Hasselblad Natural Colour Solution
While image colour and tonality are somewhat subjective, the tones displayed in an image are controlled by the following:
- Pixel size and sensitivity (Dynamic range)
- Number of bits per colour channel in the A/D conversion process
- Colour profile applied to the resulting image data
The Hasselblad Natural Colour Solution (HNCS) was developed for serious users who demand the utmost colour accuracy. This system delivers the best possible natural colours from the selected chip without having to select from multiple presets. The HNCS delivers smoother tonal transitions and a more analogue film-like image quality straight out of the camera, thanks to its 16 bit colour depth, compared to the majority of smaller DSLR sensors.
HNCS in real terms
When we spoke to many of our photographers, the feedback was unanimous:
- Solutions where you have to sacrifice some colours in order to get others right are unacceptable.
- Correct colours need to be in every shot, regardless of subject.
- A simplified colour temperature control that ensures the correct rendition of colours is important.
- A one-profile solution is ideal in order to keep colour management simple.
To satisfy these requirements we produced a new colour look-up-table (LUT). Based on the adjusted colour balance input, we can now render the image with this new LUT.
We have also recalculated the algorithms and taken aspects of CMOS sensitivity and filter characteristics into account to produce the most natural colours possible so that even skin tones show significant benefits from this approach. This process culminated in the creation of the Hasselblad RGB profile. For users with a wider gamut requirement, the Hasselblad L*RGB profile was also created, which incorporates LAB colour data for an extended colour space.
Example: Generic profile
Example: Portrait profile
Example: Product profile
Hasselblad has developed the Hasselblad Natural Colour Solution (HNCS) in order to enable photographers able to achieve colours and subject matter that look perceptually correct. This has been established by the clever combination of math with a series of colour transformations of the original raw RGB colour data coming from the calibrated sensor of a Hasselblad digital camera.
The standard solution for achieving colour that looks “right” is to provide different profiles for different subject matter – one profile for optimum skin tones, another profile for the best rendition of industrial products, and another for the most pleasing reproduction of foliage. Unfortunately, photographic tasks often defy such rigid classifications – how are we to deal with a subject involving both skin and foliage? There may be a generic profile as an alternative to a specific one, but generic profiles are a compromise – there’s nothing completely wrong, but nothing completely right either. With HNCS, Hasselblad strives for a universal profile to relieve the photographer from having to choose between different profiles optimized for different subject matter but without the compromises typical for generic profiles.
At the end of a colour handling, contrast is optimised in order to use most of the available digital levels to represent the important parts of the image in between highlights and dark points. We call this discipline to apply a “film curve,” as film has always used this trick to deliver the best possible contrast. The colours are now finished and described in a device independent of colour space. The final transformation to work with the colours in Photoshop or to output the colours on a specific device is done by simply choosing the Hasselblad RGB input profile in the Phocus imaging software.
Accurate Colour Matching
To summarise, the process of applying different disciplines of mathematics to have basic RGB colour values allows for consistent and proper colour matching to what the human eye sees. With the use of HNCS, the needs for choosing a specific colour profile in order to capture accurate colours correctly is gone. The HNCS colours simply come out correctly.