Image Sensor / Optical Components

Those lenses are okay for diffusers, or for light distribution, but not high enough quality for use in this project.

What could be very useful, is a custom ‘fly eye’ lens for giving an even lighting field from an LED light, with no vignetting in the corners, like Nikon use on their microscopes.

Jeff, thanks for the MC500 recommendation (and Peter for the Componon-S recommendation, and everybody else in this forum for all the information you are sharing)! I had a MC500 delivered and have been doing some experimentation with it. I have little to compare it to - it does have some noise and its dynamic range is poorer than my Canon 7D’s - but I think its value for money is unbeatable.

I managed to control the MC500 programmatically in Linux using V4L4J. I can at least enabled/disable auto-exposure, set manual exposure values and the capture resolution. My plan is to take two or three captures of each frame with bracketed exposures and then combine them into an HDR image.

Here is an early experiment.

Short exposure:

image

Longer exposure:

image

HDR image (created with Photoshop) :

![image](upload://22yHojn8joWsLUWot530tbIJOvw.jpeg)

The settings of the camera were default, the lightning is just a white cob led bounced from a white sheet of paper and the captured frame is only using around 2000 x 1500 pixels of the 2500 x 2000 pixel sensor, so I think there is room for improvement still. I think the multi-exposure capture solves the problem with MC500’s limited dynamic range (or any sensor’s in comparison to film, for that matter). There is some sensor noise apparent in the darker areas, but I think the “long exposure” should have been a bit longer still to get more detail and less noise in the shadows.

There is some “glow” around the bright areas in the longer exposure frame, not sure what’s causing it and if anything can be done to remove it. The lens is a Componon-S 50mm f2.8 set to f5.6 aperture.

I’m continuing my experimentation, but I am optimistic about being able to tweak the MC500 to suit my requirements.

Nice work!

The glow is most likely blooming on the MC500 itself, we noticed the same thing here, its well depth is relatively poor, so blooming tends to be the result. As stated, for the price, the MC500 is good value, especially when experimenting to get everything else working, and you can get quite reasonable results, but a better sensor does deliver considerably better results, and with a single pass, and you can always upgrade once the rest of the system has the bugs ironed out.

Thanks! I did some more testing. The blooming I am seeing seems to be caused by most part by the sensor heating. Turning off the lighting between shots considerably lessens it it and leaving the light on quickly brings it back. See attached examples of a longer exposure and an HDR merge - little or no blooming there.

image


image

In case the blooming can be overcome with turning the lighting on just for the captures, and merging exposure bracketed captures can solve the limited dynamic range, then I think it is mostly a question of optimizing capture time (1-2 frames a second vs 15 frames a second?) or optimizing the cost (<$100 for MC500, $700-$1000 for a better sensor?). Would be interesting to compare the image quality of a MC500 HDR merge such as this with one of the more expensive devices!

It will be interesting to see if the sensor heats up and has the issue anyway with a longer capture session.

The challenge with merging images is not ending up with flicker and colour variation frame to frame, if you can sort that out, it becomes a viable method.

You can also get around some of the noise by frame stacking as well, but the hassle and extra time post-processing, and the extra capture time soon outweighs the one off cost of spending the extra $350 or so for a better sensor, where you can capture in a single pass at 24fps in real time, with less noise and better image quality, and less post-processing required - especially if you are digitising a lot of film.

If you have plenty of free time, or only have a few films, or if on a tight budget and want a better than ‘pointing a video camera at the wall’ result, then the MC500 certainly delivers.

Maybe interesting: Imec Presents Compact Lens-Free Digital Microscope. Can’t find anything about color caption. See: http://www.imec-nl.nl/nl_en/press/imec-news/lensfree-microscopy.html

Holographic microscopes don’t capture in colour AFAIK, and the images won’t work well with film, but it is interesting tech.

Hi, we are a library and archive that specializes in automotive materials. We have successfully crowdfunded the money to build a kinograph to digitize our (mostly 16mm) films. We originally set a goal of $3500 to build the kinograph and buy a camera and have surpassed our goal by a decent amount which gives us plenty of wiggle room to get a high quality camera for our setup. We originally budgeted around $2500 for the camera in hopes of securing one that not only fits our needs for this project but that could also be used in other library activities as we like to make videos and such. The short list is below.


