The Science Behind the Camera Test: Understanding How to Make Better Images through Research by Michael Kent

Share this post

Reference: StudentFilmmakers Magazine, May 2007. The Science Behind the Camera Test: Understanding How to Make Better Images through Research by Michael Kent. Pages 44, 46 – 49.

As a recent graduate of television studies, I possibly have a different perspective to many established DP’s and shooters. Having shot a few shorts, a full season of race car TV, production managed a feature, and shot four documentaries, all on DVD – to call me ‘seasoned’ would be a gross overstatement. However, I am at that interesting stage in my career where the basics have become basic, and I am anxious to learn more of the industry’s subtleties.

Over these various projects I have become infatuated with the science and engineering side of the business – a path
I would not have seen myself following in my first year of university. But after some real world experience, I seemed to run into problems on set that required technical, scientific or engineering solutions. I’m not alone on this – I have many friends who are backed into the same corner, but that is arguably what keeps this business so exciting. It is constantly changing, but we have ourselves to blame for this. With all the advancement we have made in terms of
software upgrades, plug-in creations, and the like, we must never forget that at the end of the day, it is shot quality that counts. I have found so much software that makes the editing life simple – Boris FX, After Effects, Magic Bullet – I just can’t get enough of them, but on one of my latest docs the one scene that caught my eye amidst all of the
CG and titles was a smooth, three minute long Steadicam walk that was properly exposed, with every color accurately
captured.

Capturing stunning images should be our utmost goal. My best friend and rig-flyer Sean Sealey has been advancing
his Steadicam career at an astonishing pace. He’s been offered some fantastic work that he’s turned down which I found very surprising. I’ve never called him on it, but I’ve gone to bed many nights thinking “Gosh, what a fool! He could have nailed a feature and networked with a feature crew!” Instead of countering my discontent, Sean spends his time training his eye with a Nikon D50. His rationale makes perfect sense. A first impression can only be made
once; Sean ensures that the first impression he makes is always flawless – even if it means perfecting his eye for composition, exposure and balance over the next 10 years. That is why he will always be an “A” list contender. Not everyone can afford to buy a D50 and start training their eyes, but what most people can afford when working
on set is to ensure that their cameras are aligned properly. Producing camera alignment tools is both an art and a science.

Misaligning a camera with a home made or inaccurate chip chart will produce inferior images and be costly to your career. It was an interesting happenstance that introduced me to DSC Laboratories. My university uses DSC products and on graduating my professor told me that DSC was looking for good people. Working with DSC has
opened my eyes to the benefits accurate camera alignment can make to a DP’s career.

The Science Behind DSC Lab’s Products For over 40 years, DSC Laboratories has been perfecting the science of image improvement. We customize patterns for nearly every television station and blockbuster feature shot digitally, and do both through strategic, environmentally sound practices. We have created nearly 50 different patterns in front and rear-lit configurations that can test everything from color balance to lens resolution on any type of digital video camera. Much of what we do, I cannot publish, but what I can, I will share.

Before getting into the fine details, it is always good scientific practice to set up a baseline. For our purposes the baseline revolves around understanding why the need for accurate test patterns exists.

There are three elements that need to be ‘baselined’, and they all involve the flow of how the image is originally perceived to how the final product is viewed on screen.

The Human Eye Without getting into too much biological detail, recreating what we see from behind the lens can only be achieved if we understand why we see what we see. The eye perceives images through photosensitivity – interpreted through rods and cones in the retina. The fovea is a spot within the retina that is densely packed with
cones – all of which have a specific red, green or blue protein separation, and thus help produce a visible color spectrum. If all three cones are activated at the same time, they tell our brain we see white light.

If some are firing at full intensity and some are not, we see colors in between white and black. Rods are very sensitive to light information and do not “see” color. Cones have color sensitivity, but require more light to function.

