Increasing Image Size (interpolation)

Read five articles on interpolation techniques and you’ll probably get five different recommendations … read the claims from software developers and you’ll find three or four are the ‘very best’. So, despite having ‘up-sized’ hundreds of images you might still be wondering … “have I done this the best possible way?” … “is there a way to get better results?”

Let’s look at the do’s and dont’s and compare the results of the most frequent recommendations, and hopefully put this to bed once and for all (or until the next ‘amazing’ piece of software hits the streets).

Quality
I’m assuming that the objective is a ‘professional’ image (fit for printing in large sizes) rather than something for a screen saver … for instance; Alamy require submissions of 48Mb jpeg files, so if you’re using anything less than an 18 megapixel camera you have to interpolate images – which must then get past Alamy’s pixel-peeping quality control. So, the givens will include that you’re working from a native RAW file, the original is decently sharp and exposed and that there’s a minimum of ‘corrective’ processing required on the image.

Sharpening should always be the last step in the workflow – so interpolate before sharpening. Excessive processing (in sharpening, softening, exposure and noise reduction) can cause ‘banding’ and blocky artifacts to appear after interpolation.

[sample images have no sharpening, in-camera or during processing]

[artifacts caused by any number of 'excessive' adjustments]

 

8 bit or 16 bit
Pretty much a no-brainer really. You’re after the best quality so start out with the most data in the image. 16 bit images will result in better interpolation.

A=original, 100% crop of colour gradient
B=15% increase of 16bit image
C=15% increase of 8 bit image – showing start of artifacts/blockiness
[within limitations of web image]

 

Genuine Fractals or Photoshop’s Bicubic
A seemingly never ending debate as is evident on many forums. So let’s look at some results.

A=original colour gradient, 100% crop
B=35% increase using Genuine Fractals
C=35% increase using Photoshop Bicubic Smoother

A=original with sharp detail, 100% crop
B=25% increase using Genuine Fractals
C=25% increase using Photoshop Bicubic Smoother

You might find one is preferable, but nothing in it to get your knickers in a twist.

 

PS Bicubic Smoother or PS Bicubic Sharper
Some photographers recommend using the Bicubic Sharper option for up-sizing images, suggesting the results are ’sharper’. (Despite that PS describes the two methods as ‘best for enlargement’ and ‘best for reduction’).

A=original, 100% crop
B=25% increase using Photoshop Bicubic Smoother
C=25% increase using Photoshop Bicubic Sharper

Comparing these images, it would appear the Bicubic Sharper option results in a better ‘clarity’ than the Bicubic Smoother method.

Bear in mind however that Bicubic Sharper obviously includes some sharpening of the image (as the name suggests). Obviously therefore, you would want to compare the results of Bicubic Smoother plus sharpening to the results of Bicubic Sharper.

However, as stated earlier, sharpening should be the very last step in the workflow and, if you are supplying the image to a stock agency, sharpening should be left to the buyer – who’s the only one who can decide how much sharpening is required for the printing process to be used.

So, Bicubic Sharper is not recommended for stock images.

 

One step or Stair interpolation
There was a time when many recommended increasing the size of the image in multiple small steps rather than in a large single step. I have not compared results (as I suspect it may have been necessary because of the capabilities of interpolation software at the time). It now seems to be generally accepted that stair interpolation will not produce better results.

 

Interpolation and Printing
Finally, in deciding which process to use for interpolation, you must consider the printing method to be used.

Evidently, the use of a stochastic raster (‘fractal’ or ’shape’ – as used in Genuine Fractals, S-Pline, etc) in the interpolation process is fine for inkjet printing and for photographic prints but a disaster for litho printing. So, if your image is destined for print, and you don’t know which print process, the safer option is Bicubic Smoother.

 

Conclusion
Certainly in terms of images for stock agencies, simply using the Bicubic Smoother in Photoshop for interpolation is as good as it gets … with the added benefit of saving all the cost of 3rd party plug-ins (which is not inconsiderable).

Royalty Free or Rights Managed?

So, you’ve decided you’re going to sell your photographs to the world! Good Luck!

A leading stock photographer was asked in an interview what advice he would give to photographers getting into the stock business. His reply …. “don’t”!

Browse any forum frequented by the professional stock fraternity and you’ll find endless accounts of ‘falling revenues’, ‘tough times’, ‘too much competition’ and of course, the endless debate about micro vs macro agencies ( basically, royalty free or rights managed ).

Let me clear up some confusion about the micro/macro – RF/RM thing.

