{"id":2302,"date":"2016-08-01T05:09:46","date_gmt":"2016-08-01T09:09:46","guid":{"rendered":"https:\/\/www.kaspersky.com.tr\/blog\/?p=2302"},"modified":"2017-09-21T13:50:17","modified_gmt":"2017-09-21T10:50:17","slug":"how-facial-recognition-works","status":"publish","type":"post","link":"https:\/\/www.kaspersky.com.tr\/blog\/how-facial-recognition-works\/2302\/","title":{"rendered":"\u0130nsan vs makine: y\u00fcz tan\u0131ma"},"content":{"rendered":"<p><span class=\"embed-youtube\" style=\"text-align:center; display: block;\"><iframe class=\"youtube-player\" type=\"text\/html\" width=\"640\" height=\"390\" src=\"https:\/\/www.youtube.com\/embed\/Y1lnrGIbweY?version=3&amp;rel=1&amp;fs=1&amp;showsearch=0&amp;showinfo=1&amp;iv_load_policy=1&amp;wmode=transparent\" frameborder=\"0\" allowfullscreen=\"true\"><\/iframe><\/span><\/p>\n<p>2010 y\u0131l\u0131nda d\u00fcnyan\u0131n en b\u00fcy\u00fck y\u00fcz tan\u0131ma sisteminin sahipleri (Facebook) bir foto\u011fraftan portreleri tan\u0131may\u0131 <a href=\"http:\/\/edition.cnn.com\/2010\/TECH\/social.media\/07\/02\/facebook.recognition\/\" target=\"_blank\" rel=\"noopener nofollow\">\u00f6\u011frendiler<\/a>: bu sosyal a\u011f foto\u011fraflardaki y\u00fczleri arad\u0131 ve bu alanlar\u0131 etiketledi. Bazen hatalar yapt\u0131. D\u00f6rt y\u0131l sonra Facebook,<a href=\"http:\/\/www.theverge.com\/2014\/7\/7\/5878069\/why-facebook-is-beating-the-fbi-at-facial-recognition\" target=\"_blank\" rel=\"noopener nofollow\"> %97 do\u011fruluk oran\u0131yla<\/a> bir foto\u011frafta tasvir edilen ki\u015fiyi bulabiliyor.<\/p>\n<p><span class=\"embed-youtube\" style=\"text-align:center; display: block;\"><iframe class=\"youtube-player\" type=\"text\/html\" width=\"640\" height=\"390\" src=\"https:\/\/www.youtube.com\/embed\/l4Rn38_vrLQ?version=3&amp;rel=1&amp;fs=1&amp;showsearch=0&amp;showinfo=1&amp;iv_load_policy=1&amp;wmode=transparent\" frameborder=\"0\" allowfullscreen=\"true\"><\/iframe><\/span><\/p>\n<p>Bu Facebook i\u00e7in b\u00fcy\u00fck bir ilerleme ancak algoritmas\u0131 y\u00fczde \u00fc\u00e7l\u00fck k\u0131s\u0131mda insan beynine yeniliyor. Biri bize k\u00f6t\u00fc \u00e7\u00f6z\u00fcn\u00fcrl\u00fckte bir foto\u011frafta tan\u0131d\u0131k birini sorsa bunu <a href=\"http:\/\/web.mit.edu\/sinhalab\/\" target=\"_blank\" rel=\"noopener nofollow\">bilgisayarlardan daha iyi yapar\u0131z<\/a>. Bu foto\u011fraflar de\u011fi\u015fik bir a\u00e7\u0131dan \u00e7ekilmi\u015f olsa bile. <\/p>\n<p>Bilgisayarlar\u0131n genelde insanlardan daha do\u011fru olmas\u0131 kadar nadir g\u00f6r\u00fclen bir \u015fey bu. Bu t\u00fcr meydan okumalar\u0131 \u00e7\u00f6zmekte neden daha iyiyiz ve neden bilgisayarlar ayn\u0131 \u015feyi yapmaya \u00e7al\u0131\u015f\u0131yor?