{"id":52835,"date":"2019-11-17T00:00:00","date_gmt":"2019-11-16T21:00:00","guid":{"rendered":"https:\/\/prohoster.info\/blog\/blog_prohoster\/preimushhestva-oblachnogo-raspoznavaniya-lits"},"modified":"2020-02-18T14:00:38","modified_gmt":"2020-02-18T11:00:38","slug":"preimushhestva-oblachnogo-raspoznavaniya-lits","status":"publish","type":"post","link":"https:\/\/prohoster.info\/sq\/blog\/administrirovanie\/preimushhestva-oblachnogo-raspoznavaniya-lits","title":{"rendered":"P\u00ebrpar\u00ebsit\u00eb e njohjes s\u00eb fytyrave n\u00eb re","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"<p><img decoding=\"async\" alt=\"P\u00ebrpar\u00ebsit\u00eb e njohjes s\u00eb fytyrave n\u00eb re\" src=\"\/wp-content\/uploads\/2019\/11\/cca3a82389e2e5a0c33c065afdc35513.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<i>E ardhmja e af\u00ebrt<\/i><\/p>\n<p>Ekzistojn\u00eb disa metoda q\u00eb funksionojn\u00eb sistemet e njohjes s\u00eb fytyr\u00ebs, por n\u00eb p\u00ebrgjith\u00ebsi b\u00ebhet fjal\u00eb p\u00ebr nj\u00eb teknologji q\u00eb \u00ebsht\u00eb n\u00eb gjendje t\u00eb identifikoj\u00eb nj\u00eb person nga nj\u00eb imazh digjital ose nj\u00eb kuad\u00ebr nga nj\u00eb burim video.<\/p>\n<p>Shum\u00eb pronar\u00eb smartphone-\u00ebsh p\u00ebrdorin \u00e7do dit\u00eb njohjen e fytyr\u00ebs, por n\u00eb pajisje mobile, shpejt\u00ebsia e njohjes nuk \u00ebsht\u00eb kritike, nd\u00ebrs\u00eb numri i p\u00ebrdoruesve rrall\u00eb kalon nj\u00eb ose dy persona. P\u00ebr sistemet zyrtare dhe ato n\u00eb ambientet e hapura (n\u00eb njohjen masive) p\u00ebrdoren teknologji t\u00eb tjera. <\/p>\n<p>S\u00eb fundmi u diskutua n\u00eb Habr\u00eb <noindex><a rel=\"nofollow\" href=\"https:\/\/habr.com\/ru\/news\/t\/474190\/\">lajm<\/a><\/noindex>: kafet\u00eb n\u00eb rrjetin e Mosk\u00ebs \"Pravda Kafe\" dhe OneBucksCoffee kan\u00eb filluar t\u00eb testojn\u00eb sh\u00ebrbimin e njohjes s\u00eb fytyr\u00ebs n\u00eb dyqanet e tyre. <\/p>\n<p>Kafet\u00eb p\u00ebrdorin zgjidhjen ton\u00eb teknike. Dhe sot ne do t\u00eb flasim m\u00eb n\u00eb detaje p\u00ebr t\u00eb. Natyrisht, kemi folur tashm\u00eb p\u00ebr vet\u00eb teknologjin\u00eb, por ka disa novitete - zgjidhja \u00ebsht\u00eb b\u00ebr\u00eb realisht cloud. Dhe kjo gjith\u00e7ka ndryshon.<br \/>\n<noindex><a rel=\"nofollow\" name=\"habracut\"><\/a><\/noindex><\/p>\n<h2>Si funksionon teknologjia e njohjes s\u00eb fytyr\u00ebs<\/h2>\n<p>\nE para q\u00eb duhet t\u00eb b\u00ebj\u00eb sistemi \u00ebsht\u00eb t\u00eb nxjerr\u00eb fytyr\u00ebn n\u00eb kuadrin dhe, duke p\u00ebrdorur algoritme, t\u00eb siguroj\u00eb se \u00ebsht\u00eb pik\u00ebrisht nj\u00eb fytyr\u00eb njeriu. <\/p>\n<p>Pas detektimit fillestar ndodh p\u00ebrcaktimi i karakteristikave t\u00eb ndryshme individuale sipas pikave t\u00eb ngurta - p\u00ebr shembull, merret parasysh distanca mes syve dhe dhjet\u00ebra parametra t\u00eb tjer\u00eb. <\/p>\n<p>M\u00eb pas, algoritme t\u00eb tjera k\u00ebrkojn\u00eb n\u00eb bazat e t\u00eb dh\u00ebnave t\u00eb krijuara m\u00eb par\u00eb dhe japin nj\u00eb p\u00ebrqindje ngjashm\u00ebrie me mostrat e k\u00ebrkuara t\u00eb t\u00eb dh\u00ebnave. N\u00ebse p\u00ebrqindja e ngjashm\u00ebris\u00eb \u00ebsht\u00eb mjaft e lart\u00eb, fytyra konsiderohet e njohur.<\/p>\n<p>N\u00ebse nuk thellohemi n\u00eb detaje (fotot p\u00ebr analiz\u00eb duhet gjithashtu t\u00eb normalizohen, para se t\u2019i kalojm\u00eb n\u00eb rrjetin neurolgjik, i cili lexon nj\u00eb discriptor t\u00eb caktuar), kompleksiteti kryesor i zgjidhjes n\u00eb k\u00ebt\u00eb moment nuk p\u00ebrfshin vet\u00eb teknologjit\u00eb (algoritmet), por realizimin.