Micro Stock

• Typically these are royalty free (RF) images.
• The user pays a once-off fee for unlimited use of the image.
• The fee is normally $1 – $5 and your commission is 25c – 60c.
• If RF images feature people, they must have a model release.
• If RF images feature property, they must have a property release.
• RF images cannot feature trademarks and the like.
• Micro stock agencies rely on high volume turnover.
• Typical RF images are small, used in adverts and websites and usually feature products and concepts

Macro Stock

• Typically these are rights managed (RM) images.
• The user pays a fee based on intended use for a specified time and is not entitled to use the image for any other purpose.
• The fee is normally $200 – $600 and your commission is 50% – 80%.
• RM images without model/property releases are restricted to editorial use only.
• RF images for editorial use can feature trademarks and the like.
• Sales volumes of macro stock are much lower than micro stock.
• Typical RM images are large (40Mb+), used in magazines, newspapers, books and themes include travel, landscape, topical / current events etc.

Many image libraries will offer you a choice of RF or RM.

Here’s what I think is the best advice to photographers getting into the stock business …

• Shoot what you enjoy and do best. Do not attempt to produce images that meet an agency’s requirements … it’s been done, too many times, and if it’s not your ‘thing’, it’s been done better!
• Stamp your style on your shots. Every element that makes your work unique improves your sales prospects!
• Find niche themes. Even if your ‘thing’ has been done a zillion times (they all have) make yours ’specialised’ … people – hands instead of faces … landmarks – local instead of international … flowers – indigenous instead of universal …

When you’ve got that clear in your mind go and look at the image libraries and select a couple that fit your profile … not the other way round. If at any stage you find that your images and the library are battling to find common ground – dump them! Use someone else … do not try to become their kind of photographer, it simply doesn’t work (and they’ll have plenty of those anyway) … besides, there’s the dog and tail wagging thing!

Like many photographers, I don’t do what I do to sell it for 25c. So I can’t give you first hand advice on the merits of micro stock sites. There are the obvious … Shutterstock, Dreamstime, iStockphoto, Fotolia, et al … I guess it’s all down to their market share, because at 25c a throw you want volume above all else. Use Alexa to compare trafffic on these sites.

For those considering RM stock libraries, I have individual articles for;
Photographers Direct
Age Fotostock
MyLoupe
Alamy

Equipment Review Sites

Photography, especially the serious kind, is an expensive game. Buying decisions can keep the costs down or make you weep!

I suppose some people might walk into a camera shop and end up buying something the counter sales person recommends … alternatively, instead of nodding repeatedly during that sales pitch, just grab the nearest photo mag, turn to a random page and pick the product in the first advert – 6 of one …

Then again, maybe you’ve decided what sort of photography you’re into and know the basic range of equipment that’s suitable ( if you haven’t yet arrived in this murky place what on earth are you doing looking into product specs! ).

Still here? Ok, so you’ve decided it’s wise to have a look at what others are saying about the equipment you might consider – point & shoot vs DSLR / brand A vs B / this lens or that lens … so many ways to drive yourself crazy!

Here’s the smell of the coffee … they’re not all going to agree, some will tell you nothing, others will boggle your mind with science, some are even paid to tell it the way it isn’t!

So how do you get any help at all from this ocean of info out there?

• Read, read and when that’s done, read some more. You’ll get the general drift and sort the wheat from chaff … at least in terms of pros and cons.
• Short list those sites that convincingly reinforced the general consensus of good or bad.
• Remove the sites from your short list that weren’t quite upfront about the cons. (This will certainly eliminate all the manufacturers sites … great for looking up technical specs but little else.)
• Push the sites you find to be full of gobbledygook to the bottom of the list. That stuff will eventually become meaningful to you and then you will need them.
• Eliminate those sites that don’t have reviewers / authors who appear to be pro’s. That waves good bye to all the ‘reviewed by users’ sites … who are those people??
• Eliminate those sites that review only (or predominantly) single brands … there’s vested interest in them thar sites!
• Find any sites that actually recommended alternatives? Put them at the top of your list!
• Refine your list after reviewing other equipment and in no time at all you’ll have a list of review sites that you have faith in. That’s as good as it gets!

At the end of the day, when you think you’ve found the right gear, borrow, beg or loan it some place and try it out for yourself. That’s essential! It might be great … but if it doesn’t work for you (too heavy, too small etc) what’s the point!

There are a number of excellent review sites out there. The main ones I use (because they suit my equipment range) are photozone , dpreview and dxomark. Together, they provide a great combination of lab science and field usage but most of all, tell it like it is!