Na'urori masu aunawa da lokaci na haske (ToF) na kamara suna ba da hanya mai sauri da dacewa don samun bayanan muhalli na 3D ta hanyar auna lokacin tafiya na hasken da aka fitar da shi. Wannan takarda ta gabatar da cikakkiyar hanyar simulashi don kimanta aikin na'urar, fahimtar kurakuran gwaji, da bincika tasirin gani cikin zurfi. Simulashin yana da mahimmanci don gano iyakokin na'urar, inganta ƙarfin aunawa, da haɓaka iyawar gane tsari a aikace-aikacen duniya inda hayaniya da rikitattun abubuwan gani suka yawaita.
Na'urorin ToF suna lissafin nisa a kowane pixel ta hanyar auna lokacin da haske ke ɗauka daga tushe zuwa wani abu sannan ya koma na'urar gano shi.
D-ToF yana auna lokacin tafiya na gajerun bugun haske kai tsaye. Don nisa har zuwa mita 50, wannan yana buƙatar gajerun bugun haske da lokutan fallasa (misali, 10 ns don 1.5 m), yana aiki a cikin kewayon GHz. Wannan sau da yawa yana haifar da ƙarancin rabo na sigina zuwa hayaniya (SNR), kamar yadda aka lura a cikin wallafe-wallafen da suka danganci (Jarabo et al., 2017).
Wanda kuma aka sani da ToF na tushen lokaci (P-ToF), wannan hanyar kai tsaye tana daidaita tushen haske kuma tana haɗa sigina da aka karɓa. Yawancin kyamarorin ToF na zamani suna amfani da ka'idar Amplitude Modulated Continuous Wave (AMCW) ko Continuous Wave Intensity Modulation (CWIM). Ana auna canjin lokaci tsakanin sigina da aka fitar da waɗanda aka karɓa, yawanci ta amfani da na'urar haɗa photon (PMD) a kowane pixel tare da Lock-In demodulation (Schwarte et al., 1997; Lange, 2000). Hoto na 1 yana kwatanta abubuwan tsarin.
Hoto na 1: Ka'idar aunawa na na'urar ToF na tushen kamara ta amfani da AMCW (an daidaita daga Druml et al., 2015). Zanen yana nuna na'urar hoto na 3D, tushen haske mai daidaitawa (LED/VCSEL), ruwan tabarau, matrix na pixel, na'urar canzawa A/D, mai sarrafa jerin gwano, mai sarrafa mai masaukin baki, da lissafin taswirar zurfi da ya biyo baya.
Babban gudummawar shine hanyar simulashin da ke ba da damar bincika tasirin gani cikin zurfi.
Simulashin yana amfani da tushen bin diddigin haske a cikin samfurin gani na geometric. Wannan yana ba da damar bin diddigin hasken kowane haske daga tushe(s) ta cikin wurin, yana lissafin hulɗa da abubuwa da yawa da ruwan tabarau na kamara kafin ya isa na'urar gano shi.
Lissafin zurfi ya dogara ne akan tsawon hanyar gani (OPL), wanda aka ayyana shi azaman samfurin tsawon hanyar geometric da ma'anar refractive na matsakaici: $OPL = \int n(s) \, ds$. Wannan shine babban ma'auni don zurfi, yana ba da damar simulashin nau'ikan na'urorin ToF daban-daban (D-ToF, C-ToF) da tallafawa kimantawar hoto na wucin gadi.
An aiwatar da hanyar ta amfani da Zemax OpticStudio don bin diddigin hasken gani mai inganci da samfurin ruwan tabarau, haɗe tare da Python don samar da wuri, sarrafa bayanai, bincike, da aiwatar da samfuran na'urar (misali, demodulation, hayaniya).
An tsara tsarin don lissafin rikitattun abubuwan gani na duniya waɗanda ke ƙalubalantar na'urorin ToF.
Yana kwaikwayon tsangwama na hanyoyi da yawa (MPI), inda haske ke tunani daga saman da yawa kafin ya isa na'urar, babban tushen kuskuren zurfi. Mai bin diddigin hasken yana bin waɗannan hanyoyi masu rikitarwa.