<\/p>\n<p><strong>Beyinlerimiz ciddi al\u0131\u015ft\u0131rmalar yapt\u0131<\/strong><br>\nBeynimizin bir k\u0131sm\u0131n\u0131n sadece y\u00fcz tan\u0131maya kullan\u0131ld\u0131\u011f\u0131 <a href=\"http:\/\/www.sciencemag.org\/news\/2012\/10\/identifying-brains-own-facial-recognition-system\" target=\"_blank\" rel=\"noopener nofollow\">biliniyor<\/a>. Bu b\u00f6lgenin ad\u0131 fusiform gyrus, temporal lobun ve oksipital lobun bir par\u00e7as\u0131. \u0130nsanlar do\u011fduklar\u0131 andan itibaren y\u00fcz tan\u0131maya ba\u015flarlar. Bebekler ya\u015famlar\u0131n\u0131n ilk g\u00fcn\u00fcnden itibaren bu yeteneklerini geli\u015ftirirler. D\u00f6rt ayl\u0131k olduklar\u0131nda amcalar\u0131n\u0131 ve teyzelerini <a href=\"http:\/\/news.stanford.edu\/news\/2012\/december\/infants-process-faces-121112.html\" target=\"_blank\" rel=\"noopener nofollow\">birbirinden ay\u0131rmaya ba\u015flarlar<\/a>.<\/p>\n<p>G\u00f6zler, yanaklar, burun, a\u011f\u0131z ve ka\u015flar birbirimizi tan\u0131mam\u0131z i\u00e7in anahtar olan y\u00fcz \u00f6zelliklerimizdir. Ten de ayr\u0131ca \u00f6nemlidir, \u00f6zellikle yap\u0131s\u0131 ve rengi. Beynimizin y\u00fczleri bir b\u00fct\u00fcn olarak tan\u0131d\u0131\u011f\u0131n\u0131 s\u00f6ylemekte fayda var. Yani g\u00f6zleri, burunlar\u0131 ayr\u0131 ayr\u0131 de\u011fil <a href=\"https:\/\/en.wikipedia.org\/wiki\/Face_perception#In_individuals_with_autism_spectrum_disorder\" target=\"_blank\" rel=\"noopener nofollow\">tek bir b\u00fct\u00fcn olarak g\u00f6r\u00fcyor<\/a>. Bu y\u00fczden y\u00fcz\u00fcn\u00fcn bir yar\u0131s\u0131 bir ka\u011f\u0131tla kapal\u0131 insanlar\u0131 rahatl\u0131kla tan\u0131yabiliyoruz. Ancak, iki \u00fcnl\u00fc insan\u0131n y\u00fczlerini birle\u015ftiren basit bir kolay yap\u0131ld\u0131\u011f\u0131nda foto\u011fraftakinin kim oldu\u011funu anlamam\u0131z biraz zaman alabiliyor.<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/media.kasperskydaily.com\/wp-content\/uploads\/sites\/91\/2016\/08\/06014032\/joliepitt.gif\" alt=\"joliepitt\" width=\"1280\" height=\"840\" class=\"alignnone size-full wp-image-2303\"><\/p>\n<p>Do\u011fdu\u011fumuzdan itibaren beynimiz y\u00fczleri haf\u0131zada saklar. Zaman ge\u00e7tik\u00e7e genel bir tema olu\u015fturup y\u00fcz tan\u0131mada bu temay\u0131 kullan\u0131yoruz. Biri bu temay\u0131 \u00e7izebilseydi, san\u0131r\u0131m \u015fu \u015fekilde g\u00f6r\u00fcn\u00fcrd\u00fc:<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/media.kasperskydaily.com\/wp-content\/uploads\/sites\/91\/2016\/08\/06014031\/256052_900.jpg\" alt=\"256052_900\" width=\"900\" height=\"321\" class=\"alignnone size-full wp-image-2304\"><\/p>\n<p>Y\u00fcz tan\u0131ma, olu\u015fturdu\u011fumuz temayla g\u00f6rd\u00fc\u011f\u00fcm\u00fcz y\u00fcz\u00fc kar\u015f\u0131la\u015ft\u0131rarak ger\u00e7ekle\u015fir. Burun daha geni\u015f, dudaklar daha dolgun, ten rengi daha s\u0131cak ya da so\u011fuk gibi kar\u015f\u0131la\u015ft\u0131rmalar yapar\u0131z. Nadiren seyahat eden ki\u015filer di\u011fer \u0131rklar\u0131n birbirine \u00e7ok benzedi\u011fini s\u00f6yler \u00e7\u00fcnk\u00fc temalar\u0131 etraf\u0131nda bulunan ki\u015filere g\u00f6re \u00f6zelle\u015fmi\u015ftir. <\/p>\n<p>Bu arada, maymun ve k\u00f6pek gibi <a href=\"https:\/\/peerj.com\/articles\/1115\/\" target=\"_blank\" rel=\"noopener nofollow\">baz\u0131 hayvanlar da y\u00fczleri tan\u0131yabilirler<\/a>. Koklamak onlar i\u00e7in \u00f6nemli bir bilgi kayna\u011f\u0131d\u0131r ayr\u0131ca beyinde kaydedilen g\u00f6r\u00fcnt\u00fcleri de di\u011fer canl\u0131lar\u0131 tan\u0131mak i\u00e7in kullan\u0131rlar.  