<\/p>\n<p>Sistemet e njohjes s\u00eb fytyr\u00ebs po zhvillohen n\u00eb disa drejtime, t\u00eb klasifikuara sipas qasjes ndaj p\u00ebrpunimit t\u00eb informacionit. Ndonj\u00ebher\u00eb \u00ebsht\u00eb e v\u00ebshtir\u00eb t\u00eb zgjidhet se cila sistem \u00ebsht\u00eb m\u00eb e p\u00ebrshtatshme p\u00ebr nj\u00eb detyr\u00eb t\u00eb ve\u00e7ant\u00eb. <\/p>\n<h2>Diversity of systems<\/h2>\n<p>\n<img decoding=\"async\" alt=\"P\u00ebrpar\u00ebsit\u00eb e njohjes s\u00eb fytyrave n\u00eb re\" src=\"\/wp-content\/uploads\/2019\/11\/ae40cad194655e1f24ee9f9cb014f3a9.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<br \/>\nT\u00eb dh\u00ebnat mund t\u00eb p\u00ebrpunohen n\u00eb cloud, n\u00eb server\u00ebt lokal\u00eb t\u00eb vendosur n\u00eb perimetrin e siguris\u00eb s\u00eb nd\u00ebrmarrjes, ose direkt n\u00eb kamera. <\/p>\n<p>N\u00eb rastin e fundit, analiza kryhet nga kamera vet\u00eb, dhe n\u00eb server arrin informacioni i p\u00ebrpunuar. Avantazhi kryesor i sistemit \u00ebsht\u00eb sakt\u00ebsia e lart\u00eb dhe mund\u00ebsia e lidhjes s\u00eb nj\u00eb numri t\u00eb madh pajisjesh n\u00eb nj\u00eb server.<\/p>\n<p>N\u00eb pamje t\u00eb par\u00eb, kjo teknologji duket e thjesht\u00eb dhe e leht\u00eb p\u00ebr tu zgjeruar, por ka disa disavantazhe. Nj\u00eb prej tyre \u00ebsht\u00eb \u00e7mimi i lart\u00eb. P\u00ebr m\u00eb tep\u00ebr, aktualisht nuk ka nj\u00eb standard unik p\u00ebr p\u00ebrfaq\u00ebsimin e informacionit q\u00eb kamerat e specializuara e d\u00ebrgojn\u00eb n\u00eb server. Seti i t\u00eb dh\u00ebnave mund t\u00eb ndryshoj\u00eb ndjesh\u00ebm midis ofruesve t\u00eb ndrysh\u00ebm.<\/p>\n<p><img decoding=\"async\" alt=\"P\u00ebrpar\u00ebsit\u00eb e njohjes s\u00eb fytyrave n\u00eb re\" src=\"\/wp-content\/uploads\/2019\/11\/7e05d83ea7bbb2f8e9a8af5aa89e67e9.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<i>Sistemi \"i thjesht\u00eb\" i njohjes s\u00eb fytyr\u00ebs nga <noindex><a rel=\"nofollow\" href=\"https:\/\/www.businesswire.com\/news\/home\/20180226005603\/en\/Panasonic-Launch-Face-Recognition-Server-Software-Deep\/?feedref=JjAwJuNHiystnCoBq_hl-Q-tiwWZwkcswR1UZtV7eGe24xL9TZOyQUMS3J72mJlQ7fxFuNFTHSunhvli30RlBNXya2izy9YOgHlBiZQk2LOzmn6JePCpHPCiYGaEx4DL1Rq8pNwkf3AarimpDzQGuQ%3D%3D&amp;utm_campaign=crowdfire&amp;utm_content=crowdfire&amp;utm_medium=social&amp;utm_source=pinterest\">Panasonic <\/a><\/noindex><\/i><\/p>\n<p>Sistemet q\u00eb bazohen n\u00eb kamerat IP me funksionin e analiz\u00ebs s\u00eb integruar video, jan\u00eb m\u00eb pak t\u00eb njohura se zgjidhjet serverike. Por gjithashtu, duke p\u00ebrdorur nj\u00eb sistem tradicional t\u00eb bazuar n\u00eb regjistrues dhe\/ose nj\u00eb server lokal, nuk do t\u00eb kursesh dot. <\/p>\n<h2>Programet dhe \u00e7mimet* e Njohjes s\u00eb Fytyr\u00ebs <\/h2>\n<p>\n<i>*Sipas informacionit nga burime t\u00eb hapura. <br \/>\n<\/i><br \/>\nDuke marr\u00eb n\u00eb konsiderat\u00eb kompleksitetin e algoritmeve dhe \u00e7mimin e lart\u00eb t\u00eb pajisjeve serverike p\u00ebr modulet e video-analiz\u00ebs, sistemet e njohjes s\u00eb fytyr\u00ebs kan\u00eb mbetur p\u00ebr nj\u00eb koh\u00eb t\u00eb gjat\u00eb nj\u00eb luks i shtrenjt\u00eb. <\/p>\n<p>Shtes\u00eb, \u00e7mimi i zgjidhjes ndikohet nga trafiku i madh t\u00eb dh\u00ebnash q\u00eb gjenerohet gjat\u00eb funksionimit \u2013 p\u00ebrve\u00e7 shpenzimeve p\u00ebr server\u00eb t\u00eb fuqish\u00ebm duhej t\u00eb investohej n\u00eb pajisje t\u00eb avancuara rrjeti dhe kanale komunikimi t\u00eb sh\u00ebndosha.