Yana samfurin jigilar haske ta cikin kayan da ba su da ƙarfi (misali, gilashi, filastik), inda watsawa a ƙarƙashin saman da tunani na ciki ke faruwa, suna shafar lokaci da girman da aka auna.
Ya haɗa da tasirin ruwan tabarau kamar kuskuren spherical, kuskuren chromatic, da karkatarwa. Waɗannan kurakuran suna canza hanyar gani da gaban igiyar ruwa, suna tasiri daidaiton aunawar lokaci/zurfi a kowane pixel.
Takardar ta nuna manyan fasalulluka akan wurin gwaji na 3D mai sauƙi. Duk da yake ba a yi cikakken bayani game da takamaiman sakamako na ƙididdiga ba a cikin abin da aka ba da shi, nunin yana iya nuna ikon simulashin na:
Abubuwan da aka fitar na simulashin za su haɗa da taswirorin haske, taswirorin lokaci, da taswirorin zurfi na ƙarshe, tare da ma'aunin kuskure da ke kwatanta sakamakon simulashin da gaskiya.
Ingancin simulashin ya dogara ne akan ingantaccen samfurin zahiri. Manyan ma'auni sun haɗa da:
Tsawon Hanyar Gani (OPL): $OPL = \sum_{i} n_i \cdot d_i$, inda $n_i$ shine ma'anar refractive kuma $d_i$ shine nisan geometric a cikin sashi $i$.
Canjin Lokaci don C-ToF: Canjin lokaci da aka auna $\phi$ yana da alaƙa da OPL da mitar daidaitawa $f_{mod}$: $\phi = 2 \pi \cdot 2 \cdot \frac{OPL}{c} \cdot f_{mod} = 4 \pi f_{mod} \frac{OPL}{c}$, inda $c$ shine saurin haske. Ma'auni na 2 yana lissafin tafiya biyu. Zurfi $z$ shine: $z = \frac{c \cdot \phi}{4 \pi f_{mod}}$.
Samfurin Sigina: Sigina mai haɗin kai $S$ a wani pixel don PMD mai famfo da yawa ana iya samfurin shi kamar: $S_k = \alpha \int_{0}^{T} I_{emit}(t) \cdot I_{demod,k}(t - \tau) \, dt + \eta$, inda $\alpha$ shine albedo/reflectance, $I_{emit}$ shine ƙarfin da aka fitar, $I_{demod,k}$ shine aikin demodulation don famfo $k$, $\tau$ shine jinkirin lokaci mai daidaitawa da OPL, $T$ shine lokacin haɗawa, kuma $\eta$ shine hayaniya.
Wannan aikin ba kawai wani kayan aikin simulashin ba ne; gada ce ta dabara tsakanin ƙirar gani mai kyau da rikitattun gaskiyar fahimtar ToF. Ta hanyar goyon bayan Tsawon Hanyar Gani (OPL) a matsayin babban ma'auni mai haɗawa, marubutan sun wuce nisa mai sauƙi na geometric. Wannan canji ne mai zurfi. Yana magance kai tsaye ƙafar Achilles na kasuwanci na ToF: kurakuran tsarin daga tsangwama na hanyoyi da yawa (MPI) da kaddarorin kayan, waɗanda su ne abubuwan da suka dogara da OPL. Hanyarsu tana ɗaukar jigilar haske a matsayin ɗan ƙasa na farko, yana ba da damar warware dalilin da yasa taswirorin zurfi suka gaza a kusurwoyi, kusa da gilashi, ko ƙarƙashin hasken yanayi—matakin bincike da ya rasa daga yawancin takaddun bayanan mai siyarwa.