Garip olan, insan\u0131n en yak\u0131n arkada\u015f\u0131 olan k\u00f6pekler sadece y\u00fcz\u00fcm\u00fcze bakarak ruh halimizi anlamakla kalmay\u0131p, <a href=\"http:\/\/www.wikihow.com\/Teach-a-Dog-to-Smile\" target=\"_blank\" rel=\"noopener nofollow\">bizden g\u00fcl\u00fcmsemeyi \u00f6\u011frenebiliyorlar<\/a>.<br>\n<strong><br>\nBilgisayarlar y\u00fczleri nas\u0131l tan\u0131r?<\/strong><br>\nG\u00fcl\u00fcmseme ile y\u00fcz tan\u0131mlama aras\u0131ndaki ba\u011flant\u0131 nedir?<\/p>\n<p>Yaz\u0131l\u0131mlar y\u00fcz\u00fcn kameraya bakt\u0131\u011f\u0131 iki portreye bak\u0131p ikisinin ayn\u0131 ki\u015fi olup olmad\u0131\u011f\u0131n\u0131 belirleyebilirler. Bu \u00e7\u00f6z\u00fcmler portre ressamlar\u0131n\u0131n kulland\u0131\u011f\u0131 teknikleri kullan\u0131r: insan y\u00fcz\u00fcnde bulunan d\u00fc\u011f\u00fcm noktalar\u0131 denilen noktalar\u0131 analiz ederler. Bu noktalar birbirinden ba\u011f\u0131ms\u0131z y\u00fczlerimizi belirlemek i\u00e7indir. Farkl\u0131 metotlarla bir y\u00fczde 80 ila 150 d\u00fc\u011f\u00fcm noktas\u0131 bulunabilir.<\/p>\n<p>\u00d6rnek olarak, ressamlar ve yaz\u0131l\u0131mlar, g\u00f6zler aras\u0131ndaki mesafeyi, burnun geni\u015fli\u011fini, g\u00f6z \u00e7ukurlar\u0131n\u0131n derinli\u011fini, yanaklar\u0131n \u015feklini, \u00e7enenin uzunlu\u011funu \u00f6l\u00e7erler.<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/media.kasperskydaily.com\/wp-content\/uploads\/sites\/91\/2016\/08\/06014029\/image_image_77893.jpg\" alt=\"image_image_77893\" width=\"1244\" height=\"1600\" class=\"alignnone size-full wp-image-2305\"><\/p>\n<p>Modelden ba\u015f\u0131n\u0131 \u00e7evirmesini istedi\u011finizde ya da bak\u0131\u015f a\u00e7\u0131n\u0131z\u0131 de\u011fi\u015ftirdi\u011finizde hesaplanan de\u011ferler de\u011fi\u015fir. Y\u00fcz tan\u0131ma i\u015flemini ger\u00e7ekle\u015ftiren algoritmalar iki boyutlu uzayda \u00e7al\u0131\u015f\u0131r, bak\u0131\u015f a\u00e7\u0131s\u0131 tan\u0131mlama i\u00e7in kritik bir noktad\u0131r. Gizli mi kalmak istiyorsunuz? G\u00f6zlerinizi ve yanaklar\u0131n\u0131z\u0131 g\u00fcne\u015f g\u00f6zl\u00fc\u011f\u00fcn\u00fcz\u00fcn arkas\u0131na saklay\u0131n ve \u00e7enenizi bir \u015falla kapat\u0131n. <a href=\"https:\/\/www.kaspersky.com\/blog\/findface-experiment\/11916\/\" target=\"_blank\" rel=\"noopener nofollow\">FindFace servisini test etti\u011fimizde<\/a>, servisin sadece kameraya bakan y\u00fczleri tan\u0131d\u0131\u011f\u0131n\u0131 g\u00f6rd\u00fck.