CNN na tushen yanki
CNNs-Based CNNs (R-CNNs) dangi ne na abubuwan gano abubuwa waɗanda suka fara ba da shawarar yankunan ɗan takara a cikin hoto, sannan yi amfani da CNN don rarrabewa da daidaita kowane abu daidai.
Dubawa
They turned image classification into full object detection, locating and labeling many objects at once.
Zurfafa nutsewa
Rarraba hoto ya amsa 'menene a wannan hoton?' amma ganewa dole ne kuma ya amsa 'a ina, kuma nawa?' R-CNN na asali (2014) ya yi amfani da algorithm na waje (Binciken Zaɓi) don ba da shawara a kusa da yankuna 2,000, ya karkatar da kowannensu zuwa ƙayyadaddun girman, kuma ya gudanar da CNN akan kowane ɗayan, wanda yake daidai amma jinkirin jinkiri. Mai sauri R-CNN ya haɓaka wannan ta hanyar gudanar da CNN sau ɗaya akan dukkan hoto da abubuwan haɗawa kowane yanki (RoI pooling). Saurin R-CNN sannan ya maye gurbin Neman Zaɓan tare da ƙwararren hanyar sadarwa na yanki (RPN), yana mai da duka bututun daga ƙarshe zuwa ƙarshe kuma kusa da ainihin lokaci. Mask R-CNN ya ƙara shi don fitar da abin rufe fuska-pixel ga kowane abu da aka gano.
Fahimtar Fasaha
Babban maɓalli mai inganci shine haɗawar RoI: maimakon sake kunna CNN akan kowane akwatin da aka tsara, hanyar sadarwar tana ƙididdige taswirar fasali guda ɗaya don hoton, sannan amfanin gona kuma ta canza fasalin fasalin cikin kowane yanki na sha'awa zuwa tsayayyen grid. RPN na R-CNN mai sauri yana zamewa akan wannan taswirar fasalin yana tsinkayar maki 'abu' da gyare-gyaren akwatin don akwatunan anka da aka saita masu girma dabam da ma'auni, samar da shawarwari kusan kyauta.
Dabarun Tasiri
Gudu da sikelin
Kayayyakin AI na iya sarrafa aiki da bincike, ganowa, da ayyuka masu alama a sikelin.
Gina zaɓuɓɓuka
Ƙungiyoyin ƙirƙira za su iya samar da ra'ayoyi cikin sauri tare da ƙarancin bita da hannu.
Ƙungiya da aikin aiki
Ayyuka na iya amfani da siginar hoto da bidiyo waɗanda a baya suke da wahalar aiwatarwa.
Makomar CNNs na tushen yanki
Masu gano matakan R-CNN guda biyu sun kasance masu ƙarfi a inda daidaito ya fi dacewa, amma masu gano matakai guda ɗaya (YOLO, SSD) da masu gano na'urorin Transformer kamar DETR, waɗanda ke tsallake ginshiƙan ƙirar hannu da shawarwari gabaɗaya, suna ƙara shahara ga sauri da sauƙi. Yanayin yana zuwa ƙarshen-zuwa-ƙarshe, ba tare da anga ba, gano tushen tambaya. Har yanzu, ainihin ra'ayoyin layin R-CNN, abubuwan da aka raba da kuma tunanin matakin yanki, suna ci gaba da tasiri ga tsarin yanki, bidiyo, da tsarin gano 3D.
Aiwatar da Gaskiyar Duniya
Ganowa da ƙididdige samfura a kan ɗakunan ajiya don sarrafa kaya
Misali na sel ko gabobin jiki a cikin binciken likita ta amfani da Mask R-CNN
Gano lahani da wuraren su akan layin samar da masana'anta
Gano motoci masu yawa da masu tafiya a ƙasa a cikin ciyarwar kamara mai sarrafa kanta
Hatsari & Tsare-tsare
Haƙƙoƙin hoto da yarda na iya zama haxarin doka idan ba a fayyace ba.
Ayyukan samfuri na iya bambanta a ko'ina cikin haske, ƙididdiga, da mahalli.
Ƙarya tabbataccen ƙila ba za a iya lura da shi ba sai dai idan an kula da ƙofofin amincewa.
Taswirar Hanya
Ƙayyade ma'auni na karɓa don daidaito, tunowa, da farashi na kuskure.
Gwada tare da bayanan da suka dace da ainihin yanayin samarwa.
Ƙara bita na ɗan adam don ƙarancin amincewa ko tsinkaya mai tasiri.
Bi diddigin ƙirar ƙira kuma sake ingantawa bayan canje-canjen kamara ko saitin bayanai.
Ci gaba da Bincike
Free newsletter
Get the daily AI briefing
Three verified AI stories every weekday morning, written in plain English. Free forever, no ads.
One email each weekday. Unsubscribe in one click. We never sell or share your address.
Test yourself
Take the Region-Based CNNs quiz
Instant feedback on every answer, and a shareable certificate with a verifiable ID once you pass a course.
Support free AI education. AI Understanding is a 501(c)(3) nonprofit — no ads, no paywall, ever. Make a donation
Jagora na gaba
Samfuran Ƙarfafa Maki
Tambayoyin da ake yawan yi
What is Region-Based CNNs?
CNNs-Based CNNs (R-CNNs) dangi ne na abubuwan gano abubuwa waɗanda suka fara ba da shawarar yankunan ɗan takara a cikin hoto, sannan yi amfani da CNN don rarrabewa da daidaita kowane abu daidai. Sun mai da rarrabuwar hoto zuwa cikakken gano abu, ganowa da kuma yiwa abubuwa da yawa lakabi lokaci guda.
Menene ainihin ra'ayin CNN na yanki?
R-CNNs suna ba da shawarar yankuna waɗanda za su iya ƙunsar abubuwa, sannan su gudanar da CNN don rarraba kowane yanki da kuma daidaita akwatin da ke kan iyaka.
Me yasa ainihin R-CNN (2014) ya kasance a hankali?
Asalin R-CNN ya gudanar da CNN a kansa akan kusan shawarwarin yanki 2,000 akan kowane hoto, wanda yayi tsada sosai.
Menene Saurin R-CNN ya gabatar don maye gurbin Neman Zaɓaɓɓen?
Mai Saurin R-CNN ya gabatar da hanyar sadarwa ta Shawarar Yanki da aka koyo, yana mai da tsarin samar da wani yanki na hanyar sadarwa mai horarwa kuma cikin sauri.
Menene haɗin gwiwar RoI ke cim ma?
RoI pooling yana fitar da ƙayyadaddun grid fasali daga taswirar fasali guda ɗaya a kowane yanki, guje wa ƙididdigewa da saurin ganowa.
Wane ƙarin iyawa Mask R-CNN ya ƙara?
Mask R-CNN yana ƙara sauri R-CNN ta hanyar tsinkayar madaidaicin abin rufe fuska-pixel ga kowane abu, yana ba da damar rarrabuwar misali.