<\/p>\n<p>Aktualisht, n\u00eb tregun rus jan\u00eb disa lojtar\u00eb t\u00eb m\u00ebdhenj q\u00eb ofrojn\u00eb algoritme t\u00eb kualitetit t\u00eb analiz\u00ebs dhe p\u00ebrpunimit t\u00eb t\u00eb dh\u00ebnave video. Ato ndajn\u00eb t\u00eb nj\u00ebjtin interes n\u00eb projekte q\u00eb lidhen me biznesin e madh. Shpjegimi i k\u00ebsaj fokusimi \u00ebsht\u00eb i thjesht\u00eb \u2013 \u00e7mimi i zgjidhjes kalon shum\u00eb p\u00ebrtej mund\u00ebsive t\u00eb bizneseve t\u00eb vogla dhe t\u00eb mesme. <\/p>\n<ul>\n<li>ISS<\/li>\n<\/ul>\n<p>\nSoftware-i \"SecurOS Face\". <\/p>\n<p>\u00c7mimi i licenc\u00ebs p\u00ebr modulin e kapjes s\u00eb fytyr\u00ebs \u00ebsht\u00eb 41 275 rubla p\u00ebr kanal. Software-i instalohet n\u00eb serverin e njohjes s\u00eb fytyr\u00ebs ose n\u00eb serverin e detektimit t\u00eb tyre. <\/p>\n<p>\u00c7mimi i licenc\u00ebs p\u00ebr modul t\u00eb njohjes s\u00eb fytyr\u00ebs p\u00ebr 1000 njer\u00ebz n\u00eb baz\u00eb \u00ebsht\u00eb 665 760 rubla. Instalohet n\u00eb serverin e njohjes s\u00eb fytyr\u00ebs.<\/p>\n<ul>\n<li>Sigur<\/li>\n<\/ul>\n<p>\nProdhues rus i pajisjeve dhe software-it p\u00ebr sistemet e kontrollit t\u00eb aksesit. <\/p>\n<p>\u00c7mimi i licenc\u00ebs p\u00ebr modulin e verifikimit t\u00eb fytyr\u00ebs p\u00ebr nj\u00eb kamer\u00eb \u00ebsht\u00eb 50 000 rubla. <\/p>\n<p>Kostoja e licenc\u00ebs p\u00ebr modulin e identifikimit t\u00eb fytyr\u00ebs p\u00ebr nj\u00eb kamer\u00eb \u00ebsht\u00eb 7,000 rubla. <\/p>\n<p>\u00c7mimi i licenc\u00ebs p\u00ebr databaz\u00ebn deri n\u00eb 1,000 fytyra \u00ebsht\u00eb 294,000 rubla.<\/p>\n<ul>\n<li>ITV<\/li>\n<\/ul>\n<p>\nSoftueri 'Inteligjenca' p\u00ebr njohjen e fytyrave me memorie p\u00ebr 1,000 referenca fytyrash n\u00eb databaz\u00eb \u00ebsht\u00eb 314,000 rubla. <\/p>\n<p>Nuklea i sistemit \u00ebsht\u00eb 20,300 rubla. Lidhja e kanaleve t\u00eb videos \u00ebsht\u00eb 6,000 rubla.<\/p>\n<ul>\n<li>Macroscop<\/li>\n<\/ul>\n<p>\nModuli i njohjes s\u00eb fytyrave Macroscop Basic me nj\u00eb madh\u00ebsi databaze deri n\u00eb 1,000 fytyra \u00ebsht\u00eb 240,000 rubla. <\/p>\n<p>Licenca p\u00ebr t\u00eb punuar me nj\u00eb kamer\u00eb IP \u00ebsht\u00eb 16,500 rubla.<\/p>\n<p>S\u00eb fundmi, zgjidhjet nga Macroscop ishin n\u00eb p\u00ebrdorim p\u00ebr sigurimin e objekteve jasht\u00ebzakonisht t\u00eb r\u00ebnd\u00ebsishme me num\u00ebr t\u00eb madh njer\u00ebzish: stadiume, aeroporte, fabrika. Por tani kompania ofron produktin e saj edhe p\u00ebr shitjen me pakic\u00eb. \u00c7mimi \u00ebsht\u00eb 94,000 rubla p\u00ebr module (regjistruesit nuk shiten).<\/p>\n<ul>\n<li>TRASSIR<\/li>\n<\/ul>\n<p>\nSoftueri kushton 79,000 rubla + 32,000 rubla p\u00ebr regjistruesin. Klient\u00ebt e kompanis\u00eb jan\u00eb kryesisht kompani t\u00eb m\u00ebdha (fabrika, kompani nxjerr\u00ebse, universitete, komplekse sportive). Por kompania fokusohet kryesisht n\u00eb mbik\u00ebqyrjen tradicionale, dhe jo n\u00eb njohjen e fytyrave. Megjithat\u00eb, regjistruesit e tyre t\u00eb videos jan\u00eb t\u00eb p\u00ebrshtatsh\u00ebm edhe p\u00ebr k\u00ebto detyra.<\/p>\n<ul>\n<li>FindFace<\/li>\n<\/ul>\n<p>\nKompania zhvillon dhe shes vet\u00ebm softuer t\u00eb specializuar p\u00ebr njohjen e fytyrave. Ju do t\u00eb duhet t\u00eb zgjidhni konfigurimin e server\u00ebve p\u00ebr ruajtjen dhe p\u00ebrpunimin e t\u00eb dh\u00ebnave vet\u00eb.<\/p>\n<ul>\n<li>Ivideon<\/li>\n<\/ul>\n<p>\nSh\u00ebrbimi i kuq t\u00eb videomonitorimit dhe videoanalitik\u00ebs, i cili ofron sh\u00ebrbime p\u00ebr bizneset me buxhet t\u00eb kufizuar. Sh\u00ebrbimi <noindex><a rel=\"nofollow\" href=\"https:\/\/ru.ivideon.com\/face-recognition\/\">Ivideon Faces<\/a><\/noindex> punon pothuajse me \u00e7do kamer\u00eb, \u00e7mimi p\u00ebr lidhjen e nj\u00eb pajisjeje \u00ebsht\u00eb nga 3,150 rubla me analiz\u00ebn deri n\u00eb 100 fytyra unike n\u00eb dit\u00eb dhe regjistrimin themelor n\u00eb arkivin n\u00eb re p\u00ebr 5 dit\u00eb. <\/p>\n<h2>P\u00ebrzgjedhja e 'harduerit' p\u00ebr sistemet e Njohjes s\u00eb Fytyrave <\/h2>\n<p>\nMe nj\u00eb kamer\u00eb Full HD p\u00ebr p\u00ebrpunimin e videostreamit, q\u00eb p\u00ebrmban 10 fytyra n\u00eb korniz\u00eb, k\u00ebrkohet nj\u00eb nukle i procesorit me frekuenc\u00eb 2.8 GHz. N\u00ebse ka pak fytyra n\u00eb korniz\u00eb (nga 1 deri n\u00eb 3), nj\u00eb nukle procesori mund t\u00eb merret leht\u00ebsisht me p\u00ebrpunimin e dy videostreamve.