<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/media.kasperskydaily.com\/wp-content\/uploads\/sites\/91\/2016\/08\/06014027\/rec.jpg\" alt=\"rec\" width=\"1280\" height=\"840\" class=\"alignnone size-full wp-image-2306\"><\/p>\n<p>Bu \u015fekilde \u201cd\u00fcz g\u00f6rseller\u201d ile \u00e7al\u0131\u015fan y\u00fcz tan\u0131ma sistemini kand\u0131rabilirsiniz. Ama \u015funu da bilmekte fayda var, daha tutarl\u0131 \u00e7al\u0131\u015fan algoritmalar gelmeye ba\u015fl\u0131yor. <\/p>\n<p><strong>S\u0131radaki ne?<\/strong><br>\nBeyinlerimiz biz b\u00fcy\u00fcd\u00fck\u00e7e y\u00fcz tan\u0131ma \u00fczerine al\u0131\u015ft\u0131rmalar yapar. \u201cBiz\u201d ve \u201conlar\u201d aras\u0131ndaki farklar\u0131 anlamak hayatta kalmam\u0131z i\u00e7in en \u00f6nemli yeteneklerden biridir. Modern bilgisayarlar insanlar gibi \u00f6\u011frenebilir ve kendilerini programlayabilirler. Makine y\u00fcz tan\u0131mas\u0131n\u0131 geli\u015ftirmek i\u00e7in yaz\u0131l\u0131mc\u0131lar <a href=\"https:\/\/www.ted.com\/talks\/jeremy_howard_the_wonderful_and_terrifying_implications_of_computers_that_can_learn?language=en\" target=\"_blank\" rel=\"noopener nofollow\">bilgisayarlar\u0131n kendi kendilerine \u00f6\u011frenebilece\u011fi algoritmalar yaz\u0131yor<\/a> ve onlar\u0131 y\u00fczlerce insan portresiyle besliyor. Bu g\u00f6rselleri bulmak zor de\u011fil. Sosyal medyada, foto\u011fraf bar\u0131nd\u0131rma sitelerinde, foto\u011fraf stoklar\u0131nda ve di\u011fer web kaynaklar\u0131nda bunlardan \u00e7ok var.<\/p>\n<blockquote class=\"twitter-tweet\" data-width=\"500\" data-dnt=\"true\">\n<p lang=\"en\" dir=\"ltr\">You can\u2019t replace your face, says facial recognition \u2013 <a href=\"https:\/\/t.co\/tW6vdmxPWE\" target=\"_blank\" rel=\"noopener nofollow\">https:\/\/t.co\/tW6vdmxPWE<\/a> <a href=\"https:\/\/t.co\/dKXmOVdJ33\" target=\"_blank\" rel=\"noopener nofollow\">pic.twitter.com\/dKXmOVdJ33<\/a><\/p>\n<p>\u2014 Kaspersky (@kaspersky) <a href=\"https:\/\/twitter.com\/kaspersky\/status\/723502848114307072?ref_src=twsrc%5Etfw\" target=\"_blank\" rel=\"noopener nofollow\">April 22, 2016<\/a><\/p><\/blockquote>\n<p><script async src=\"https:\/\/platform.twitter.com\/widgets.js\" charset=\"utf-8\"><\/script><\/p>\n<p><a href=\"https:\/\/en.wikipedia.org\/wiki\/Three-dimensional_face_recognition\" target=\"_blank\" rel=\"noopener nofollow\">Algoritmalar 3-D modellerle \u00e7al\u0131\u015fmaya ba\u015flad\u0131\u011f\u0131ndan beri<\/a> y\u00fcz odakl\u0131 tan\u0131ma sistemi de daha verimli oldu. Y\u00fcze bir \u0131zgara yans\u0131tmak ve videolardan y\u00fczleri se\u00e7mek i\u00e7in olan yaz\u0131l\u0131mlar insan y\u00fcz\u00fcn\u00fcn farkl\u0131 a\u00e7\u0131lardan nas\u0131l g\u00f6z\u00fckt\u00fc\u011f\u00fcn\u00fc anlayabiliyor. Bu arada, insan beynindeki temalar da \u00fc\u00e7 boyutludur.  Bu teknoloji hala geli\u015ftirme a\u015famas\u0131nda olsa da marketlerden orijinal \u00e7\u00f6z\u00fcmler bulabilirsiniz.