Das Patent von Sony sieht so aus:
Zitat:
Quelle: http://www.imaging-resource.com/ee_u...tent_index.png
„Sony’s patent publication (which means it has been filed and acknowledged, but not yet denied or approved), lays out the groundwork for being able to create unique microlenses with different profiles (shapes), so that the microlenses over the phase-detection pixels could direct light coming only from the desired direction onto the surface of the pixel, without requiring shading. This technology is significant and unique for the fact that the patent implies this would be done on a pixel-by-pixel basis.
This change would give the pixels and overall system much more information to work with in low-light situations."
Quelle: http://www.imaging-resource.com/news...his-patent-bec
|
Das ältere Patent 2013 mit dem Trick der polarisierten Mikrolinsen:
http://www.sonyalpharumors.com/sony-...-af-technique/
Zitat:
http://i1266.photobucket.com/albums/...ps1a76f56d.png Bildquelle: http://i1266.photobucket.com/albums/...ps1a76f56d.png
Description, self-interpretation of the patent literature
Patent Publication No. 2013-57769
Published 2013.3.28
Filing date 2011.9.8
Pixel in the phase difference AF (image plane phase difference AF)
It is understood the amount of deviation of focus
It is necessary to shield the pixel partially, disadvantage in low light
It can not detect the phase difference in a low-contrast
Sony patents
Pixel does not have the sensitivity to polarization
Using a polarizer, to have a sensitivity to a specific direction of polarization
Polarization pixels
It is arranged to overlap the polarizer to the pixel
It is placed in pixels which are not read out at the time of the green pixels and videos of the Bayer array
It is not necessary to shield the pixel, advantageous in low light
A plurality of types of polarizer is used
What the angle of the polarization main axis is different, those passing the TE wave (p-polarized light) or equivalent only, TM wave (s-polarized light) that passed through the equivalent only, etc
For example, using the luminance information of two pixel pairs polarization angle difference was 90 °
Using brightness information of the pixel polarizer is arranged a (≒ polarization information) in AF
Perform the AF even at low contrast
To divert the algorithm of contrast AF
By the brightness information that changes in the subject and the in-focus state, the focus determination
Quelle mit Google Translator übersetzt: http://translate.google.com/translat...p%2F2013-04-01
Originalquelle aus egami-Blog: http://egami.blog.so-net.ne.jp/2013-04-01
|
Läuft offenbar über speziell gekrümmte Mikrolinsen an bestimmten Pixeldioden oder Polfilter/Polarisierung an jeder Pixeldiode, scheint wohl mehrere Patente zu geben.
Bei Canons Dual Pixel CMOS AF (PDAF auf dem Sensor mit einem zweigeteilten Dualen Pixel die sich eine Mikrolinsen-Farbfilterfarbe teilt):
http://www.canon.de/for_home/product...pixel_cmos_af/
PDF-Datei zum Canon EOS 70D DUAL PIXEL CMOS AF: https://www.canon.de/images/EOS%2070...83-1070371.pdf
Kurzerklärung in englischer Sprache:
http://www.betterphotography.com/ind...sensor-and-why
Ein Pixel wird zweigeteilt, scheint aber unproblematisch zu sein, wenn es um's Rauschen geht (logisch, nur weil zwei Pixel geteilt, werden, heißt es noch lange nicht, dass mehr Rauschen ist, denn die teilen sich die selbe Farbinformation von der Mikrolinse, dies ist mathematisch schon eine Addition, der Informationen, ergo null Problem):
Zitat:
Canon DUAL PIXEL CMOS AF Technik:
Quelle: http://www.imaging-resource.com/PROD...N-70D-DAF1.PNG
„The breakthrough: Canon Dual Pixel CMOS AF. With the sensor used in the EOS 70D, Canon broke the mold, asking why it shouldn't be possible to equip every sensor pixel with phase-detect capability. Of course, in doing so, they couldn't arbitrarily throw away half of the light falling on the chip, by shading the pixels the way conventional on-chip phase-detect systems do. That would cost them a full stop of ISO sensitivity, and who knows what odd optical artifacts might result as well.