<\/p>\n<p>Nga ky shembull, duket se edhe n\u00eb nj\u00eb sistem t\u00eb thjesht\u00eb duhet t\u00eb kemi nj\u00eb rezerva t\u00eb caktuar n\u00eb 'harduer'. N\u00ebse n\u00eb objekt hyjn\u00eb nj\u00ebkoh\u00ebsisht 15 njer\u00ebz n\u00eb vend t\u00eb 10, at\u00ebher\u00eb do t\u00eb k\u00ebrkohet nj\u00eb nukle e dyt\u00eb me performanc\u00eb t\u00eb ngjashme. <\/p>\n<p>Prandaj, p\u00ebr pun\u00ebn e sistemit tradicional, duke marr\u00eb parasysh ngarkesat maksimale, \u00ebsht\u00eb e domosdoshme t\u00eb mbani kapacitete dyfish rezerv\u00eb.<\/p>\n<p>P\u00ebr t'ju ndihmuar t\u00eb kuptoni m\u00eb leht\u00eb se sa kushton nj\u00eb sistem tradicional i njohjes s\u00eb fytyr\u00ebs, do t\u00eb marr\u00ebim si shembull nj\u00eb dyqan dhe do t\u00eb llogarisim kostot e nj\u00eb sistemi tradicional dhe nj\u00eb sistemi n\u00eb re p\u00ebr njohjen e fytyr\u00ebs.<\/p>\n<h2>Llogaritja e kostove: \u00e7mimi i sistemit tradicional t\u00eb njohjes s\u00eb fytyr\u00ebs<\/h2>\n<p>\n<img decoding=\"async\" alt=\"P\u00ebrpar\u00ebsit\u00eb e njohjes s\u00eb fytyrave n\u00eb re\" src=\"\/wp-content\/uploads\/2019\/11\/b5726140f7c784705d0bbf96386e4c6d.jpg\" style=\"display:block;margin: 0 auto;\" \/><br \/>\n<br \/>\nSupozoni se ne po implementojm\u00eb nj\u00eb sistem njohjeje t\u00eb fytyr\u00ebs n\u00eb nj\u00eb zinxhir farmacish me 16 pika. N\u00eb mesatare, \u00e7do farmaci vizitohet nga 500 bler\u00ebs \u00e7do dit\u00eb.<\/p>\n<p>P\u00ebr t\u00eb njohur fytyrat n\u00eb m\u00ebnyr\u00eb t\u00eb plot\u00eb, n\u00eb \u00e7do objekt v\u00ebzhgimi mund t\u00eb vendoset nj\u00eb kamer\u00eb rotative ose nj\u00eb kamer\u00eb me objektiv mekanik.<\/p>\n<p>N\u00eb rastin e p\u00ebrdorimit t\u00eb nj\u00eb sistemi tradicional, kostot do t\u00eb jen\u00eb si m\u00eb posht\u00eb:<\/p>\n<ol>\n<li>P\u00ebr \u00e7do farmaci \u00ebsht\u00eb e nevojshme t\u00eb ket\u00eb t\u00eb pakt\u00ebn nj\u00eb regjistrues video t\u00eb specializuar. \u00c7mimi i tij n\u00eb treg \u00ebsht\u00eb rreth 40,000 rubla.<\/li>\n<li>P\u00ebr \u00e7do regjistrues \u00ebsht\u00eb gjithashtu e nevojshme nj\u00eb disk i fort\u00eb special (mos e ngat\u00ebrroni me nj\u00eb HDD t\u00eb zakonsh\u00ebm p\u00ebr PC) me kapacitet jo m\u00eb pak se 4 TB, p\u00ebr t\u00eb regjistruar videon me nj\u00eb rezolucion prej 1920x1080 n\u00eb intensitet t\u00eb lart\u00eb l\u00ebvizjeje. \u00c7mimi mesatar n\u00eb treg \u00ebsht\u00eb 10,000 rubla.<\/li>\n<li>N\u00eb buxhet duhet t\u00eb p\u00ebrfshihet edhe \u00e7mimi p\u00ebr sh\u00ebrbimin e sistemit t\u00eb v\u00ebzhgimit (p.sh. vizita e nj\u00eb teknik p\u00ebr rregullimin e gabimeve, p\u00ebrdit\u00ebsimin e softuerit ose nd\u00ebrrimin e HDD). \u00c7mimi i k\u00ebtyre sh\u00ebrbimeve \u00ebsht\u00eb 12,000 rubla\/vit (vizit\u00eb \u00e7do tremujor) p\u00ebr \u00e7do objekt (sip\u00ebrfaqja e \u00e7mimeve t\u00eb nj\u00eb prej organizatave t\u00eb instalimit).<\/li>\n<li>\u00c7mimi minimal i softuerit funksional p\u00ebr njohjen e fytyr\u00ebs \u00ebsht\u00eb mesatarisht 120,000 rubla p\u00ebr kamer\u00eb (licenc\u00eb pa kufizime n\u00eb koh\u00eb).<\/li>\n<li>Sipas t\u00eb dh\u00ebnave nga kompania Backblaze, rreth 50 % e t\u00eb gjith\u00eb disqeve t\u00eb fort\u00eb k\u00ebrkojn\u00eb z\u00ebvend\u00ebsim pas 6 vjet\u00ebsh p\u00ebrdorimi. Pra, pas 5 vjet\u00ebsh p\u00ebrdorimi t\u00eb pand\u00ebrprer\u00eb, rreth 8 disqe do t\u00eb d\u00ebshtojn\u00eb, dhe n\u00ebse n\u00eb nj\u00eb sistem t\u00eb till\u00eb nuk parashikohet rezervim, mesatarisht duhen planifikuar shpenzime shtes\u00eb p\u00ebr 1.6 disqe n\u00eb vit, ose 16,000 rubla\/vit.