Hankali yana da kyau na masana'antu: Ayyana gaskiyar ƙasa (OPL ta hanyar bin diddigin haske) → Kwaikwayi aunawar da ba ta cika ba na na'urar (ƙara daidaitawa/ragewa, hayaniya) → Bincika bambancin. Wannan kwararar tana kama da mafi kyawun ayyuka a cikin halayyar na'urar amma tana amfani da shi a hankali a cikin simulashin. Amfani da Zemax don gani da Python don dabaru na na'urar yana haifar da bututu mai sassauƙa, mai sassa. Duk da haka, sarkar hankali tana da rauni mai rauni: takardar tana nuna ƙarfi amma ba ta yi cikakken bayani game da fassarar daga taswirar OPL mai kyau na simulashin zuwa ƙimar pixel na ƙarshe, mai hayaniya, da aka rage ba. Tsalle daga gani na zahiri zuwa na'urorin lantarki na na'urar shine muhimmin mu'amala inda yawancin kurakura ke haifuwa, kuma zurfin samfurinsa ya kasance ba a bayyana ba.
Ƙarfi: Cikakkiyar hanyar aikin shine fasalin kashe shi. Simulashin MPI, wucewa da haske, da kurakuran ruwan tabarau a cikin tsari ɗaya ba kasafai ba ne. Wannan cikakkiyar ra'ayi yana da mahimmanci, saboda waɗannan tasirin suna hulɗa ba tare da layi ba. Aiwatar da aiki ta amfani da Zemax mai daidaitaccen masana'antu yana ba da sahihanci nan da nan da canzawa ga ƙungiyoyin R&D. Idan aka kwatanta da masu fitarwa na ilimi kamar Mitsuba ko Blender Cycles, waɗanda suka mai da hankali kan ingancin gani, wannan bututu an gina shi da manufa don aunawa.
Kurakurai & Makafin Gani: Giwa a cikin ɗaki shine farashin lissafi. Cikakken bin diddigin haske na geometric don rikitattun wurare, watsawa da hanyoyi da yawa sanannen tsada ne. Takardar ba ta yi shiru game da dabarun hanzari ba (misali, bin diddigin hanyoyi biyu, taswirar photon) ko aikin da za a iya samu, wanda ke iyakance amfanin da ake ganinsa don ƙira mai maimaitawa. Na biyu, da alama tana gefen igiyar ruwa. Tasirin kamar haɗin kai, tsangwama a cikin fina-finai na bakin ciki, ko rarrabuwa—waɗanda ke ƙara dacewa don ƙananan na'urori da jerin VCSEL—sun waje da samfurin gani na geometric. Yayin da filin ke tafiya zuwa dToF na tushen SPAD tare da lokacin picosecond, wannan ya zama babban iyaka. A ƙarshe, tabbatarwa da bayanan na'urar na duniya ta zahiri kawai an nuna shi ne; ba tare da ma'aunin kuskure na ƙididdiga da kyamarori na zahiri ba, ikon hasashen simulashin ya kasance magana.
Ga masu haɗa tsarin ToF da masu ƙira, wannan takarda tana ba da zanen aiki. Aiki na 1: Karɓi tunanin bincike na OPL-centric. Lokacin gyara kurakuran zurfi, da farko zayyana bambance-bambancen hanyar gani da ake zargi a cikin wurin ku. Aiki na 2: Yi amfani da wannan tsarin simulashin a cikin lokacin ƙira-don-sana'a. Kar kawai kwaikwayi ruwan tabarau mai kyau; kwaikwayi shi tare da jurewa sannan ka bincika kasafin kuɗin kuskuren zurfi. Aiki na 3: Tura tsarin ƙarin. Haɗa shi da kayan aikin sarrafa ƙira na lantarki (EDA) don haɗa kai da simulashin tushen hayaniya na gani da na lantarki. Nan gaba na ToF yana cikin wannan haɗin gwiwar ƙira. Al'ummar bincike yakamata su gina akan wannan ta hanyar buɗe tushen irin waɗannan bututun, kamar yadda Stanford's Open3D ko MIT's aikin hoto na wucin gadi ya ba da damar binciken jigilar haske. Manufa ta ƙarshe ita ce "tagwayen dijital" don na'urorin ToF—wannan takarda mataki ne na tushe a wannan shugabanci, amma nauyin tabbatarwa, hanzari, da haɗawa ya rage.
Tsarin simulashin da aka tsara yana buɗe hanyoyi da yawa don aikin gaba da aikace-aikace: