6by9
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 7149
Joined: Wed Dec 04, 2013 11:27 am
Location: ZZ9 Plural Z Alpha, aka just outside Cambridge.

Re: Solution planned for the bad lenses of the V2 8MP RasPi

Mon May 09, 2016 5:09 pm

Mettauk wrote:Any official news before I return 4 of the new V2 cameras i purchased?
Analysis still ongoing as it does not show up on all modules.
I've asked those at Pi Towers to make an official comment - I'm only a volunteer.
Software Engineer at Raspberry Pi Trading. Views expressed are still personal views.
I'm not interested in doing contracts for bespoke functionality - please don't ask.

User avatar
jbeale
Posts: 3476
Joined: Tue Nov 22, 2011 11:51 pm
Contact: Website

Re: Solution planned for the bad lenses of the V2 8MP RasPi

Mon May 09, 2016 5:39 pm

6by9 wrote:Analysis still ongoing as it does not show up on all modules.
I've asked those at Pi Towers to make an official comment - I'm only a volunteer.
Thank you, and I appreciate the information. I understand "perfect" will not happen, and "good enough" is subjective, but if someone official is willing to post a full-frame full-res image from v2.1 camera showing an outdoor scene including distant scenery (or even indoors, if objects are at least 3 m away), I think that would be a useful baseline.

None of the 4 new v2.1 cameras I got (2 normal, 2 NoIR) perform to my satisfaction for objects >= 3m away (except at the very center of the frame where it is a bit soft, but acceptable). All of my older v1.3 cameras do significantly better than the v2.1 for resolution towards the edges of the frame, on faraway objects.

User avatar
jbeale
Posts: 3476
Joined: Tue Nov 22, 2011 11:51 pm
Contact: Website

Re: Solution planned for the bad lenses of the V2 8MP RasPi

Tue May 10, 2016 2:21 am

I compared v1 (normal) and v2 (NoIR) cameras from the exact same distance, 5.60 meters away from a target that is 251 x 253 mm in size.

There are three crops from each camera: center, lower left corner, and lower right corner. Notice the v2 camera shows much more magnification in the corners than the v1 camera, but also distortion and blurryness. This particular v1 camera isn't good at LR corner, but OK at LL corner. This is the target I used, printed on 11x17 inch ("ledger") size paper: http://www.bealecorner.org/red/test-pat ... attern.pdf

Image

6by9
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 7149
Joined: Wed Dec 04, 2013 11:27 am
Location: ZZ9 Plural Z Alpha, aka just outside Cambridge.

Re: Solution planned for the bad lenses of the V2 8MP RasPi

Tue May 10, 2016 5:44 am

Software Engineer at Raspberry Pi Trading. Views expressed are still personal views.
I'm not interested in doing contracts for bespoke functionality - please don't ask.

User avatar
Mettauk
Posts: 235
Joined: Mon Dec 10, 2012 12:40 pm
Location: Zarg

Re: Solution planned for the bad lenses of the V2 8MP RasPi

Tue May 10, 2016 8:17 pm

Having read Eben on the V2 focus/distortion issues and having ranted a bit I wonder what solution is going to take place?
IMG_20160510_213059.jpg
V2 with M12 mount
IMG_20160510_213059.jpg (26.87 KiB) Viewed 5795 times
Having replaced the mount, glued the lens down and fiddeled around a bit I managed to get a V2 to work as well as the V1 with a wider view. :-( I will not be buying any more official V2 cameras as it really is not worth the hastle or cost. Will the V1 continue to be available until a V2.xx is made?
As humans we have been the same for a very very long time, technology changes how we do... not who we are as people.

User avatar
DougieLawson
Posts: 35814
Joined: Sun Jun 16, 2013 11:19 pm
Location: Basingstoke, UK
Contact: Website Twitter

Re: Solution planned for the bad lenses of the V2 8MP RasPi

Tue May 10, 2016 8:38 pm

As soon as any remaining supply of V1 cameras is sold that's it, no more - ever. The Omnivision imaging sensor is no longer made for that device.

Camera V2 is your only choice unless you can find third party/cloned/OEM cameras. Hacking the hardware to change the focus is the only way you're going to get a CSI camera with the focal length you need.
Note: Having anything humorous in your signature is completely banned on this forum. Wear a tin-foil hat and you'll get a ban.