<\/li>\n<\/ol>\n<p>\n<b>Shpenzimet kapitale (pa marr\u00eb parasysh \u00e7mimin e kamerave) do t\u00eb p\u00ebrb\u00ebjn\u00eb 2,928,000 rubla\/vit.<\/b><\/p>\n<h2>Shpenzimet p\u00ebr sistemin n\u00eb re<\/h2>\n<p>\nN\u00eb rastin e sistemit n\u00eb re, \u00e7mimi i tarif\u00ebs p\u00ebr v\u00ebzhgimin me njohjen e 500 fytyrave n\u00eb dit\u00eb do t\u00eb jet\u00eb <b>4,750 rubla\/muaj (57,000 rubla\/vit) p\u00ebr kamer\u00eb, ose 912,000 rubla\/vit p\u00ebr 16 kamera<\/b>.<\/p>\n<p>Kujtojm\u00eb se pronari i rrjetit nuk do t\u00eb ket\u00eb nevoj\u00eb t\u00eb blej\u00eb asnj\u00eb \"harduer\" t\u00eb shtes\u00eb. Shpenzimet p\u00ebr mir\u00ebmbajtjen teknike gjithashtu nuk jan\u00eb t\u00eb nevojshme, sepse t\u00eb gjitha server\u00ebt e cloudit mbahen nga ofruesi i sh\u00ebrbimeve cloud n\u00eb qendr\u00ebn e t\u00eb dh\u00ebnave.<\/p>\n<p>Ka nj\u00eb kursim m\u00eb shum\u00eb se 3 her\u00eb gjat\u00eb vitit t\u00eb par\u00eb t\u00eb funksionimit t\u00eb sistemit.<\/p>\n<h2>P\u00ebrmbledhje nd\u00ebrmjet\u00ebse dhe \"pluset\" shtes\u00eb<\/h2>\n<p>\nN\u00eb llogaritjet e m\u00ebsip\u00ebrme ka nj\u00eb nuanc\u00eb t\u00eb r\u00ebnd\u00ebsishme: pas 3 vjet\u00ebsh, sistemit tradicional do t\u2019i dal\u00eb m\u00eb lir\u00eb n\u00eb totalin e shpenzimeve krahasuar me njohjen e fytyrave n\u00eb cloud. Duhet t\u00eb merret parasysh dy faktor\u00eb.<\/p>\n<p><b>S\u00eb pari<\/b>, pajisjet q\u00eb blen pronari i rrjetit do t\u00eb humbasin vler\u00ebn pas 3 vjet\u00ebsh p\u00ebrdorimi. Por me siguri do t\u00eb shfaqen teknologji dhe algoritme t\u00eb reja m\u00eb t\u00eb avancuara p\u00ebr njohjen e fytyrave, q\u00eb do t\u00eb funksionojn\u00eb n\u00eb \"harduer\" m\u00eb t\u00eb fuqish\u00ebm. Dhe pas 3 vjet\u00ebsh, me shum\u00eb gjasa do t\u00eb jet\u00eb e nevojshme t\u00eb z\u00ebvend\u00ebsohet plot\u00ebsisht pajisja n\u00eb pikat e aksesit.<\/p>\n<p>Me sistemin cloud nuk \u00ebsht\u00eb e nevojshme t\u00eb b\u00ebni k\u00ebt\u00eb \u2013 sh\u00ebrbimi p\u00ebrmir\u00ebsohet dhe p\u00ebrdit\u00ebsohet vazhdimisht fal\u00eb zhvillimit t\u00eb algoritmeve dhe rritjes s\u00eb fuqis\u00eb llogaritime t\u00eb qendrave t\u00eb t\u00eb dh\u00ebnave. Mb\u00ebshtetje p\u00ebr standardet e siguris\u00eb gjithashtu nuk \u00ebsht\u00eb e lidhur me harduerin. <\/p>\n<p><b>S\u00eb dyti<\/b>, kursimi i mjeteve n\u00eb vitet e para do t\u00eb lejoj\u00eb q\u00eb k\u00ebto para t\u00eb riciklohen disa her\u00eb, duke sjell\u00eb fitime shtes\u00eb p\u00ebr biznesin.<\/p>\n<h2>E kaluara, e tashmja dhe e ardhmja e njohjes s\u00eb fytyrave n\u00eb cloud<br \/>\n<\/h2>\n<p>\nEvolucioni i sistemeve t\u00eb njohjes \u00ebsht\u00eb p\u00ebrshpejtuar n\u00eb vitet e fundit. P\u00ebrs\u00ebri, jo shum\u00eb koh\u00eb m\u00eb par\u00eb, n\u00eb vend t\u00eb algoritmeve t\u00eb komplikuara dhe rrjeteve neuronale, nj\u00eb punonj\u00ebs i zakonsh\u00ebm i sh\u00ebrbimit t\u00eb siguris\u00eb thjesht krahasonte fytyrat e regjistruara nga programi me bazat e t\u00eb dh\u00ebnave dhe sh\u00ebnonte se kush ishin k\u00ebta njer\u00ebz.<\/p>\n<p>P\u00ebr m\u00eb tep\u00ebr, sistemet funksiononin p\u00ebrmes server\u00ebve lokal\u00eb. Prandaj, p\u00ebr t\u00eb funksionuar sh\u00ebrbimi, p\u00ebrdoruesi duhej t\u00eb instalohej nj\u00eb PC t\u00eb dedikuar ose nj\u00eb regjistrues t\u00eb specializuar t\u00eb videos. Kjo n\u00ebnkupton shpenzime shtes\u00eb p\u00ebr pajisjet dhe kostot indirekte p\u00ebr veprimin e tyre.<\/p>\n<p>Njohja e fytyrave n\u00eb cloud nuk k\u00ebrkon blerjen dhe konfigurimin e ndonj\u00eb pajisjeje tjet\u00ebr, p\u00ebrve\u00e7 kamerave, dhe do t\u00eb funksionoj\u00eb me ato kamera q\u00eb tashm\u00eb jan\u00eb instaluar n\u00eb objekt.