXoJIdFhVTbNJ4teks dfjwBpUH4UFm2oiCc6dVR2LQadkeOlmtOibVeinX/NgjUuLiH/LSymcE/KoA +ld+RQqPFltJq5GpMQp+ho+ROmOFcVO+6967iLv6MpgY+UB0AWnyUwZHbHiN cdVyDnhYt25d7QVIoUFIr+Xns87dhDETFdNY0rJA+Rjyc+UDWCV5sKw8bNKK GdZzWBkHHRZ0xDgnRg5vMBPvTI0xw7yVpDOi5lB+brdoR6w5Mdf56TuA8DNw uNqwfj6QW6GEW97Yn6RZG9XKWy1IqFdPRk+8OXR0RBl/refCqajRo0c/E3bE hiGNtnrg5+e3fft21ZP5S1HJ/HTsIoyewIsP0XDtUZVlRyInTirsxHLT4n0o PGRmqo+Yg/MxDw8PSE1nCZ4O3cXYmOcHIxyEBwQH1KZle8IMkNOsrWhN22S1 7JBUq44r/gv4GTSSBXLkd8HP88IexRHgiOJ38qvglPOAWpYekLONJneJKu8o 7OnevfvTq/k4aps3b9ZFCNqoUaP4Z/bu3Vvh7zw8vL28jvbplwmumZ/A4sJj eX0O2Tnkbr+BiiQrdoDaW4nTmlLRTxvtADAY3shJVg851FHw00WUHeqwJM1p K/PTtM3tJq2ymrdLqpF7/QF+evd7CB6532BpxZFy3aPO1x8qL1IaOEzMQ5lk ibrEzR1L6sRAEkZPSmjcVLsCGeGZMGHCM6RG0f75z39iFVr7qcCXsfsU9+rV S8WPj7f3id79MsE1M1keqISHCo54PPsPTOs/kOhP3wFynsVKOljVUUXIkuzI oXJHKTzu2E3OqgAbFBzcguYQ2ekshjoIT3POgBywxi3Bspq1TapRO5f8WKrY w5EfcTUs/YJpNJyjAivOvFClHSxlmrSDyyNZyZ1f2sQqG+whB5IwfOyJXnHF fLx171tRgPBg++ijj7QIoUs1Go2HDx+G5zRv3lzLz7FecZnDRstBsnIFOzmA /TnNIb9WDJV5t8XNksfE8tNVXA2whxIe6qrQQyE2bWiEjNi0FgOe20BOyw63 W0o+C7GhskMM+XmldVKNF/y5a8vnmJ+4hwPgC8ZzZXP8jjj5ImWXcfGiyyZz vvG8TJGjoYgMObGKl1YgUJDSBw3PGjK6nlF9sjkOEMQVBYGMun3yySf86YG8 LytatOiXX37Zrl07FT++RH84fuKHsVAnTVxEOgSPmzyHFctFy7g4J4ab7uRS LZEcVjfGbUcu1GH8oOYwtSHMUEOHxXwWMPNKa7Jt1ILC00Lkp7oL/ESc69k3 I06ayWVr2Fjxk+wcLM72xnFLTfrJLIk/qAHcOn8VPwPEFQjCiPFv1mnopnff Coh5CoQW3Xbw4EF+hRj/Ob29vf39/flrwOL655vxQzLYlCjAg/oMh0JU8sFy baevkp+ecaLyEKlRRMuKYqBKdkhKRWMbERvmrUiELDKDUoPYoOCgzwJsmrS8 3aj57ZebkT7dZjVqkVS1liv8xMRm4HcBhFTfkRkQgnUJRIjEe/3FWWD5OfKK JlmdmPcfMFQYOvpI995Fvb20a5srVar07ANm5w1cFcSHWoSK0Kb4/JSfWwOG ZgyVzpJg8aF6SoI7vEx8xAydqwpKSbrMT+eesuxgtMPScLEG2FnMrfgguVkb ojPNJH6gjz4LyGnYTLQGr9xu0BQ6T+q/kli1pgv+61yP2Ax+ei6Wq2L1HSgv 0ceVtKi9vajkxg5QY6aa+OMWdd8DfgaPelnjueChj4/PiRMnCgQS5+3XX3+N iFBfPUbbIJALK1nqzpAR6VjwUdYGRaFmFR4860E1K8GvtcC5cpzrZNOdbD4C EyuxEthJLAAyhwXygg6LaQ64J/RWosNqLhrCU/8VsNS6jW/Xa/Kk3iuJVWr6 h+Wen26gP/Fy/tiH+3aYF8RK5dDYgfKagT4aeJhSsV8ZO2EkbpAwdGRywya6 C+O1F0j587Tr16/XqKFzf0wVP5UDg9KHjnqAa91Z0sovTlb5LHndhXJKC8SH TYCqqjpY6mHFQJaSQ3iMaiMGOVKQ3FwKdWh4TDstZNmpT2QntV4Twk+dl1Nf akj0p0oNl/jpndELq51c/sjWQLLfiBjm0QSBVddjpcIXW9Sk9GiI0+OBQy/2 jivrV0zruVq1alUgYOS83bt3r1mzZtpqA/8t6hpMwvAxD0jYw5a1Dxb50Y15 2PEU15dypR7gR6wnY8DTQzFXzs9EcAiRmIdRBIaC80obUXa0/AA8gA1VHtgS fuo1SaxcPdf8hIef69wzI6avIvIHX4xfiixGwqRSubatV5yi2MU0mdXB2NJK yNRAfIaMGFChotZz+fn5Pe3lGfnSMjMze/Qg94xz17uWuLunp6+X19bGzSHA