Any DMs sent on Twitter will be answered next month.

This is a doctor free zone.

User avatar
jbeale
Posts: 3476
Joined: Tue Nov 22, 2011 11:51 pm
Contact: Website

Re: Solution planned for the bad lenses of the V2 8MP RasPi

Tue May 10, 2016 9:28 pm

This solution looks rather promising: viewtopic.php?f=43&t=145815&start=200#p971545

If you don't like that approach, the older v1 boards with OV5647 modules are still available (for now) with M6x0.35, M12x0.5, and CS lens mounts from various 3rd parties (at a range of pricing and quality control). As pointed out, OV5647 is now discontinued so this would be a "lifetime buy" situation.

ianj
Posts: 21
Joined: Mon May 12, 2014 2:24 am

Re: Solution planned for the bad lenses of the V2 8MP RasPi

Fri May 13, 2016 1:42 pm

UGH! And here I was coming to look for information as to why my "new" security camera was so blurry. Hopefully something is being done beyond attempting to adjust the focus... where's the QC these days :lol:

User avatar
jbeale
Posts: 3476
Joined: Tue Nov 22, 2011 11:51 pm
Contact: Website

Re: Solution planned for the bad lenses of the V2 8MP RasPi

Fri May 13, 2016 2:14 pm

ianj wrote:UGH! And here I was coming to look for information as to why my "new" security camera was so blurry. Hopefully something is being done beyond attempting to adjust the focus... where's the QC these days :lol:
You did see Eben's statment linked above, right? viewtopic.php?f=43&t=145815&start=200#p971091

gkreidl
Posts: 6048
Joined: Thu Jan 26, 2012 1:07 pm
Location: Germany

Re: Solution planned for the bad lenses of the V2 8MP RasPi

Fri May 13, 2016 2:39 pm

The best optical resolution you could get in theory (with this lens) is
1.3 * 2.0 = 2.6 micron
With a cheap lens it's at least 3 micron.
The sensor's pixel size is 1.1 micron. You can never expect really sharp 8 MB pictures.
It would require a much greater sensor (2.2 pixel size or more) and a good lens with f/1.4 - f/1.8 to get really sharp images.
Sensor and lens would cost a lot more money, of course.
Minimal Kiosk Browser (kweb)
Slim, fast webkit browser with support for audio+video+playlists+youtube+pdf+download
Optional fullscreen kiosk mode and command interface for embedded applications
Includes omxplayerGUI, an X front end for omxplayer

User avatar
jbeale
Posts: 3476
Joined: Tue Nov 22, 2011 11:51 pm
Contact: Website

Re: Solution planned for the bad lenses of the V2 8MP RasPi

Fri May 13, 2016 3:24 pm

gkreidl wrote:The best optical resolution you could get in theory (with this lens) is 1.3 * 2.0 = 2.6 micron With a cheap lens it's at least 3 micron.
Based on what theory? What is your criterion for "resolution" (what MTF threshold?) The pixel size on the v2.1 sensor is 1.12 microns, should we expect 3 pixels of blur?

Below is a center-of-frame crop from the v2.1 camera, saved as a JPEG and only later blown up to 4x so there are JPEG artifacts but looking at the left ring, I think there is some contrast at the 2 pixel level. At the edge of the frame it does get much worse, on my samples anyway. If I was able to get this level of resolution across the frame, I would be more than satisfied. Some outdoor photos from people who have refocused their v2.1 lenses suggest they are able to get good resolution across the frame (even looking through a window, so I may just be unlucky.

Image
Last edited by jbeale on Fri May 13, 2016 3:38 pm, edited 1 time in total.

gkreidl
Posts: 6048
Joined: Thu Jan 26, 2012 1:07 pm
Location: Germany

Re: Solution planned for the bad lenses of the V2 8MP RasPi

Fri May 13, 2016 3:34 pm

jbeale wrote:
gkreidl wrote:The best optical resolution you could get in theory (with this lens) is 1.3 * 2.0 = 2.6 micron With a cheap lens it's at least 3 micron.
Based on what theory? What is your criterion for "resolution" (what MTF threshold?) The pixel size on the v2.1 sensor is 1.12 microns, should we expect 3 pixels of blur?