Sony A7s II
Specs for the camera are linked above
Price - $3,000 (plus Lens)
Still Resolution - 4240x2384
Video - 4K
Shutter - Electronically Controlled, Vertical-Traverse, Focal-plane type


Panasonic GX8 – this looks to be our front runner due to the balance of cost and specs
The specs are on the link above, but you have to scroll down the page a bit to find the,.
Price - $1,000 (plus Lens)
Still Resolution – 5184x3888
Video - 4K
Shutter – Electronic + Mechanical and a Full Electronic Mode


Leica V-Lux (typ 114) - PDF link
Price - $1,200
Still Resolution – 5472x3648
Video - 4K
Shutter – Mechanical mode and a Full Electronic Mode


I had sent a previous list of cameras that could be used to Matthew to gain some insight and he sent back some notes stressing the 3-5 fps that we might top out at due to the rolling shutters. Anybody have experience with these cameras that can give insight to a reliable source on what fps they can hit with the electronic shutter. At 3-5 fps it basically puts us at a 9 minute film taking about 45 minutes to an hour or so to digitize, which while not ideal, it is feasible for us during the work day. That said we don't have concrete answers on the speeds they can hit yet.

After seeing some of the camera and sensor options here on the forum we were wondering if it would be beneficial to build a camera/lens setup rather than buy one of these cameras. Our concerns with this: Would the “DIY” camera setup significantly improve the speed and/or quality over the Sony, Panasonic, or Leica models? Some of the ones listed here still only do about 3fps which those cameras should be able to achieve. Since the budget we have for this may be higher than a budget build kinograph we obviously have a higher ceiling for a “DIY” type camera, but is there anything that will fit our needs and improve upon the options already provided by the cameras above? Another option we have floating is the idea to buy a camera from our list and then possibly upgrade to a “DIY” camera once the tech is to a point where it makes sense to switch or if the community finds a solution that works great for the setup.

I think basically what it boils down to is if you had the budget of up to $3-5k to spend on the camera for your setup what would some of your options be for the best quality output?

Those are standard digital cameras, which according to the discussion here may not be the best use of funds. Most solutions here are looking to use a dedicated “computer vision” camera with a special software interface to the computer to capture the images.

One thing to keep in mind is that if those cameras only operate via a physical shutter, you’ll wear it out very quickly. If they can be set to use electronic shutter only, then they might be able to be used. Would you be using them in still mode or video mode? I would think still mode using an electronic shutter and a fast shutter speed might possibly work.

Also, I don’t think you can just go out and buy a stock lens and get it to work. Most kinograph type solutions are using lenses mounted backward with various extension tubes, etc.

The machine vision cameras can range from anywhere from $265 on up to several thousand depending on what you want. But, I’m not sure anyone here has a ready-to-use software package to trigger the lighting, activate the camera, and advance the film.

All 3 of those cameras have a full electronic shutter mode. We were going to shoot for still mode with the electronic shutter as a means of capture. The cameras above also have cable release triggers. This would be similar to Matthew Epler’s original design that he used with a digital camera. For us this would be an easy way to get our feet wet while still getting good image quality.

As stated in my previous post, what type of cameras such as the more DIY machine vision camera you speak of, would be comparative to the types of image resolution we could capture using one of the cameras I had listed? You said it can cost from $265 on up depending on what we want, but part of the problem is we don’t necessarily know what is out there to fully know what we want. We just aren’t that familiar with that style of camera. Any options or ideas are welcome.

Yeah, all the camera options can be confusing, and I’m still learning myself. I’m more interested in the budget/DIY home-use scenario. I made an excel sheet of Point Grey Research’s cameras, but I can’t upload it to this forum, and don’t have access to Google Docs while at work to create a sharable link.

In the below table are some of PGR’s offerings. The color versions are the same price, so the one you choose would determine if you want to deal with 3 exposures for each color using a mono camera vs. having a full color camera taking a single exposure. I’ve filtered out anything with a rolling shutter of any sort. Some of these are the same camera, just different interface (USB3 vs. GigE), and some models from different lines appear to have the same sensor and specifications, so the difference is mainly the physical housing, I’d guess (i.e. the 2k and 4k options I’ve listed). There’s other options beyond these models, as this is just a small sample.