The way we measure how sensitive cones are to light energy is through wavelengths. Nanometers are distance
markers that help us quantify what wavelengths of light make up our visible spectrum. The spectral band of colors
goes from Ultra-violet (below 400nm) up to Infrared (above 700nm). Individuals perceive color information differently, but for the most part, the human visible spectrum ranges from 400nm (blue) to as high as about 675nm (red). As evidenced in this diagram, all colors we can see within that range fall in this spectrum. Camera manufacturers must take these factors into account when designing a color camera’s taking characteristics.

The Camera

While the peak sensitivity of human vision is about Blue 440nm, Green 540nm and Red 580nm visually “true” Red for most people has a spectral peak of about 635nm.

A camera and television set must reproduce the color at the same wavelength if we are to accurately recreate that color. This is a difficult task, given the characteristics of monitors and TV sets, also the fact that they are seldom set up properly. CCDs transmit information by converting light to voltage levels that vary with brightness. Ensuring
that a digital image when recreated will be of a particular spectral curvature becomes a complicated science. In most cases, cameras try to mimic the cones found in the fovea with three CCDs, one each for each of the Red, Green and Blue primary colors.

In taking a picture of a banana the yellow color will activate primarily the red and green sensors – by the time we see
that same banana on television, its color will incorporate the characteristics (i.e., color space) of the television set, which, in theory, should be similar to the “look” of the original banana.

Data from CCDs is converted to digital voltage levels whose characteristics can be adjusted using a camera’s color matrix.

CCDs interpret images based on specific color thresholds engineered into their algorithmic circuitry. Optimizing these thresholds largely depends on how deep the camera’s menu will go. Higher end cameras allow an enormous amount of matrix setting control, but without accurate monitoring materials, it is impossible to line up these colors with any form of consistency.

The Waveform Monitor

The waveform monitor is one of two important tools that make up an effective camera shooter’s toolkit. A waveform
monitor displays a video signal in millivolts (mV) for HD and IRE (Institute of Radio Engineers) for NTSC. 700mV is equal to 100 IRE units.

Using a precision DSC grayscale aligning a camera is usually quite simple. First, light the ChromaDuMonde or other DSC chart evenly, and then adjust camera settings to produce equal spacing of the grayscale’s step signals between 0 and 700mV – this should result in accurate grayscale alignment.

When selecting a test pattern, dynamic range becomes very important. In the early days of television, the dynamic range of a camera was very limited, to about 20 or 25:1. Grayscale pattern had a similar low dynamic range.

Unfortunately, such test charts are still being sold to align modern cameras which have a dynamic range of
3000:1 or higher. The progression rate between most 9 step charts and a modern 11 step pattern is also different. The lightest step on a 9 step chart is 60% reflection where a similar 11 step chart has a reflection of 90% (on the first step). Obviously, different densitometric curves between test patterns will result in different image reproduction from cameras. Aligning to an inaccurate or grayscale of limited dynamic range can result in poor quality images that are beyond redemption!

The need for an accurate Grayscale chart is quintessential to setting precise gamma and exposure. The white chip on
a chart is normally set to 700mv (100 IRE) and true black close to 0 mV. Setting the black can be tricky. NTSC has a basic black level set up of 7.5 IRE above true black. It would seem logical to set the blacks at 0; however, absolute black is an impractical goal because all surfaces reflect some light, however little. Even the blacks found on DSC CamAlign charts with their special ultra high reflective gloss surface, while way below regular charts do reflect some light.

In addition, with front lit test pattern configurations lens flare will contribute light and in turn, raise the black level.

Similarly with DSC’s rear lit Combi patterns where black is true black (opaque), zero black is still impossible to obtain.

The Vectorscope

The vectorscope is the other ‘must have’ tool that should be part of every shooter’s toolkit. Only after setting a camera’s exposure levels and tracking can a shooter move on to optimizing color. DSC uses the latest International Telecommunications Union standard for HD color ITU-R BT.709, (SMPTE 274M). ITU 709 standard largely replaces a number of previous standards. While the NTSC colorimetry standard had an excellent wide color gamut (when TV sets were made to the standard) they didn’t sell, because the picture was too dim and had to be viewed in a darkened room. For this reason SMPTE C and Europe’s EBU were born and for 25 years DSC made charts to both these standards.