u9dvsCKHjdObD2UrvmJidQrLmGqBBKmiZZXy8PzQyEcUn+ZUhfj0qgnNsEic 3JyEOtRnEX5eaojkiPrzUuPE6NzzYws/1ykmA5WTfRcCEvewh7RoX21xCmni ieLcnDBw2JdtOrjpXcln2bJlBcKDa23MmDH8KYeK3wItXm1p9IowbNQ95rb4 MLKvRnlUcxMY9oieS5qbUC7xYmstGDkKzWFBsiraadiUYIPwoElhD4EH7EVi T2q/nFipmiv+q2OPjB6xiiI5rTmI83Rswk7UKKkjihKVKbYQTiRH7j/o3T8z buALwaFa8YmOji4QDPLS3njjDe0ch4xQEbfFdRoIQ0Y+wDII4ydWqjCzdaf4 A2T8sHVfxNi6nRjF0gtWUqaFHTU8zdrKrgoM1AbhAWyoh0LNIbJTvwmEPamS 20J4Ul+o7zo/7bukd5Pgxzon7ZNqOV88Z+VQ5uOQJezzi29jYhldQv+hyXUb qA44FngPHDhQIAzksW3ZssUhQvR7TahaQxgw7FHcIDml5bMt5rliYjXKgzFP d4dui5sDvd2Cqwo2VcIDxsU5hByIk+tJ2FBvRbYvSvDUbpAnftp2Tu/SS6mc MXL+SHBiP42e4i8F0OJqpPAEqdLeS8SMKlVGr7ibMbGG4iW0l4Fq3rx5gYx+ vjQgv3hx/UtuulO/3M0eAcKb1W/wnT7x6lXuPZUVZnnFaYy8dIdVlXFWq00n eekFK+80Q3LacNi0wEqyKD4NxPScuCqmNig4/JaKD+GnVoPEilXLml3ih03m dpJi/s49FWd5MH7Y4gGKlmI5Ad3KFbDufYS+g6dWUa8tzPsFVP8M7ciRIxaL +jxrMbSjP5A6waEXu8cK8cO4GQpJnNG/8z6rs2Z9Msuz2nVRBzystgzGBAfs 5WbEW2FVWart3EbBQWPAgNWqh7KTWrMu6des+6RG3cQKLvBjP9emc3qnHvK8 P6+fSBRLJ+U1/D3l/bjsVkVUt96ZPeNOt+9W3MfXTXNKRf5eSaOg2tWrV198 8UVHeT3sN5co+e/W7cGX3Y2NF6fU+YRLKztY6mFxsl6qJdd5mvHi00LSnKZM c2S3BcrDw0OwaYDMEKteh2xr1HGVn/Bzrdqna5erMRNXmHDraVVcqUyiSOgz IJ5e0f35Ex/WHj161K1bN0d5Pfnunh4b678MQWAGhEA9FDMUHDl6k+l8bq4b KvO1wcYtRM3hcytkRkrPRbVh2CA5zKq9mFm1dmKFyq7w07JdOoOHLVdj60nw S+HCEpmo7mqueJnq0jMrJu7HFm3JbSI1J5Pq3qL9L90mTZrkLClzcxtdsRKE glk9+4nwsDwdl36pKoRah0VNDnvkVItLsjh4bvNBMkuykB80FTzUnlR/yRV+ bOHnmrdJx/CsTSfFShJ+6SwzhS51x1MXtVAJMXExVrub3iKNb775pkBG+ak2 PCNDF6Ei9Fs3CAm90LGb0Kt/muz6lbKDxxkPu674NJeUB6wxnT1vRN2WMs+S ZYdFyAAMwFNb8lk1ZM0Rt9Qyq76QUC66rNnsOj9YCWdL9NtIHV1TgYQdeliy uvb+d+Pmbu5FtGeyv/TSSwUyvs+gffbZZ2XLlnUSUYf4FfuowStCTL/7IkI9 FOUd1ayWPCXaVtYcJIcv72A9kCIk5+agOWB8eq51W4gN6bx0q3ItsMzKNROi XOKnWev01mJgL5/ToWs8P0ydmNej/i6tY3ehe2wbY5j2SLo9hWuI/alaSkpK lSpVHEbURUic9HqVGk+6xz7u2ltRW+antBwFPCzPQs0hi3aay7KDzDB76WXZ W/HwID+S5tyqWvtWlVq3Kta4ValmZvQLCZGVXODn7CutHtAVRzL2yp+AYp6O GXN2begXl35BWZ1j/t2ouVsRHfGx2WyPHz8ukJF9Zu3+/ftdu3Z1NM2BvqyN yXypbSehS2+ZHEW2xcU5aGyVKa674BdgAD8MHua50DDOQYdVQ+G2CDNo0TWJ AT8Va2RWrJUQUdEl/9VW1J/mUqiGOEmLkWSieBVqwy1eYvLbtrPQsUcbg/rS zV4u3W7yWTaQjqNHjx45cuTw4cMnTpyA3Dwv7zZz5kxH4RD6MlOJ4h/VbwwI 3Wf8sLXKuvNZDB5QG34Nhqq8g6ZNtYCfai+JmsOwYQb8lK+WWb6aK/xY7eea tUkHd9ymEymJN+ewb85WlYjBvyJAaiOt6mdC1LpTVvtuXzZs4uahzmbhx+jn 53flypW8DMrTa3hTOb5BJBMdHR0bG7t79+4c3h9c1cBT+/r6Olo4BPIE8eH0 ipUfd+qe2aE7NyshYdOCi5YJPy2Uiwabikk6Lzss2kFmWJ5VA/mh3qoyx0yF 6qg8AM+t8lUzy1dNCC9f1mLJBT8WaxWz5YS9fHrthkRtaKlcUWFoLi1Dai6r