The solution was to split each sensor pixel into two sub-pixels, each receiving light from the half of the lens' light cone that it normally would in a sensor with conventional microlenses. I'm not enough of an optical scientist to be able to describe in detail how light travels through microlenses, but it seems reasonable that light rays striking a microlens from one side of the frame end up hitting the same side of the underlying photodiode. It's not clear at this point whether Canon figured out a way to insert a light barrier vertically into the microlens structure to improve the separation between light rays arriving from opposite directions, if they somehow altered the microlens structure to achieve the same end, or whether conventional microlens structures inherently create enough separation on their own. Whatever they did, it apparently does give them enough separation between light arriving from different directions that they can perform effective phase detection.
However they're manipulating the incoming light, they've split the photodiode for each pixel into two halves, so they can read out image data corresponding to top-arriving and bottom-arriving light rays separately. Based on what Canon has told us, they can also read out both sides simultaneously, for normal imaging. In fact, it appears that they can read out data from the two pixel halves separately and together at the same time, when they're recording video. That sounds a little tricky to do on a pixel-by-pixel basis, but perhaps they're taking advantage of the much lower resolution of video to do the two different types of readout on alternating rows of pixels.
Canon 70D review -- Dual AF pixels
We believe that the split pixels are still managing to collect the same amount of light overall as a conventional, un-split structure, based on a study of the Canon 70D's low light/high-ISO performance. Were they only collecting only a portion of the incoming light, we'd have expected that to show up as significantly decreased sensitivity/higher noise, and we didn't see any such increase. It's possible, though, that sensitivity / noise improvements in other areas of the sensor and imaging pipeline are sufficient to mask any deficit caused by the split-pixel design. Only Canon knows for sure if there's a drawback at the sensor level, but the important thing to note is that overall we saw slightly better high ISO performance from the 70D than its predecessor, while at the same time it offers the uniquebenefits of Dual Pixel CMOS AF.
The Canon 70D's Dual Pixel AF operation varies slightly, depending on whether it's in still shooting or video modes: When running in still capture mode, the camera determines the degree of focus adjustment needed before commanding the lens to move, while in video capture mode, focus determination and lens movement occur continuously and simultaneously."
Quelle: http://www.imaging-resource.com/PROD...L_PIXEL_AF.HTM
|
Offenbar häufig in Foren und selbst unter Canon Fans verwechselt (guter Kumpel von mir, meinte als Canon User wirklich ernsthaft, die 70D cutet die Pixel dank Dual Pixel und daher mehr Rauschen, dem ist offenbar nicht so):
Kritik hagelte es da unter anderem beim dpreview-Interview seiner Zeit und woanders bei den Canon-Sensoren im Allgemeinen, denn die können bis heute keine 14 Bit:
Zitat:
„One thing we know from our own testing is that Canon DSLR sensors can’t quite compete with some modern sensors from Sony in terms of dynamic range. How important to you is developing sensor technology?
We are very focused on getting the best image quality. I’m not sure what measurements you’re looking at but when it comes to dynamic range for example we consider image quality as a whole, from low to high ISO sensitivities and on balance we consider our sensors to be the best.
My ideal camera is one that can take a picture in any environment from complete darkness to the brightest sunshine.
So in your opinion your sensors are currently the best on the market?
Yes. In the EOS 7D Mark II for example the sensor we’ve used is improved compared to the previous generation, especially at high ISO and in shadows. There’s less noise.
Currently no Canon camera offers more than 22MP.
Do your DSLR customers ask for higher resolution?
Yes. We know that many of our customers need more resolution and this is under consideration. In the very near future you can expect us to show something in terms of mirrorless and also a higher resolution sensor."
Quelle: http://www.dpreview.com/articles/707...ry-near-future
|
Neueres Interview von Canon:
http://www.dpreview.com/articles/530...aying-speed-up
Warum ich das mit Canon hier alles vergleiche? Weil sowohl Sony als auch Canon, die einzigen Marktteilnehmer sind, die PDAF auf dem Sensor gut können!