<\/p>\n<p>Mimik \u00e7al\u0131\u015fmalar\u0131 da ayr\u0131ca dikkatleri \u00fczerine \u00e7ekti. Duygular\u0131n ger\u00e7ek\u00e7i sahnelenmesi oyun end\u00fcstrisi i\u00e7in bir alt\u0131n madeni ve dikkate de\u011fer say\u0131da \u015firket, karakterlerini daha inand\u0131r\u0131c\u0131 yapmak i\u00e7in \u00e7ok \u00e7al\u0131\u015f\u0131yor. Bu y\u00f6ndeki \u00f6nemli ad\u0131mlar \u00e7oktan at\u0131ld\u0131. Ayn\u0131 teknoloji, y\u00fcz tan\u0131ma sistemine de b\u00fcy\u00fck katk\u0131lar sa\u011flayacak. Bu \u00e7\u00f6z\u00fcmler, insan mimiklerini de tan\u0131maya ba\u015flad\u0131\u011f\u0131nda, foto\u011fraftaki komik g\u00fcl\u00fc\u015f\u00fcn muhtemelen sokaktaki gen\u00e7 k\u0131z taraf\u0131ndan \u00e7ekildi\u011fini bilecek. <\/p>\n<p><span class=\"embed-youtube\" style=\"text-align:center; display: block;\"><iframe class=\"youtube-player\" type=\"text\/html\" width=\"640\" height=\"390\" src=\"https:\/\/www.youtube.com\/embed\/9ymDg2NI584?version=3&amp;rel=1&amp;fs=1&amp;showsearch=0&amp;showinfo=1&amp;iv_load_policy=1&amp;wmode=transparent\" frameborder=\"0\" allowfullscreen=\"true\"><\/iframe><\/span><\/p>\n<p>3D modellerden farkl\u0131 olarak, geli\u015ftiriciler farkl\u0131 a\u00e7\u0131larla da \u00e7al\u0131\u015f\u0131yor. \u00d6rnek olarak, Identix \u015firketi y\u00fcz tan\u0131ma i\u00e7in FaceIt Argus ad\u0131n\u0131 verdikleri biometrik teknolojiyi <a href=\"https:\/\/www.linkedin.com\/pulse\/20140802213837-343420151-how-facial-recognition-systems-work\" target=\"_blank\" rel=\"noopener nofollow\">geli\u015ftirdiler<\/a>. Cildin yap\u0131s\u0131n\u0131 test ediyor: \u00e7izgileri, g\u00f6zenekleri, yaralar\u0131 ve bunu gibi di\u011fer \u015feyleri. FaceIt Argus\u2019un yap\u0131mc\u0131lar\u0131 programlar\u0131n\u0131n ikizler aras\u0131ndaki farklar\u0131 anlayabilece\u011fini \u00f6ne s\u00fcr\u00fcyor ancak bu hen\u00fcz m\u00fcmk\u00fcn de\u011fil.<\/p>\n<p>Bu sistemin g\u00f6z k\u0131rpma, ka\u015f \u00e7atma ve g\u00fcl\u00fcmseme gibi y\u00fczdeki de\u011fi\u015fikliklere duyarl\u0131 oldu\u011fu ve b\u0131y\u0131k ya da sakal uzamas\u0131n\u0131, g\u00f6zl\u00fcklerin \u015feklini alg\u0131layabildi\u011fi s\u00f6yleniyor. FaceIt Argus di\u011fer y\u00fcz tan\u0131ma sistemleriyle birlikte kullan\u0131ld\u0131\u011f\u0131nda do\u011fruluk oran\u0131 y\u00fczde 20 ile 25 oran\u0131nda art\u0131yor. Ancak bu teknoloji d\u00fc\u015f\u00fck \u00e7\u00f6z\u00fcn\u00fcrl\u00fckl\u00fc ve lo\u015f \u0131\u015f\u0131kta \u00e7ekilmi\u015f foto\u011fraflarda ba\u015far\u0131s\u0131z oluyor.<\/p>\n<p>Bunu gizlemek i\u00e7in ba\u015fka bir teknoloji var. Karlsruhe Institute of Technology\u2019nin (Almanya) bilgisayar uzmanlar\u0131 yeni bir teknik geli\u015ftirdi, <a href=\"http:\/\/www.dailymail.co.uk\/sciencetech\/article-3178864\/Nowhere-hide-Facial-recognition-technology-identify-people-complete-DARKNESS-reading-thermal-signatures.html\" target=\"_blank\" rel=\"noopener nofollow\">k\u0131z\u0131l \u00f6tesi portreleri tan\u0131yan bir teknoloji<\/a>. D\u00fc\u015f\u00fck \u0131\u015f\u0131k hatta karanl\u0131kta bile \u00e7al\u0131\u015f\u0131yor. <\/p>\n<p>Bu teknoloji insanlar\u0131n termal yap\u0131s\u0131n\u0131 analiz edip di\u011fer foto\u011fraflar\u0131ndaki ile %80 maksimum do\u011fruluk oran\u0131nda kar\u015f\u0131la\u015ft\u0131r\u0131yor. G\u00f6rselin say\u0131s\u0131ndaki art\u0131\u015f, algoritman\u0131n daha verimli \u00e7al\u0131\u015fmas\u0131n\u0131 sa\u011fl\u0131yor. Sadece bir g\u00f6rsel oldu\u011fundaysa do\u011fruluk oran\u0131 %55\u2019e d\u00fc\u015f\u00fcyor.<\/p>\n<p><span class=\"embed-youtube\" style=\"text-align:center; display: block;\"><iframe class=\"youtube-player\" type=\"text\/html\" width=\"640\" height=\"390\" src=\"https:\/\/www.youtube.com\/embed\/a_W_ZmyYlj0?version=3&amp;rel=1&amp;fs=1&amp;showsearch=0&amp;showinfo=1&amp;iv_load_policy=1&amp;wmode=transparent\" frameborder=\"0\" allowfullscreen=\"true\"><\/iframe><\/span><\/p>\n<p>\u0130lk bak\u0131\u015fta bu matemati\u011fi yapmak kolay gelmemi\u015f olabilir: normal \u0131\u015f\u0131kta \u00e7ekilen ve k\u0131z\u0131l \u00f6tesi \u0131\u015f\u0131kta \u00e7ekilen y\u00fczler aras\u0131nda bir ba\u011flant\u0131 yok. K\u0131z\u0131l \u00f6tesi \u0131\u015f\u0131kta \u00e7ekilen portre ile g\u00fcn \u0131\u015f\u0131\u011f\u0131nda \u00e7ekilen portre birbirinden olduk\u00e7a farkl\u0131 g\u00f6z\u00fck\u00fcyor. <\/p>\n<p>K\u0131z\u0131l \u00f6tesi foto\u011fraflar\u0131n duyarl\u0131l\u0131\u011f\u0131 tene, \u00e7evre \u0131s\u0131s\u0131na ve ki\u015finin ruh haline ba\u011fl\u0131d\u0131r. Ayr\u0131ca k\u0131z\u0131l \u00f6tesi foto\u011fraflar normal foto\u011fraflardan daha d\u00fc\u015f\u00fck kalitelidir. Ki bu da durumu daha da zorla\u015ft\u0131r\u0131r.<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/media.kasperskydaily.com\/wp-content\/uploads\/sites\/91\/2016\/08\/06014026\/2AEFE49A00000578-3178864-Computer_scientists_have_developed_a_technology_that_can_recogni-m-29_1438184452186.jpg\" alt=\"2AEFE49A00000578-3178864-Computer_scientists_have_developed_a_technology_that_can_recogni-m-29_1438184452186\" width=\"634\" height=\"159\" class=\"alignnone size-full wp-image-2307\"><\/p>\n<p>Bu problemi \u00e7\u00f6zmek i\u00e7in bilim adamlar\u0131 makine \u00f6\u011frenmesi algoritmas\u0131na d\u00f6nd\u00fcler ve sistemlerini 82 ki\u015finin 1586 foto\u011fraf\u0131yla \u201cbeslediler\u201d.<\/p>\n<p><strong>Bu her yerde!<\/strong><br>\nBu g\u00fcnlerde, y\u00fcz tan\u0131ma sistemi neredeyse t\u00fcm d\u00fcnyada kullan\u0131l\u0131yor. Uber, taksi s\u00fcr\u00fcc\u00fclerini kontrol etmek i\u00e7in \u00c7in\u2019de buna benzer bir \u00e7\u00f6z\u00fcm <a href=\"https:\/\/www.techinasia.com\/uber-china-facial-recognition\" target=\"_blank\" rel=\"noopener nofollow\">geli\u015ftirdi<\/a>. NEC ve Microsoft y\u00fcz tan\u0131ma i\u015flemini ve IoT\u2019u birle\u015ftirerek m\u00fc\u015fterileri daha iyi tan\u0131ma yoluna <a href=\"http:\/\/www.biometricupdate.com\/201604\/nec-combines-facial-recognition-technologies-with-azure-iot-technology\" target=\"_blank\" rel=\"noopener nofollow\">girdiler<\/a>. Ayn\u0131 zamanda, Rus 2ch.ru forumundan troller porno y\u0131ld\u0131zlar\u0131na sald\u0131rmak i\u00e7in y\u00fcz tan\u0131ma sistemini <a href=\"https:\/\/www.kaspersky.com\/blog\/findface-deanon\/11921\/\" target=\"_blank\" rel=\"noopener nofollow\">kulland\u0131lar<\/a>.