Below is a center-of-frame crop from the v2.1 camera, blown up to 4x. Looking at the left ring, I think there is some contrast at the 2 pixel level. At the edge of the frame it does get much worse, on my samples anyway. If I was able to get this level of resolution across the frame, I would be more than satisfied. Some outdoor photos from people who have refocused their v2.1 lenses suggest they are able to get good resolution across the frame (even looking through a window, so I may just be unlucky.
http://www.truetex.com/raspberrypi
Minimal Kiosk Browser (kweb)
Slim, fast webkit browser with support for audio+video+playlists+youtube+pdf+download
Optional fullscreen kiosk mode and command interface for embedded applications
Includes omxplayerGUI, an X front end for omxplayer

ianj
Posts: 21
Joined: Mon May 12, 2014 2:24 am

Re: Solution planned for the bad lenses of the V2 8MP RasPi

Fri May 13, 2016 3:42 pm

jbeale wrote:
ianj wrote:UGH! And here I was coming to look for information as to why my "new" security camera was so blurry. Hopefully something is being done beyond attempting to adjust the focus... where's the QC these days :lol:
You did see Eben's statment linked above, right? viewtopic.php?f=43&t=145815&start=200#p971091
I must have missed it, thanks for the link.

Will attempt to adjust the focus and hope it doesn't scratch or break as others have encountered... :|

User avatar
jbeale
Posts: 3476
Joined: Tue Nov 22, 2011 11:51 pm
Contact: Website

Re: Solution planned for the bad lenses of the V2 8MP RasPi

Fri May 13, 2016 4:00 pm

Ok, he states his number is based on the Rayleigh criterion for green light. However, experimentally I see line-space pairs (barely) resolved which have a 3 pixel period. 3 x 1.12 = 3.36 microns. Does that agree with theory? Let's check.

According to [1] the Rayleigh criterion is equivalent to a line-to-line separation of 0.61 (lambda / NA). Let's say green light 550 nm = 0.55 um and (in air) the f-number (N) = 1/(2 * NA). If we have f-number=2.0 then NA = 0.25, therefore Rayleigh says the just-resolved line-line separation is 0.61 * (0.55 um / 0.25) = 1.34 microns, and on our sensor that's only 1.2 pixels. So my results at 3 pixels are well within what diffraction-limit theory says. Let me know if you see any errors in my math.

Anyway, my results look fine to me at the center of the frame. If I could get 3-pixel-period line-space pairs resolved across the frame, it would be great.

[1] http://myplace.frontier.com/~stevebrain ... s/MTF2.pdf

rambutan2000
Posts: 3
Joined: Sat Oct 05, 2013 1:34 am

Re: Solution planned for the bad lenses of the V2 8MP RasPi

Fri Dec 09, 2016 11:04 pm

I'm starting down a long road of using rpi cameras for texture capture, small scale photogrametry etc. I've purchased these before:
http://www.ebay.com/itm/OV5647-Camera-B ... 5731.m3795

And tried out my own various lenses which worked well. However I haven't seen anyone manufacturing the equivalent for the v2 camera. I'm going to try something like this approach:
http://www.truetex.com/raspberrypi

But using an inexpensive mount such as these:
http://www.m12lenses.com/CS-Mount-Lens-Holders-s/60.htm

User avatar
jbeale
Posts: 3476
Joined: Tue Nov 22, 2011 11:51 pm
Contact: Website

Re: Solution planned for the bad lenses of the V2 8MP RasPi

Sat Dec 10, 2016 10:31 pm

The RPi designers prevented 3rd-party compatible v2 boards by using hardware copy-protection on the board. As a result, the only v2 board you can buy is the genuine article licensed by RPi Trading, like the Pi itself.