E0sn2clod1qSTlqrjkL7bkR8iuiIT8+ePfMyxJ9++unvv/+el3dw0gYNGsQP NOaJPEvwhGPHjuX2bUHQ8BKvuuEQ+rLGwcFnmrYROnRP42uDDBs51eKSdL7I wxDiC8tgfJ5OlUcMdZCfSqLmUIqq3SpXGSyzXJUEe7lc8WOxWCtarJ8Gm1OC zVes5e/UrHvnldZ0eredLES8IvGGbo4j6knrTv+t39jN010V+eBwQHriwshm ZWW9/fbb4eHhS5cudeHlOWkQkvE3Q9HRCqnFxMTkdp3tvXv3Onfu7MiX4ZuX 8fXdVbOO0K7rA1yAirNaWOFp0lLmh1txSgyxYYVlkJ0XlXUeEvC8pAiVefFh /JSvmhd+oi3Wf4Vafww2g50ItlwOi0ytVDOt3it3wJeRX0QH+RRFMBVCuMgW 5ahFO6F1pxiTWVd8IiMjgYRcHfnMzMwNGzaUK1fOLZ/uueOoqS4m5qjhGfHQ Jk6cmJNrJPJt3rx5zhCiCz/iLfbU5u2Flh3kmAeTdF1+IOwRy4NSbRmnsZiB w6r+EgiOyE/VF0R+KnHOi7othOdWVOXMKOAnd/4L+KlksX4eajkWYjkYYvku xPJ9cNixINMFg+1mVNW02g3uwHchctRRXMnPzjvT8POodacjDZq6e3kW0buS xuuvv56rA75z504kx+1pXjwT2+TJk7NdIc9/F2gREREQMuXqfzlw4EBAQIAj UHHWvnzJUl+/1FBo0eEezqQTfmiezhV5RM+lKhKyCg9OZqHPqv4i5eclWuGp SYo8lajmEHIoNrCNigZybkVUuhVeKTO8UqIt1/oD/HxB+fmJ2o+AUIj1h2Dz 0SBzSqj1ur3i7aovpjVodqdZG9GRoUcGaeVZat5WaNtliFXnVAU8YrpXsdBt 4OZwUhKbv7//U426wXmZTCZV8JxDisaNGwcimfP/67fffoPfgqPU3osG1R6e HnMjyz9+pc0TnFtnhcG6jXG5u1gkfFHK0Ik1kGbS6zF+xGgHDDpEeWjOhZoD 23KS7ERGEwuvdCsiOjMi2gX/peLnUIjlcKgVDED6AfYEhf0SbP7NFHGzXNU7 NevdgV8Epvn8IskmLR82bXfu5WZ+Pt664tOoUaOcHF4IjyHGxqEB5HCm9eOP P84FDblv169fHzVq1IwZMyZNmtSlS5eKFSti0uRklLGhO6tWrdrBgwdz9T9O mTLFmS+jQtQoMOhkvcZCs3ZpdNEpiZDxRAk+WsYpCT5Vp1JDYh7qucQ8HTSH j3kAHtQfUB7GT1Tl1HLRmeHRidbIfOHnSKj1KLUjFKQfqRydC7Ves1W8XblW Wp1GdyCuA0Uiy97IIm2hZcdpEeW0xyTnp+ckJSXhpbz5m6qPHDkyV0OTL+3s 2bNr1659+eWXnVDEYiES/ZYp8+233+bqv3jvvffwy+r6Mk9aqS7p45NcoYrQ uOVDyo8i4WIBMxopD8pTEqLgID8Q7ZAgp5qYqqPyEGCiZXIiKqXaKlw3R6Ya bYtNNn+JDdf4AWPwgP0sGYJ0JMRyMth80Wi/EVX5dpUXbtd5+U7jlg9atL/2 SpsQcm6Oeg0DHKJixYr98ccfTg7mjRs32rdvr/pJQt9utz98+DDXw59/DbLv AQMGeEku2ItLxDw8PCpXrjx48GAgAXh79OhRbt/80qVLTZs2dSZEHiSo7hZs OP9iA6Fhszss5nlRsWiQKA/B5iVNnl5LzNNRcNBzoeDQaDk1Mvq2vcJNc9Ql Y/g5g/VYqPVMiGWu0RaUf/wck+A5Qe0kNfjrkRDr8eCwMyHmS6bwG5GVhRca rIqqBF9W+1uCg9OuXTsnh/E///mP0WhU/RJxdfSnn36a20F5Gu2XX37BFT7Q SpYsCV8nOTn51KlT+fLmzvMyrFQHFfXbV7EqIPSAT9XZ3ASfoaPDqiwpDzIj p1rECDbhlW5ayl8xRVww2k+F2mCUf6LZE/CzwGQNzBs/hxzzAwb/HdoJuv94 qPUsUBRqq+jr5+apFh90Xnv27HF09Hbt2oVBjpa6GjVqpKen58sY5UuD7Gn3 7t3Xrl3L93f++uuvIyMj3RysQPPC7N69SN8Q043a9QWIglB5MENnk6GQnlfm 5yMwTpYDntTy1W5HRt+0VrhsigRszoTa0Kcc4WKVlNBnx89pg+0MtWvG8A3+ wSDouuJTqlSptLQ03eP2/vvvO/rpgf4ULVo0MDCwbt26ixYtcqHw+9dqDx48 wAK4IyFyx1v0+hX7AISoXpP7LzQQAx6cFQV++MIy1pYrEXhSy1WF8OaGrcLV sCiIOs7Rn/zPyu0xabjPPSt+fpHgOWuwXjPZmxQroRUfL6eXZElJSQFCnGTK /AwCBKsA0qZNm+A4P/3BLLC2f//+kJCQ7ITIfUio6WbNOkKt+rchyapamyhP FamwTMuDqaA2tKRzw1L+sin8PAwTCTyIyzhpwCBE9CAn8omf/wu1/OyAn2PK +EelP5eN4R8HmSBn16YoCMDbb7+te6ych46qxpIdiJRmz579NJzIn6Rdv34d fnFOjgyZMiviVr54iU/LVRZq1btXrbY4qx5dM7ViTYiKb9pJJvW70X7BYAO1 gWFiP3no/yI91PJzJA/+6/8MRMq0/LAUTOW8fjGIn+emKbxfybKOxMfPz89R 5jVx4sSIiAiMk3MIElMk8IkzZsy4f//+0x/PgmkQFqIQOSwzuheBgGFciCm1 YvWs8tVv2Stet5S/ao66bLRfNNhAcCCfOiP5iDMSOb9w/JykzuuE5F+O0opf XuIfXf3h+TnO6Q9+kl+NtiOhllJe3m5e+mevOM+8oEEWs3bt2tatW+PF4nJV +w0ICNi8efPTGsKCbjdu3IiPj89GiNyLlPct9nGgMd0YcdVo/5WOyHljOHRS DPZz1M4a7AweXn9OSMOaX/zoxj88Pz8r4x+w68bwWWWC3DwcOGs3N2Ajh4fr 6tWr27dvB94YSI5iAB6hP0mC//QafEFc/uG4WO3u5ukxvFRZIOduWOQFA4l2 GD8AD4rPaU6FHPGTF//1ueP68zFOfzAEgu0vNHI+b7RVKAppu97KFnr1eBcu zHL58uXk5GSIlr1oFuYIHqDrk08+yadR+lO3jIyMOXPm+Pj4OPpNFaGVRotP 0d2BRkAIXBjAw/Oj8mKMH95/4ShD/DzfZHWhfpgtP3wKhtE7RM7bAwy6aTsO MTCQl+OWmJio+6PDWHr37t15efO/XDt79mxMTAwKkf7KfHBnnh79S5ZNMVpv GiO0/JxWxj9sNBk/x54yP6oo+orR3qZ4ad3IGfmBRMnlw3XlyhW7XX3umJd0 7uqCBQvyNhp/1fb555/Xrl3bEUXuohD5bgoIgbzmkjEbflT+C+uHeeeHIOSY H/zfz4favgoy+xDx1PktYFZ16NAhlw9U/fr1dcUHdvbt2zdvg/CXb5CdRUdH u3F3f1YLkYdHj+KlYQRvmiJYIqZN3nFAj3GBrsv88Pm7Lj+q+uE1Y/i4Uv5O IufIyMgnT564dnxAuHR/X7CzSpUqLsxUPpdt6dKljnwZClGAt09C2RAYqd+N YvLFEOKd11Ep+cpffpgLU/EDdjbUeibUZvMtqpu240C7vNz08OHDunUP2Onr 65tfs5bPR1Pdw0VXiFoUL/lNqOWqKVzFjzZ5P0pG9mnxwyN0xRC+wT/EUeSM Y/3++++7dkwg6tYeE6wc5rwa8DdpkN7Cb8rJIjcPKkR+Xl7zygb9ZiSxkJPg 5yiRBVf4ibZYPwu1HFXGP3wIrVrCAZFzi+KlnETOJUqUuH79ugsHZNOmTY48 V4cOHfLjkD9vTXXakbZ5UoRmlgm6Ygw//TT5OZYzflIM1n8Hmb1o2cERPzlc rapqjx8/1r2clxtdgfb0zvz6Sze886+TWivA06F4yesmuyr4yUd+SP1Z8l+H KDxsCfQR5fwX+LJLBvurpQIciQ8O98KFC104FMnJyY5yrhUrVuTHwX4+Gy7I 1x8LT0+zT1EYvovKmS++8sMj5LL+fBpqgTdBcrQpGPtfToEZrJE++jVnLylz /+mnn3J7EB4+fKh7Bp8bva+Ky6nc36Gp7iTOWhHKz74g0w3quZzwI+tPiOX1 3K9fRX6OovKwWXiOH4bQBUP45oAQN09nkTP4IBdS7B07djgKm9977718OtLP Z7tz507p0qV1fnoe7q+WDkg12U9rZt5PaJzXEepuzuaBnyP0zC/dEtBRaa3a ZaO9YwmHkTPy06NHDxcOgm7BEPbUqlXL+QtV57Tm9hTX56AdOHDA29tbVUsE eBoWK3HJaPvVYM0JP4fzxo8q/9INoeFjfBcSVtzLW2eVPDfikEPl9gicPHlS GwSi+ODJX3fv3j127NiHH364Zs2a1157LS4urk2bNi+++GKFChX8/f0DuQYP wQ8CdfXq1WvZsmW/fv2mTp0K4dOuXbu++uqrEydOwK81V2f8/cnbuXPntDe5 A7fl7+3zQ6j5N6N62uIkV4fhPdfR/ODnMOe/ECFefy4b7LNLBzkRnyJFisCv 4JdffsntQZgzZ45WfOCt/Pz82rVrFx0dDfkXf76Vyw2XVZtMJgCsbdu28fHx QNeqVav27t37zTffpKSkAF0Fe8ZQrlpGRkblypVVh87di0ykbvE3XDOFs2U/ pwuIH76KeM5gr1m0mHPnVbFiRReOQ/Xq6guMM4TY0IM6OVrOkZMGagYvx+lX