<\/p>\n<p>Nuk \u00ebsht\u00eb e nevojshme t\u00eb mbani nj\u00eb staf specialist\u00ebsh p\u00ebr t\u00eb mbajtur pun\u00ebn e pajisjeve. Problemet me gjendjen teknike t\u00eb pajisjeve zgjidhen nga vet\u00eb ofruesi i sh\u00ebrbimit (dhe e b\u00ebn k\u00ebt\u00eb m\u00eb efikas se kompanit\u00eb e pa specializuara). <\/p>\n<p>Njohja n\u00eb re e kthen nj\u00eb sistem t\u00eb rend\u00eb dhe t\u00eb ndjesh\u00ebm nga serverat analitik\u00eb lokal\u00eb n\u00eb nj\u00eb struktur\u00eb fleksib\u00ebl dhe t\u00eb q\u00ebndrueshme n\u00eb re. Kjo do t\u00eb thot\u00eb q\u00eb sistemi i njohjes nuk varet m\u00eb nga kapacitetet e nj\u00eb serveri t\u00eb ve\u00e7ant\u00eb, i bler\u00eb dhe instaluar n\u00eb zyr\u00ebn e klientit, as nga infrastruktura IT q\u00eb ky klient disponon. Nuk ka nevoj\u00eb t\u00eb blini pajisje t\u00eb reja dhe t\u00eb zgjidhni p\u00ebr nj\u00eb koh\u00eb t\u00eb gjat\u00eb me ofruesin n\u00eb lidhje me konfigurimin dhe mund\u00ebsit\u00eb e zgjerimit.<\/p>\n<p>Reja automatikisht shp\u00ebrndan ngarkes\u00ebn n\u00eb t\u00eb gjith\u00eb infrastruktur\u00ebn e disponueshme me servera t\u00eb fuqish\u00ebm. Klienti nuk ka nevoj\u00eb t\u00eb mbaj\u00eb burime q\u00eb p\u00ebrdoren rrall\u00eb n\u00eb rezerv\u00eb p\u00ebr t\u00eb punuar gjat\u00eb periudhave t\u00eb papritura me rritje t\u00eb ngarkes\u00ebs (festat, fundjavave). M\u00eb shum\u00eb informacion mbi mund\u00ebsit\u00eb e sistemit mund t\u00eb merrni, <noindex><a rel=\"nofollow\" href=\"mailto:videoanalytics@ivideon.com\">duke u konsultuar<\/a><\/noindex> me ne. <\/p>\n<p>\"E v\u00ebrteta mbi kafen\" dhe OneBucksCoffee tani kan\u00eb shkaktuar nj\u00eb stuhi diskutimesh, por shum\u00eb shpejt nuk do t\u00eb mbeten kompani n\u00eb biznesin fizik pa videoanalitik\u00eb. Aktor\u00ebt e tregut t\u00eb konsumator\u00ebve kan\u00eb nj\u00eb nevoj\u00eb t\u00eb madhe p\u00ebr t\u00eb njohur bler\u00ebsin n\u00eb fytyr\u00eb: p\u00ebr t\u00eb personalizuar sh\u00ebrbimin dhe ofertat, p\u00ebr t\u00eb analizuar disponimin e mysafir\u00ebve, p\u00ebr t\u00eb reduktuar shpenzimet dhe p\u00ebr t\u00eb rikthyer klient\u00ebt, e jo thjesht p\u00ebr t\u00eb bler\u00eb zgjidhje teknologjike p\u00ebr t\u00eb arritur raporte.<br \/>\n<br \/>Burimi: <a content=\"nofollow\" rel=\"nofollow\" href=\"https:\/\/habr.com\/ru\/company\/ivideon\/blog\/475888\/\">habr.com<\/a><\/p>","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"excerpt":{"rendered":"<p>\u0411\u043b\u0438\u0437\u043a\u043e\u0435 \u0431\u0443\u0434\u0443\u0449\u0435\u0435 \u0421\u0443\u0449\u0435\u0441\u0442\u0432\u0443\u0435\u0442 \u043d\u0435\u0441\u043a\u043e\u043b\u044c\u043a\u043e \u043c\u0435\u0442\u043e\u0434\u043e\u0432, \u043f\u043e \u043a\u043e\u0442\u043e\u0440\u044b\u043c \u0440\u0430\u0431\u043e\u0442\u0430\u044e\u0442 \u0441\u0438\u0441\u0442\u0435\u043c\u044b \u0440\u0430\u0441\u043f\u043e\u0437\u043d\u0430\u0432\u0430\u043d\u0438\u044f \u043b\u0438\u0446, \u043d\u043e \u0432 \u0446\u0435\u043b\u043e\u043c \u0440\u0435\u0447\u044c \u0438\u0434\u0451\u0442 \u043e \u0442\u0435\u0445\u043d\u043e\u043b\u043e\u0433\u0438\u0438, \u0441\u043f\u043e\u0441\u043e\u0431\u043d\u043e\u0439 \u0438\u0434\u0435\u043d\u0442\u0438\u0444\u0438\u0446\u0438\u0440\u043e\u0432\u0430\u0442\u044c \u0447\u0435\u043b\u043e\u0432\u0435\u043a\u0430 \u043f\u043e \u0446\u0438\u0444\u0440\u043e\u0432\u043e\u043c\u0443 \u0438\u0437\u043e\u0431\u0440\u0430\u0436\u0435\u043d\u0438\u044e \u0438\u043b\u0438 \u043a\u0430\u0434\u0440\u0443 \u0438\u0437 \u0432\u0438\u0434\u0435\u043e\u0438\u0441\u0442\u043e\u0447\u043d\u0438\u043a\u0430. \u041c\u043d\u043e\u0433\u0438\u0435 \u0432\u043b\u0430\u0434\u0435\u043b\u044c\u0446\u044b \u0441\u043c\u0430\u0440\u0442\u0444\u043e\u043d\u043e\u0432 \u043a\u0430\u0436\u0434\u044b\u0439 \u0434\u0435\u043d\u044c \u0438\u0441\u043f\u043e\u043b\u044c\u0437\u0443\u044e\u0442 face recognition, \u043d\u043e \u0432 \u043c\u043e\u0431\u0438\u043b\u044c\u043d\u044b\u0445 \u0443\u0441\u0442\u0440\u043e\u0439\u0441\u0442\u0432\u0430\u0445 \u0441\u043a\u043e\u0440\u043e\u0441\u0442\u044c \u0440\u0430\u0441\u043f\u043e\u0437\u043d\u0430\u0432\u0430\u043d\u0438\u044f \u043d\u0435 \u043a\u0440\u0438\u0442\u0438\u0447\u043d\u0430, \u0430 \u0447\u0438\u0441\u043b\u043e \u043f\u043e\u043b\u044c\u0437\u043e\u0432\u0430\u0442\u0435\u043b\u0435\u0439 \u0440\u0435\u0434\u043a\u043e \u0431\u043e\u043b\u044c\u0448\u0435 \u043e\u0434\u043d\u043e\u0433\u043e-\u0434\u0432\u0443\u0445 \u0447\u0435\u043b\u043e\u0432\u0435\u043a. \u0414\u043b\u044f \u043e\u0444\u0438\u0441\u043d\u044b\u0445 \u0438 \u0443\u043b\u0438\u0447\u043d\u044b\u0445 [&hellip;]<\/p>\n","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[688],"tags":[],"class_list":["post-52835","post","type-post","status-publish","format-standard","hentry","category-administrirovanie"],"aioseo_notices":[],"aioseo_head":"\n\t\t<!-- All in One SEO 5.0.2 - aioseo.com -->\n\t<meta name=\"description\" content=\"\u0411\u043b\u0438\u0437\u043a\u043e\u0435 \u0431\u0443\u0434\u0443\u0449\u0435\u0435 \u0421\u0443\u0449\u0435\u0441\u0442\u0432\u0443\u0435\u0442 \u043d\u0435\u0441\u043a\u043e\u043b\u044c\u043a\u043e.\" \/>\n\t<meta name=\"robots\" content=\"max-image-preview:large\" \/>\n\t<meta name=\"author\" content=\"Yuri Gagarin\"\/>\n\t<link rel=\"canonical\" href=\"https:\/\/prohoster.info\/sq\/blog\/administrirovanie\/preimushhestva-oblachnogo-raspoznavaniya-lits\" \/>\n\t<meta name=\"generator\" content=\"All in One SEO (AIOSEO) 5.0.2\" \/>\n\t\t<meta property=\"og:locale\" content=\"sq_AL\" \/>\n\t\t<meta property=\"og:site_name\" content=\"ProHoster | \u041a\u0443\u043f\u0438\u0442\u044c \u043d\u0430\u0434\u0435\u0436\u043d\u044b\u0439 \u0445\u043e\u0441\u0442\u0438\u043d\u0433 \u0434\u043b\u044f \u0441\u0430\u0439\u0442\u043e\u0432 \u0441 \u0437\u0430\u0449\u0438\u0442\u043e\u0439 \u043e\u0442 DDoS, VPS VDS \u0441\u0435\u0440\u0432\u0435\u0440\u044b\" \/>\n\t\t<meta property=\"og:type\" content=\"article\" \/>\n\t\t<meta property=\"og:title\" content=\"\ud83e\udd47\u041f\u0440\u0435\u0438\u043c\u0443\u0449\u0435\u0441\u0442\u0432\u0430 \u043e\u0431\u043b\u0430\u0447\u043d\u043e\u0433\u043e \u0440\u0430\u0441\u043f\u043e\u0437\u043d\u0430\u0432\u0430\u043d\u0438\u044f \u043b\u0438\u0446 | ProHoster\" \/>\n\t\t<meta property=\"og:description\" content=\"\u0411\u043b\u0438\u0437\u043a\u043e\u0435 \u0431\u0443\u0434\u0443\u0449\u0435\u0435 \u0421\u0443\u0449\u0435\u0441\u0442\u0432\u0443\u0435\u0442 \u043d\u0435\u0441\u043a\u043e\u043b\u044c\u043a\u043e.\" \/>\n\t\t<meta property=\"og:url\" content=\"https:\/\/prohoster.info\/sq\/blog\/administrirovanie\/preimushhestva-oblachnogo-raspoznavaniya-lits\" \/>\n\t\t<meta property=\"og:image\" content=\"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg\" \/>\n\t\t<meta property=\"og:image:secure_url\" content=\"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg\" \/>\n\t\t<meta property=\"og:image:width\" content=\"350\" \/>\n\t\t<meta property=\"og:image:height\" content=\"350\" \/>\n\t\t<meta property=\"article:published_time\" content=\"2019-11-16T21:00:00+00:00\" \/>\n\t\t<meta property=\"article:modified_time\" content=\"2020-02-18T11:00:38+00:00\" \/>\n\t\t<meta property=\"article:publisher\" content=\"https:\/\/www.facebook.com\/prohoster\" \/>\n\t\t<meta property=\"article:author\" content=\"https:\/\/www.facebook.com\/prohoster\" \/>\n\t\t<!