That doesn't prevent 3rd parties from making a M12 or CS-mount lens holder that could fit over the mount on the original board (at least, after you remove the existing M6 lens in the small rectangular camera block. In fact, I myself made a 3D-printed holder for a special optic (reversed microscope obejctive) which works, although that particular optic turns out to not have great optical quality. Maybe some entrepreneur will do something similar for a M12 or CS-mount.

http://www.thingiverse.com/thing:1565909
Image

linux_author
Posts: 248
Joined: Sat Apr 02, 2016 7:04 pm
Location: Gulf of Mexico
Contact: Website

Re: Solution planned for the bad lenses of the V2 8MP RasPi

Sat Dec 10, 2016 11:46 pm

jbeale wrote:The RPi designers prevented 3rd-party compatible v2 boards by using hardware copy-protection on the board.
why was this done? doesn't make sense to me

User avatar
jbeale
Posts: 3476
Joined: Tue Nov 22, 2011 11:51 pm
Contact: Website

Re: Solution planned for the bad lenses of the V2 8MP RasPi

Sun Dec 11, 2016 12:32 am

The RPi camera board is not open hardware in any sense. There's a comment somewhere in the "semi-official v2 camera thread" but to summarize, they wanted to protect their development time and $$ investment, and their licensed partners who distribute the product.

6by9
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 7149
Joined: Wed Dec 04, 2013 11:27 am
Location: ZZ9 Plural Z Alpha, aka just outside Cambridge.

Re: Solution planned for the bad lenses of the V2 8MP RasPi

Sun Dec 11, 2016 8:28 am

jbeale wrote:The RPi camera board is not open hardware in any sense. There's a comment somewhere in the "semi-official v2 camera thread" but to summarize, they wanted to protect their development time and $$ investment, and their licensed partners who distribute the product.
viewtopic.php?f=43&t=149426&p=982590#p982590 is the official answer that you're thinking of.
Software Engineer at Raspberry Pi Trading. Views expressed are still personal views.
I'm not interested in doing contracts for bespoke functionality - please don't ask.

rambutan2000
Posts: 3
Joined: Sat Oct 05, 2013 1:34 am

Re: Solution planned for the bad lenses of the V2 8MP RasPi

Mon Dec 12, 2016 8:59 pm

Also saw this project:
CS mount adapter for Raspberry Pi Camera board
http://www.thingiverse.com/thing:875711

I agree about licensing and clones etc. But there seems to be a market for compatible cameras with higher grade optics.

User avatar
Mettauk
Posts: 235
Joined: Mon Dec 10, 2012 12:40 pm
Location: Zarg

Re: Solution planned for the bad lenses of the V2 8MP RasPi

Tue Dec 13, 2016 12:37 am

rambutan2000 wrote:... But there seems to be a market for compatible cameras with higher grade optics.
Almost anything is better than the "New" V2 modified camera with its Sony CCD. Even with the unexplained lens adjuster that comes with it.

There is not a working Pi camera available aside from clones of the old V1 camera.

Will a half decent replacement be forthcoming after the failed replacement for the obsolete first one?
As humans we have been the same for a very very long time, technology changes how we do... not who we are as people.

jamesh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 23385
Joined: Sat Jul 30, 2011 7:41 pm

Re: Solution planned for the bad lenses of the V2 8MP RasPi

Tue Dec 13, 2016 10:25 am

Mettauk wrote:
rambutan2000 wrote:... But there seems to be a market for compatible cameras with higher grade optics.
Almost anything is better than the "New" V2 modified camera with its Sony CCD. Even with the unexplained lens adjuster that comes with it.

There is not a working Pi camera available aside from clones of the old V1 camera.

Will a half decent replacement be forthcoming after the failed replacement for the obsolete first one?
The Sony camera works, it's not a failed replacement. It may not be great optically, but it does work.
Principal Software Engineer at Raspberry Pi (Trading) Ltd.
Contrary to popular belief, humorous signatures are allowed. Here's an example...
"My grief counseller just died, luckily, he was so good, I didn't care."

User avatar
Mettauk
Posts: 235
Joined: Mon Dec 10, 2012 12:40 pm
Location: Zarg

Re: Solution planned for the bad lenses of the V2 8MP RasPi

Tue Dec 13, 2016 11:57 am

jamesh wrote: The Sony camera works, it's not a failed replacement. It may not be great optically, but it does work.
That seems like a #PostTruth reply! It sort of works, but most things are out focus except either the edge or the centre, but not both. To name just one major issue.
As humans we have been the same for a very very long time, technology changes how we do... not who we are as people.

Return to “Camera board”