Jw2IBfmqXbs26Bto1+TJk4Guffv2ffvtt+fPn//jjz/+VKeE4EpX9Sh4uI8u 5X/LFMEvG+P5Ycvmf1bCA/ZLiNkFfrTzp1p+zhqs7waa3PTWOfP8DB06NLcH 4cyZM24OpgKfTfOkzYM253SBp7Db7UBXq1atYmNjwZMCXfv37//xxx/BjwBd z/L+m/Pnz3fT1FoBnrrFil8y2s9zC1aJceLzM7fyJ1/4Qf3B+PmwMgVj/EDm Fe9gnbP84d3cIMzI7XFwVPb5kzRPqWVLF4ReEREREJW1aNGif//+06dPT0pK gpjt4MGD8Bu5fv16PnrGjz76yE0zS+jmhWGP5QqdqvhFWnaoCn604nMkn/g5 KNUPVfwcJx2LwdfXUdnHCy+07uUFP8PcHoo+ffr8mfnJYXNOF+RHAQEBkZGR EHcBXYMGDZoxY8a6deuArv/+97+nT5++du1azmtcZ8+e1cbMtNrjsT3A8Icp nC1b1dUfxg+/cgMH3WV+PpX4+YnjhyF03hCe5B/qpOzjRfmpVq1abuGBhktZ dN+T9yzZ/vxz0jDCx7fNV3yyac6Df19fX6ArPDy8atWqrVu3hhgAEgpIY//5 z39+//33oF23bt1ih+vRo0famBknuSbSag+/YOw0xw+LnFX8oPjgcOeFn6OU n4N6VaDzRluL4sWzdV5jx47NLTwQM+DKXhjWbDOsUqVKGQwGyNlr1KhRv359 SKAgDoEYMt5B69GjBzwBRgR8CgyN0WhkF2bRbfTuiM+UKy/NhRa1DRJGoMti sdSsWbNbt2516tRx04uZmxcvedVkT6HVnjN6C+bxhB2t+BzOJ36OUXJ4hFB8 ToTa/i/EXIzIYzYH4Z133sktP59//jk7UPA+wcHBqsx69+7d//73v48ePXrl ypXU1NT79++7Vr3JysqC1968eRPE/4svvti8efOsWbNGjRoF+NWtWxdGB4YJ s7Nnj5CTpqu6qufA7zrS1/eUwfobt05MN/Jhk5jHlJHPoXzl50dOfCAcSjHY p5cOzFZ8QEYuX76c22E9cuTIvHnz9uzZAylM/kaYuWqPHz9OS0uDOAQkVDtA ysHyoOYJ5u5V8KjBxyju5f1FcNh1Y7jqVAsnZ1v8rCc+lB/LPFM++C82kQpb +E9rkLKPs/wajnn16tULZOjzt33zzTduDooJgIqvt3e9EqUiixYr4+Pt5+1N rljoQa6MQc0dT8V6lmgVoaXCbQGGW6aIswarNmxWrXZWxczIzyEuXHFt/SE/ /4USxN72ZKjtnQAjHBDn1Rk45sOGDSuQEc/fBhoIUZauBNEY1eMDc4RQqfrv kZUO2ct9boncabStCTFPCwyNKxPQsmTpaL8SYb5+pb19vPFGqOQsY+QKjaBF 7geWP/CQsGdemeA7BB67lh8Gz0ml8hxT8vNT/vHzk8TPTxQh2KaE2rMt+2Dw s2PHjgIZ8XxvEGk7cmEgMi1LlhEqVL9Xvmp6VOVHkZWzIqsK5apkRVXOjKzy ODI6I7LKrYiKv9jKfxcW/kGodUuQ6Q3/0BGl/TuXKPNiseKRvsVCvX3pBKIH 4UqBlgeiVcTLwZpy1Seh0jetdOBNUzjEzPypXlp+jnPwnODSriNPjR8sJJKy QLDF6lvUSdnHSzrV9MKFCwUy3Pne+vfv74gf8BfgmL63lXsYVeVmOLngrWiR 0anhlUSzVbhnr5Buj35sq5hlrfDEUuGJtcJja3mwNEs5csWAEOvnQeb9wWFr /EOmlA6MK1m2afGSVf2K2X38/L19fDDE8pAAc2C+Xl4zywTh6Tlns+PnBEVI lXkd1gQ/yM+8PNd/fpT4OR1qW+vgwlCK3wJd4lUgY/00WlJSkpMQGlzSqLLB QlRlvOAtXjaZ2U1bBWLWKGJmtaWao9JMkQ/CIh6GRT4Oi3oYVu6huRz0H4VF wv7fjPbjBut3wdbPgswfB4VtCzAm+Yck+4cmUWOdBP+Q7QGGH4Kt14ny2PES TKrrJHCey8ZOOT/uOG1ndirE7ML1D7X8YCEafiztHa+T5/np169fgYz102hf ffWVm5MrOHl6BvoUvRQefS+SXogyQsTmlr0iGMHGVv6mtTwBxsJRZIpQ2Q1T xHVq14zhV43hsP3DSPanmiJumwCnqDumSGa3qaVK9gc5pUu8vAZL1R1EPvKp Xs75wSz+pKv8HAg1H5Y8F/JzNMT6dYiltJePo5O8sGHws3HjxgIZ66fR7t69 i1fW1f++NGpNCDEJ5avcpMwQbJjmIDAWbhsWKW4lcv6g2Kj4AbtiDL9M7ZIx HLQI7aLUuSDZeWM4uzbLGYP9F8f8nFSWfbT8HFLyA+ay/vxL4ochdCbUvqBM oFt20+JYczt79myBjPVTanj1e4d66+lew6/4vfDoVJEcKjgIj0XpsxAb2Er8 /CFhg3YVTAkPGgKD24scP3htn3OGHIkPf41TVealjZwx4867/2L8nDTY6vll M2fhRZ1XeHj4c7a+fciQIU748aAh7seh1ke2iqKrQrWxKMlRas4fEjzXqOZc l8hBeK5IysPsoqQ/F6Qtu7APBjxYXmYFQ9Uiw5Nc2KO72ueQEh683cDJPM+f osH/+GGgyZOeSZ0tPzExMQUyyk+v4Z2UnUxkwM+qW8nSgrWCNkhWkcPzI/ss jezw5EBsA4YdROhXSg67KhTGzBgwOyoYHleu8+FrPirxYbXig1R/XOPnX+SS 4DI/KaH20aX8sxUfDH6WL19eIKP89NrBgwedhNBeNJEv6uV9whR+31JOxkYv VEZsbkjwXKOhMq85jB+GDYt8mJ2X+GGXpDsrgaRYpGGQxYdVe45TeNg6riN6 4vOTxI8L8bPZYq3Mrf+hmRf5X8oXzabsw/jJy3Va/pwtPT3dSQgtSpCH+7Qy wYK1oiw4XJz8hybUEQMeTbTDNIfFPExzINrBsAevJ5YiFplZwcfq6MKYzG2x RYZHsot8fnRVf8xUf74xWI+GWP4bYiZvEmrdFmDIds7CC29oHhLyXF6TuVGj RtlNpHpafYv+ERYB+bWoNhQe3lWhn2LkXNMLlRk8v9OYhyVcCA+Lefi50TNU iPiwR+m8ZHiO64XNWufFisauxc+VLNavAGm6yPD7EOvZUHvvEmWydV7IT7Nm zQpkfJ92GzlypHN+MJHfERD6JCxKFedcV8Y51ziEVDHP71yq/juXql/kEvZz HDxS2i5GztqwRzVboZot1XVebMbqJ5f8F4t/TtEzvOA//TY4LMjXJ1vnhfzM mjWrIIb3qbdt27ZluxYIfmKNipV4GBZ5S89bXeMyLN4ua/hhtFzUBMy/csyc Ufa1ysOqhcdzBk8+8oP5F/yn8AWXlQ2GX1aOZvHc3D7//PMCGd+n3fCURudn hbhTL/afYEtGWLmcwKNxW+F8hs5rjjJatup6Lk3OLpeatad36SrPYem2O2ze IY/8wPucN9qa52DOAuEpXry4a9f5+fO3hw8fBgUFOXdhKEHxJf2zzFHMSakK O5c12PBu66JSdrKFJ1vl0T29AvnRKg8vPgddzb94fuA/+iTI5OPl7eyujRw/ tWvXLpDBfTatefPm2fIDiXxpb58Uox2i6OtKzXEOz+9SwKzl5zy5m5tdZbr8 6CqPo9NzdAuGzHPlCz9nQu0TSzu7to+KHxfOFvwLtfHjx2fLjxeNoheUCXlk imRO6qoDcn6TKjwXuDxLQQ533wFaJNSRIG3MfJLeyUt3bTx/bo7WDirFJ4/8 HKX6UzEHZR/GjwtnC/6F2u7du3OynB6eUcnX74YJsIm47IAclqRfkjRHZaxC eFYZHp/WeK4zDhKu43phM0vYefE5xEU+PDx55AfS9h0BhiI5KPt4SWvGnrNp U1VLSUnJNoT2wkTe02NvoOm+KUoVHvOuik/Vedk5z81NnFXGxqf1ZIfnh/dZ xxzAc0ijPyp+fswnfi6E2nrmrOyD4lOuXLnn+2I7EEKHhYXlyIV5urctXvJ+ WASLeX5XosIzo5iYkO6VI91uQCwv655JoVKeE1yR8Bh3pVzdE3MYObzbUomP y/zg/eO+DzIHeOeo7IP8dOnSpUCG9Vm2Vq1a5YQfXLT8faj1lilCiwrGPKpV PUDOeTpDwd9oSRUtn1ZOT5zWxMw/c9Ojx7g7mTqCh8XMWuXJIz8/GOxLHN+J SZefNWvWFMiYPsv22muv5YQfLypB40oHpJuifleGxBdEhMLZ9AT6LCSHKY/E jOKmJzw2SI40Q2pjeZaq2uycn0PKU0QP5h8/3xltDf3072GqbXhS5Pfff18g Y/os2z/+8Y8cnpEKTwrx8T1vtF0zRmj91AVNbZC/xRtbknGGi5+1a8O0953k yTmmPJ9du7ZZxY+uucZPbYttU6DJyzNn9OAdyYOCHN1V+XlqFy9ezOF1q3A6 bLV/yF26El5lWn7OScvA2DIMVaqOAc/pUHLpA+2qQszZtdk6g0fLDzsz4kfH FLnGTz1reLdSZZyfYarip0kT9Xs+l+3Ro0cRERE5d2Ev+BW/aYq4pNQc8FY8 Odo7A2qlRqs52oRL5bMczXAdUq7QcBT5uMyP1WKtZrEF+hYtkuMzbp/jaVNt a9euXQ75ccd1rUGmVFMkv9ydZVhsAQ9bA8YvgOej5TOaIuEpZZGZiY8qZlaX eqRS86EciI9r/ERYbTazxcfHJ4fXncA1Y3+Te0dCe+ONN3LIjxeVoN4lypD7 kEonSoDpTmbxJ22dViqPCpuTyllR3et+6wfMuLZQyrmeHj92s8U3x/y40duj