Camera family	Interface   MODEL	        TYPE	MP	 PRICE 	        MAX RES.	SENSOR DESCRIPTION	                SHUTTER	        MAX FPS
Budget "HD" options.  These should allow capture at the minimum HD defined resolution of 1280x720:
BlackFly	GigE	    BFLY-PGE-13H2M-CS	Mono	1.3 MP	 $265.00 	1288 x 964	Sharp RJ33J4CA3DE CCD, 1/3", 3.75 µm	Global shutter	30 FPS
BlackFly	GigE	    BFLY-PGE-12A2M-CS	Mono	1.2 MP	 $295.00 	1280 x 960	Aptina AR0134 CMOS, 1/3", 3.75 µm	Global shutter	52 FPS
Chameleon3	USB3	    CM3-U3-13S2M-CS	Mono	1.3 MP	 $295.00 	1288 x 964	Sony ICX445 CCD, 1/3", 3.75 µm	        Global shutter	30 FPS
BlackFly	GigE	    BFLY-PGE-13S2M-CS	Mono	1.3 MP	 $345.00 	1288 x 964	Sony ICX445 CCD, 1/3", 3.75 µm	        Global shutter	30 FPS
Blackfly	USB3	    BFLY-U3-13S2M-CS	Mono	1.3 MP	 $345.00 	1288 x 964	Sony ICX445 CCD, 1/3", 3.75 µm	        Global shutter	30 FPS

2K options. 2K can be defined as 2048x1080(for film/video production) and most current "Full HD" tvs almost reach this as their resolution is 1920x1080:
Chameleon3	USB3	CM3-U3-31S4M-CS	        Mono	3.2 MP	 $465.00 	2048 x 1536	Sony IMX265 CMOS, 1/1.8", 3.45 µm	Global shutter	55 FPS
BlackFly	GigE	BFLY-PGE-31S4M-C	Mono	3.2 MP	 $495.00 	2048 x 1536	Sony IMX265 CMOS, 1/1.8", 3.45 µm	Global shutter	35 FPS

4K-ish options. 4K can be defined as 4096x2160(for film/video production) and 3840x2160 (4k tv resolution):
Chameleon3	USB3	CM3-U3-50S5M-CS	        Mono	5.0 MP	 $735.00 	2448 x 2048	Sony IMX264 CMOS, 2/3", 3.45 µm	        Global shutter	35 FPS
BlackFly	GigE	BFLY-PGE-50S5M-C	Mono	5.0 MP	 $795.00 	2448 x 2048	Sony IMX264 CMOS, 2/3", 3.45 µm	        Global shutter	22 FPS

For film, we’re mainly concerned with the second dimension since the film frame isn’t as wide as an HDTV screen. I’m guessing some of these cameras may not have a “widescreen” aspect ratio, but I haven’t looked that much into them.

Awesome, thanks for the list. One that I found which seems like it might work, but am not sure on the price or it’s complete functionality are these 2 cameras from Teledyne - the Falcon and Genie. Would those work for this application? And if so, am I reading correctly that they have a higher megapixel and resolution count compared to the list you just provided?

By the specs it appears that you are correct. I’d be interested in knowing the price if you find out.

This is something I’d like to focus on when the time comes - something that anyone could use whether it’s with a Kinograph or their own alternative. A platform-agnostic control software for scanning machines.

As we built, the features we need will reveal themselves, but I think a good starting point would be:

  • detection of new frame (can be any digital input signal with a falling or rising edge)
  • light on/off
  • motor speed control
  • feedback inputs (analog and digital)

M

I haven’t had a chance to research it, but do the micro 4/3’s shutter wear out the same as a DSLR?

I’d assume that any mechanical shutter will wear out, including those on 4/3rds cameras. While professional level DSLRs are often rated for 100,000 or more cycles, that can get used up pretty quickly if scanning film. Consumer level DSLRs are often rated for an even smaller cycle.

That said, you don’t have to go with an SLR and a mechanical shutter. Many of the more recent models of cameras offer an electronic shutter option, and most of the non-DSLRs are all electronic shutter.

This is the best/simplest demonstration of CCD vs. CMOS sensors I’ve found so far. Thought I’d post here for anyone who was interested:

A beautiful couple of animations, but the description is incorrect CCD vs CMOS. Global or Rolling shutters can apply to both types. See this simple explanation from Point Grey.

Jeff

Thanks for setting the record straight, @VitalSparks!

Found a couple new interesting resources.

  1. Cine2Digits lighting module. “Good for CCD-Bayer filters.”

  2. XENA’s description of their lighting system, which they say is superior to integrating spheres.