Zurück zu Sony:
PDAF im Allgemeinen:
https://photographylife.com/how-phas...utofocus-works
Fazit:
Ich denke der PDAF funktioniert wunderbar bei Sony und Canon, das Zeug wird irgendwann mal den Phasendetektions-Autofokus an externen Modulen auch bei den D-SLR vielleicht überholen, wäre nicht der Vorteil wie an SLT, dass man selbst bei Darkframes einen permanenten AF hat, den löst der Kamerasensor aus, haste kein Bild für Autofokus in dieser kurzen Zeit mehr. In der Praxis noch unproblematisch, da die Kameras nur 1/4000 und 1/8000 Sekunden packen und nur 11-14 Bilder pro Sekunde machen.
Da haste genug Pufferzeit auch rein mathematisch.
Ich habe eine Idee hier:
Warum nicht die externen PDAF-Module bei D-SLT und D-SLR Kameras irgendwann mal allesamt, wenn die Technik soweit ist zu CMOS-Bildsensoren umbauen, die PDAF auf dem Sensor nutzen und so etwas wie Kreuzsensoren und Doppelkreuzsensoren emulieren können, dann hat sich die ganze Kiste mit den nervigen Ausbau an deutlich mehr an AF-Sensoren in externen AF-Modulen erledigt.
Man darf nicht vergessen früher waren AF-Sensoren noch aus CCD-Sensoren gefertigt und optische Glaselemente ähnlich wie Mikrolinsen (oder sind diese Gräben und Vertiefungen an AF-Sensoren defacto so etwas wie Mikrolinsen? Ja ne?) haben die Sensoren nochmal mit korrekt zugewiesenen Licht versorgt um eine reibungslose Phasendetektionsautofokus-Messung zu ermöglichen, heutzutage sind es CMOS-Sensoren! Ja der AF selbst bei D-SLR und D-SLT Kameras hat schon Bildsensoren sozusagen drinne, die sind aber riesige lichtempfindliche Phasen-AF Sensoren noch und keine mehrzeilige verpixelte Bildaufnahemsensoren!!!!
Der AF ist sozusagen ein CCD/CMOS-Bildsensor, aber mit riesigen Strukturen, die sind allesamt auch sichtbar in der Verdrahtung. Bilder reiche ich nach:
Allgemeines zur Diskussionslage und den Verhalten mancher Mitforen-User:
Sollten Leute wieder mal Fehler in meiner Argumentation und Technikwissen (man unterschätze bítte mein über 10 Jahre langes Fotografiepraxis und vorheriges Technikwissen in der Computertechnik seit meiner Kindheit und Jugend nicht, ich hab sogar all den Shit meiner Physiklehrer aufgesogen, ich habe von den meisten hier im Forum und Autodidaktik jede Menge gelernt und bin dankbar und habe diverse Quellen an Wissen geschluckt und ich mag echt jeden Experten hier, nicht umsonst habe ich einige in meiner Freundesliste, da sie sehr kompetent und hilfsbereit sind!

Ich schätze all diese Leute sehr) haben, dann bitte sofort ergänzen und Korrekturen bei mir einfordern, wollen ja nicht, dass wieder sich Leute aufregen, wenn ich ab und an falsch liege. Niemand ist perfekt.


Selbst ich liege ab und an bissl falsch, aber niemals komplett daneben. Ihr kennt mich ja alle.
Habt euch bitte hier lieb und keine unnötigen Sigma oder Adapterstreits, die Adapter sind Fremdherstellerlösungen und gehören in anderen Topics/Threads. Immer schön ruhig bleiben liebe Leute, nicht immer verbal im Internet draufhauen. Danke.
Noch sonstige Fragen oder Zweifel bei mir?

=)
Danke.
Edit by DonFredo: Direkt eingebundene eigene Bilder dürfen nach den Forenregeln nur max. 600x600 Pixel und max. 200 KB haben. Sonst bitte die Galerie nutzen. Dort sind Bildgrößen bis max. 1800x1200 Pixel und max. 1024 KB möglich.