-- All in One SEO -->\n\n","aioseo_head_json":{"title":"\ud83e\udd47P\u00ebrfitimet e njohjes n\u00eb re t\u00eb fytyr\u00ebs | ProHoster","description":"E ardhmja e af\u00ebrt ka disa.","canonical_url":"https:\/\/prohoster.info\/sq\/blog\/administrirovanie\/preimushhestva-oblachnogo-raspoznavaniya-lits","robots":"max-image-preview:large","keywords":"","webmasterTools":{"miscellaneous":""},"schema":null,"og:locale":"sq_AL","og:site_name":"ProHoster | \u041a\u0443\u043f\u0438\u0442\u044c \u043d\u0430\u0434\u0435\u0436\u043d\u044b\u0439 \u0445\u043e\u0441\u0442\u0438\u043d\u0433 \u0434\u043b\u044f \u0441\u0430\u0439\u0442\u043e\u0432 \u0441 \u0437\u0430\u0449\u0438\u0442\u043e\u0439 \u043e\u0442 DDoS, VPS VDS \u0441\u0435\u0440\u0432\u0435\u0440\u044b","og:type":"article","og:title":"\ud83e\udd47\u041f\u0440\u0435\u0438\u043c\u0443\u0449\u0435\u0441\u0442\u0432\u0430 \u043e\u0431\u043b\u0430\u0447\u043d\u043e\u0433\u043e \u0440\u0430\u0441\u043f\u043e\u0437\u043d\u0430\u0432\u0430\u043d\u0438\u044f \u043b\u0438\u0446 | ProHoster","og:description":"\u0411\u043b\u0438\u0437\u043a\u043e\u0435 \u0431\u0443\u0434\u0443\u0449\u0435\u0435 \u0421\u0443\u0449\u0435\u0441\u0442\u0432\u0443\u0435\u0442 \u043d\u0435\u0441\u043a\u043e\u043b\u044c\u043a\u043e.","og:url":"https:\/\/prohoster.info\/sq\/blog\/administrirovanie\/preimushhestva-oblachnogo-raspoznavaniya-lits","og:image":"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg","og:image:secure_url":"https:\/\/prohoster.info\/wp-content\/uploads\/2021\/11\/logo-350.jpg","og:image:width":350,"og:image:height":350,"article:published_time":"2019-11-16T21:00:00+00:00","article:modified_time":"2020-02-18T11:00:38+00:00","article:publisher":"https:\/\/www.facebook.com\/prohoster","article:author":"https:\/\/www.facebook.com\/prohoster"},"aioseo_meta_data":{"post_id":"52835","title":null,"description":null,"keywords":null,"keyphrases":null,"primary_term":null,"canonical_url":null,"og_title":null,"og_description":null,"og_object_type":"default","og_image_type":"default","og_image_url":null,"og_image_width":null,"og_image_height":null,"og_image_custom_url":null,"og_image_custom_fields":null,"og_video":null,"og_custom_url":null,"og_article_section":null,"og_article_tags":null,"twitter_use_og":false,"twitter_card":"default","twitter_image_type":"default","twitter_image_url":null,"twitter_image_custom_url":null,"twitter_image_custom_fields":null,"twitter_title":null,"twitter_description":null,"schema":{"blockGraphs":[],"customGraphs":[],"default":{"data":{"Article":[],"Course":[],"Dataset":[],"FAQPage":[],"Movie":[],"Person":[],"Product":[],"ProductReview":[],"Car":[],"Recipe":[],"Service":[],"SoftwareApplication":[],"WebPage":[]},"graphName":"","isEnabled":true},"graphs":[]},"schema_type":null,"schema_type_options":null,"pillar_content":false,"robots_default":true,"robots_noindex":false,"robots_noarchive":false,"robots_nosnippet":false,"robots_nofollow":false,"robots_noimageindex":false,"robots_noodp":false,"robots_notranslate":false,"robots_max_snippet":null,"robots_max_videopreview":null,"robots_max_imagepreview":"large","priority":null,"frequency":null,"local_seo":null,"seo_analyzer_scan_date":"2026-01-24 05:03:20","breadcrumb_settings":null,"limit_modified_date":false,"reviewed_by":null,"ai":null,"created":"2021-02-28 20:36:36","updated":"2026-01-24 05:03:20","focus_keyword":null,"additional_keywords":null,"truseo_locale":null},"gt_translate_keys":[{"key":"link","format":"url"}],"_links":{"self":[{"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/posts\/52835","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/comments?post=52835"}],"version-history":[{"count":0,"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/posts\/52835\/revisions"}],"wp:attachment":[{"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/media?parent=52835"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/categories?post=52835"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/prohoster.info\/sq\/wp-json\/wp\/v2\/tags?post=52835"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}