uHCpjb9oO3DgQC748fIs7u39U6jld1lwxJmIs5o1YNoijyPZOe4AnmyzdX6G PSf2bPhx7Tpjf9GW7T2S1MfHw3166aBUYwS3VtmqNS08pzRxMj2BS5yeUIU9 P3MFn8NKfg5JFyFk/GhlR6tIP7q6/tAFfoYPH14gQ1kgLSsrC++gnVN+PD3D i/pdNNguKm/CflZZITyrKRIyfviV8Nw9i9U5+8/K09iRH3bXG3nyQpl/8aeI 6vKT2/XPueIHg5/ne9pU27p27ZpzfvAyL5sDjDckCaJmVVULtXfWPq0sFfJn cvE1w+Oa5Oswh8oR3YhakqYfldGRNsF34fpjueIHi2lnzpwpkHEsqLZs2bKc 8+MlrWu9YRKX8bC0nedHe8UM/irfKuPh4aMgVcGQkXNYuh+lKin7kfNrupbD +zcNGzwEmHGNn4iIiAIZxAJsX3zxRa6uiwiJvLun5xch5ivGCL7m40h/Tmum t7T510npGizHNfwwio4pC9H8dVpwFv6IU3iQnwUafpo3VZ8i8erESYwfaja/ okVzyM/f8PbZly5dgpQztxI0tJR/Gi3+sPxdqz9aL6Z3Pg7RHz6cdjT59bNe gMR7tKPKPVo7F2qZrOTHagrr3rWb6oAsWriQ5yfSZi9ZokS2/GDwk5ycXCCD WLBNe8nlbPjx8gz09j1psP7O1X9UppvRa88HxD47pZ0/yYKPiI7rwZNb+zXU 0ivMFsLzYzSNHjVKdTT27d1nCzPz/AT4+2c7y4OX2jh27FiBjGDBNud3ytZH yMN9UZmgO9SFMWZ0JeisZubrF82cO0vEVAueVUQdV/JzLDdGX2upY7aFcfxY jKaVmpuHQgAMXPEhtNloyn6hppubwWB4zi61kcOWkJCQa348Pav5FbtstF+g 8Q9SlMKZahaeIaR7PYSTEjZI1EmptKi7fv64Y4/mxMB57TdYTRarXRH/mCH8 0x6QKhUrcfEPQaiYn182p8v9PdaM6bb//Oc/ub20OF3X6rk/0PQHJ0H8iYHa VRy651kwF/aLssaIeT27TpSjyY4cGiD3e6h5rPLizxDkRNnsuqdoTRg3jg+B gJ+QwCB3py7subzURg7b1atXHV3X19kR83TvWqJkGp0OQ1qwo10574QiR1Nj J5T8kD8Z1KcQ5txAfL4zWCuYyfUQ+OAnplt33QPy3bffWjgXhgj5OZYgl2+s /Ny0WrVq5ZYfepkXr4Ohlus0EcNLIvw5+bkSah5vsgVyPGDws3fPXt2jAWFM /Tp1+CdHQhRkMDq6Nw0GP48fP37Go/bnafHx8bnlx0u6CfK9sEggJH/5UVle +Pkt1PyegSiPVZm5V6tcJTU11dEBeWffPngOjxA4u0D/AN3rJv2dgx9s69at c4UfL0+LT9ELZAm9YkUQH/+wWFqFELtIlOrMHW12Dx38qwvwQM5+2GCpa7YZ VOJjMK5audLJAQExeble/XDlqyCX9y9TRosQHLrExMRnNlh/wvbtt9/m5HQw VcN1rRsCQu5Ii8ryix++z9Z+5Iqcnyk8J0LNLcPsQUoMQFhqVq128+ZN58fk m2++UUdBFhspB5Upi/dZw4OAwc/x48efzUj9Odvt27ezvSiZvgR5ujcoVuJm WMRvGn7YvIYTfhyVhrTnGJ42yJMdObFLpOZsaRVmVYU9yM+e3btzcljmzp6j 8mKoQqFBwd7e3piRwUGLiop6vs8WzEnTvXdbto3eWdLz06Cwm6Zw5/yoTuc5 qwmK+KWturVrLCc6CZPQzodar9JqT22zLUgDD0jK0EGDc1joAy/WtmUr5XSY iJAtzFymVCkvOm0B0ePTHp0/fxsxYoQL/KAEDShZNi0snJw26Jgfdgr8Obqu VcvPKaf8nBEr1QqZYg6OnuJhvWSwXjVYvjXaRoXZTRarQQ+ehvXrOwmbtQ3c 3Mv1G2gRgoweKII39PHydh5K/U3avn373OhNVN1z0zzcyT3CSnl6ng4xpxsj bhvtqQYb2i3J2J7bRluq0XbDYP/DYP3DYOPtusGKdtVgu8o95O2awXo51HaN 23OD7LQAM1cMlpMG616jbViYPcpCfJZN67aMptrVa6SkpOT2yFy5fKVls+Za R4YURVislcqVb9akyZiRI1csW/bW9u17du36u9m+PXvemL+gRLFipUqWLFOq dG6taMkSQ0JMW8MiksLCE012tGRTOFiS9JDsCQsHS+D2MFtDbVV2ttKoeAhv tdBkmxRm62a21SBJliWQnrSlHWiAp36duqdOnnTtx/XgwYMB/fqpwmle1kID gwJKlylTomRJv2J/TwssXRYEOcoe7pqFWK2lLJbSrlqZPBgwE2Kxhinntvh5 CkuoqUvHjnm/mhzIC6iQ1pepFelvaU6OyV/XyHCHmV+bPCW/bj77v//9r1Wz 