DSC standardized on 709 five years ago and provides printouts showing the offsets, on the rare times people ask for them. Every primary color on a DSC chart is designed to line up in its respective vectorscope box.

This ChromaDuMonde has a total of 28 colors – the primary colors in the corners and the center fall into their vectorscope boxes. The colors in between form a straight line between the primary colors when properly aligned.

Why not just the primaries? With the advanced multi-matrix settings of modern cameras, it is possible to line up each primary in their box, but reduce its working colorspace. The image below has been aligned to a 6 color pattern using a multi-matrix menu – as you can see; each primary is in its box. The following image is what a 28 color chart looks like with those settings – clearly the intermediary colors are completely skewed although the primaries are lined up.

Why are those boxes so important?

DSC primary colors not only line up to their respective boxes, but also have uniform luminance values, an exceedingly difficult task. DSC charts are renowned for their accuracy and consistency. When aligning a camera’s matrix settings increasing or decreasing the sensitivity of a particular color will also change other colors. This often requires considerable patience in camera alignment. Always remember that the red and yellow are the most important as they largely define skin tone. Here is an example of camera operator error – if you line up red on the vector box to someone on set with a red sweater you will be altering the colorspace and gamut to reflect that red as your ‘true’ red. This is exemplified when going into the camera’s matrix settings to adjust color sensitivity. Notice that in many instances a particular color cannot be selected on its own. Instead, alterations can be made to B-Y, R-B, etc. When these settings are increased or decreased, all of the other colors on the vectorscope shift – adjusting one color incorrectly will affect the entire colorspace and this should always be taken into consideration. Lining up to an incorrect color is worse than lining up to no colors at all – much like the common error of white balancing to a sheet of white printer paper.

Most camera manufacturers preset their matrixes on the warm side and many clients using DSC charts to align $60,000+ cameras discover that red and yellow are the only colors that fall in the vectorscope boxes when they take their new cameras straight “out of the box”.

Camera matching becomes impossible without an accurate test pattern. No two cameras are identical making it unreliable to upload the settings from one camera to another. The only way to match cameras effectively is to use the same pattern in the same lighting condition. Set exposure levels, white balance (which can be done to the entire chart with colors as it is spectrophotometrically neutral) align the colors on a vectorscope and then match the second camera to a freeze frame of that pattern.

More often than not, the perfectly aligned vectorscope image is not the ideal setting based on the characteristics of the scene. However, by first aligning to a DSC pattern, it becomes an accurate baseline with the widest colorspace as a reference. Small changes from there are not always recommended, but up to the operator.
Color Correction

Many of our clients have used our test patterns in post production as well. When initially setting up a camera, common practice is to do so at 2000 lux, with both light sources at 45 degrees. This however, does not necessarily mimic real life shooting scenarios and that is why DSC Labs test patterns can also be used as color correction tools. By recording just a few seconds of a DSC test pattern at the head or tail of a scene, the lighting characteristics of that scene are recorded to tape, and editors can use this data to color correct or match scenes in post. Secondary colors are very useful in post, provided a DSC ChromaDuMonde pattern was shot on set; they are used to see changes to other colors when one color is changed dramatically.

This entire article has focused on the SRW ChromaDuMonde 28R – just one of the 50+ patterns offered by DSC Labs. Whether shooting on MiniDV or DVCPro, there is a chart designed for every application.
We hope that you will join the growing number of shooters who understand how to make better images through esearch.

Michael Kent, 24, is a graduate of the Radio and Television Arts program at Ryerson University and is the Operations Manager at DSC Laboratories. Michael has spoken at the SMPTE HD Boot Camp II, and has helped create the Educational text: Imaging Technology: Advanced Guide to Digital, Film and Television Production. Michael also owns Skinnee Pant Productions, races at an SCCA Pro Driver level and is a registered ACTRA stunt driver. He currently resides in Toronto, Ontario Canada. 

Leave a Reply

×