<\/p>\n<blockquote class=\"twitter-tweet\" data-width=\"500\" data-dnt=\"true\">\n<p lang=\"en\" dir=\"ltr\"><a href=\"https:\/\/twitter.com\/hashtag\/Trolls?src=hash&amp;ref_src=twsrc%5Etfw\" target=\"_blank\" rel=\"noopener nofollow\">#Trolls<\/a> expose <a href=\"https:\/\/twitter.com\/hashtag\/porn?src=hash&amp;ref_src=twsrc%5Etfw\" target=\"_blank\" rel=\"noopener nofollow\">#porn<\/a> stars social networking accounts <a href=\"https:\/\/t.co\/2mY8kh0JlJ\" target=\"_blank\" rel=\"noopener nofollow\">https:\/\/t.co\/2mY8kh0JlJ<\/a> <a href=\"https:\/\/twitter.com\/hashtag\/socialmedia?src=hash&amp;ref_src=twsrc%5Etfw\" target=\"_blank\" rel=\"noopener nofollow\">#socialmedia<\/a> <a href=\"https:\/\/t.co\/2tejy4TDZ9\" target=\"_blank\" rel=\"noopener nofollow\">pic.twitter.com\/2tejy4TDZ9<\/a><\/p>\n<p>\u2014 Kaspersky (@kaspersky) <a href=\"https:\/\/twitter.com\/kaspersky\/status\/723590897028440064?ref_src=twsrc%5Etfw\" target=\"_blank\" rel=\"noopener nofollow\">April 22, 2016<\/a><\/p><\/blockquote>\n<p><script async src=\"https:\/\/platform.twitter.com\/widgets.js\" charset=\"utf-8\"><\/script><\/p>\n<p>Y\u00fcz tan\u0131ma sisteminin geli\u015ftirilmesiyle gizlilik hakk\u0131nda t\u00fcm bildiklerimizi tekrar g\u00f6zden ge\u00e7irmemiz gerekecek. Bu bug\u00fcn ya da bir y\u0131lda bile olmayacak bir \u015fey ama zaman, buna haz\u0131rlanma zaman\u0131. Her \u015feyi bir kenara b\u0131rak\u0131n, y\u00fcz\u00fcn\u00fcz\u00fc de\u011fi\u015ftiremezsiniz de\u011fil mi? <\/p>\n<p>Gizlili\u011finizi tehlikeye atan teknolojilerden size ne zarar gelece\u011fini merak ediyorsan\u0131z, \u0130ngiliz mini dizisi \u201cBlack mirror\u201d izlemenizi \u00f6neririz, \u00f6zellikle \u201c<a href=\"https:\/\/en.wikipedia.org\/wiki\/Fifteen_Million_Merits_(Black_Mirror)\" target=\"_blank\" rel=\"noopener nofollow\">Fifteen Million Merits<\/a>\u201d b\u00f6l\u00fcm\u00fcn\u00fc.<\/p>\n<p><span class=\"embed-youtube\" style=\"text-align:center; display: block;\"><iframe class=\"youtube-player\" type=\"text\/html\" width=\"640\" height=\"390\" src=\"https:\/\/www.youtube.com\/embed\/zLZHdK6l55I?version=3&amp;rel=1&amp;fs=1&amp;showsearch=0&amp;showinfo=1&amp;iv_load_policy=1&amp;wmode=transparent\" frameborder=\"0\" allowfullscreen=\"true\"><\/iframe><\/span><\/p>\n","protected":false},"excerpt":{"rendered":"<p>2010 y\u0131l\u0131nda d\u00fcnyan\u0131n en b\u00fcy\u00fck y\u00fcz tan\u0131ma sisteminin sahipleri (Facebook) bir foto\u011fraftan portreleri tan\u0131may\u0131 \u00f6\u011frendiler: bu sosyal a\u011f foto\u011fraflardaki y\u00fczleri arad\u0131 ve bu alanlar\u0131 etiketledi. Bazen hatalar yapt\u0131. D\u00f6rt