5iA4zikqtL+6ATlgEPp+9tln+UIOa5C7ffCPf7Ro1gz/i0KQnieD0UTNqV+n zs6dO5/q5d9PnTqVsGZN21atIZdHlgrtL23AT+OGLy9buvTgwYNPD5vCVtgK W2ErbIWtsBW2wlbYClthK2yFrbAVtsJW2Arb37D9P2++Iw4= "], {{0, 192}, {192, 0}}, {0, 255}, ColorFunction->RGBColor], BoxForm`ImageTag["Byte", ColorSpace -> "RGB", Interleaving -> True], Selectable->False], DefaultBaseStyle->"ImageGraphics", ImageSizeRaw->{192, 192}, PlotRange->{{0, 192}, {0, 192}}]\);

Partition de l'image en segments avec une distance de couleur colorée de 1/8 ou plus.

In[2]:=
Click for copyable input
colorDistance = 1/8;
In[3]:=
Click for copyable input
{colors, masks} = Transpose@DominantColors[ MeanShiftFilter[img, 1, colorDistance, MaxIterations -> 12], Automatic, {"Color", "CoverageImage"}, ColorCoverage -> 0, MinColorDistance -> colorDistance ]
Out[3]=

Supprimez les segments qui sont trop petits ou trop minces.

In[4]:=
Click for copyable input
cleanMasks = Map[Opening[#, 1] &, masks]
Out[4]=
In[5]:=
Click for copyable input
validColors = Map[(ImageMeasurements[#, "Total"] > 0) &, cleanMasks]
Out[5]=

Étendez les segments colorés dans les interstices possibles en utilisant l'algorithme de grossissement et coupe.

In[6]:=
Click for copyable input
components = GrowCutComponents[img, Pick[cleanMasks, validColors], MaxIterations -> 5];

Convertissez tous les segments de couleur en masques binaires.

In[7]:=
Click for copyable input
completeMasks = Reverse@Map[Image, Differences@ Table[UnitStep[components - k], {k, Max[components] + 1, 1, -1}]]
Out[7]=

Convertissez les masques binaires en objets de BoundaryMeshRegion extrayez leurs polygones comme FilledCurve. Groupez tous les objets de FilledCurve et leurs couleurs correspondantes dans une expression graphique.

In[8]:=
Click for copyable input
MaskToFilledCurve[mask_Image?BinaryImageQ] := With[ {bm = ImageMesh[mask, Method -> "LinearSeparable"]}, GraphicsComplex[ MeshCoordinates[bm], With[ {lines = MeshCells[bm, 1]}, FilledCurve[Split[lines, (#1[[1, -1]] === #2[[1, 1]]) &]] ] ] ]
In[9]:=
Click for copyable input
icon = Graphics[ MapThread[ {#1, MaskToFilledCurve[Binarize[#2]]} &, {Pick[colors, validColors], completeMasks} ] ]
Out[9]=

L'icône graphique obtenue est évolutive et consomme moins de mémoire.

In[10]:=
Click for copyable input
ByteCount[img]
Out[10]=
In[11]:=
Click for copyable input
ByteCount[icon]
Out[11]=

Exemples connexes

de en es ja ko pt-br ru zh