y\u0131l<\/p>\n","protected":false},"author":522,"featured_media":2308,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[1287],"tags":[938,836,939,20,942,943,878,941,940],"class_list":{"0":"post-2302","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-news","8":"tag-algoritma","9":"tag-atm-guvenligi","10":"tag-biometrik","11":"tag-facebook","12":"tag-findface","13":"tag-gelecek","14":"tag-teknoloji","15":"tag-yuz-tanima","16":"tag-yuz-tanimlama"},"hreflang":[{"hreflang":"tr","url":"https:\/\/www.kaspersky.com.tr\/blog\/how-facial-recognition-works\/2302\/"},{"hreflang":"en-us","url":"https:\/\/usa.kaspersky.com\/blog\/how-facial-recognition-works\/7357\/"},{"hreflang":"en-gb","url":"https:\/\/www.kaspersky.co.uk\/blog\/how-facial-recognition-works\/7382\/"},{"hreflang":"es-mx","url":"https:\/\/latam.kaspersky.com\/blog\/how-facial-recognition-works\/7316\/"},{"hreflang":"es","url":"https:\/\/www.kaspersky.es\/blog\/how-facial-recognition-works\/8564\/"},{"hreflang":"it","url":"https:\/\/www.kaspersky.it\/blog\/how-facial-recognition-works\/8512\/"},{"hreflang":"x-default","url":"https:\/\/www.kaspersky.com\/blog\/how-facial-recognition-works\/12073\/"},{"hreflang":"pt-br","url":"https:\/\/www.kaspersky.com.br\/blog\/how-facial-recognition-works\/6402\/"},{"hreflang":"de","url":"https:\/\/www.kaspersky.de\/blog\/how-facial-recognition-works\/8090\/"},{"hreflang":"ja","url":"https:\/\/blog.kaspersky.co.jp\/how-facial-recognition-works\/11385\/"},{"hreflang":"en-au","url":"https:\/\/www.kaspersky.com.au\/blog\/how-facial-recognition-works\/12073\/"},{"hreflang":"en-za","url":"https:\/\/www.kaspersky.co.za\/blog\/how-facial-recognition-works\/12073\/"}],"acf":[],"banners":"","maintag":{"url":"https:\/\/www.kaspersky.com.tr\/blog\/tag\/algoritma\/","name":"algoritma"},"_links":{"self":[{"href":"https:\/\/www.kaspersky.com.tr\/blog\/wp-json\/wp\/v2\/posts\/2302","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.kaspersky.com.tr\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.kaspersky.com.tr\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.kaspersky.com.tr\/blog\/wp-json\/wp\/v2\/users\/522"}],"replies":[{"embeddable":true,"href":"https:\/\/www.kaspersky.com.tr\/blog\/wp-json\/wp\/v2\/comments?post=2302"}],"version-history":[{"count":1,"href":"https:\/\/www.kaspersky.com.tr\/blog\/wp-json\/wp\/v2\/posts\/2302\/revisions"}],"predecessor-version":[{"id":3881,"href":"https:\/\/www.kaspersky.com.tr\/blog\/wp-json\/wp\/v2\/posts\/2302\/revisions\/3881"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.kaspersky.com.tr\/blog\/wp-json\/wp\/v2\/media\/2308"}],"wp:attachment":[{"href":"https:\/\/www.kaspersky.com.tr\/blog\/wp-json\/wp\/v2\/media?parent=2302"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.kaspersky.com.tr\/blog\/wp-json\/wp\/v2\/categories?post=2302"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.kaspersky.com.tr\/blog\/wp-json\/wp\/v2\/tags?post=2302"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}