Making the most of a point and shoot camera

Making the most of a “Point and Shoot” digital camera.

This article is aimed at those who have a point and shoot camera and want to know the steps to get the best possible quality from it, for high quality printing. Some people have passed their Royal Photographic Society Fellowship with a P&S with a set of small prints, anything is possible, when you know how!

The problem

The sensors in small cameras are tiny. As you can see in this table:

Sensor	Width mm	Height mm	Crop Factor
1/(2.5)”	5.8	4.3
1/(1.8)	7.2	5.3
2/3”	8.8	6.6
4/3”	18	13.5
APSC Canon	22.2	14.8	1.6
APSC Nikon	23.7	15.7	1.5
APSC Canon1	28.7	19.1	1.3
35mm Full Frame	36	24	1

A typical point and shoot uses a 1/(2.5)’ sensor – A full frame sensor is 34x larger – 34x more light! A Canon 450D for example is 12x larger than a Canon G911but about the same price.

Not all pixels are created equal! Bigger = best, but popular perception is that more = best.

The reasons this is a problem,

The image has to be enlarged several times to get to the same size as a 35mm frame, any aberrations are multiplied too.
Most importantly the amount of light falling on the sensor is so much less – proportional to the area of the sensor. This means that the signal must be amplified. Think of this like amplifying a cassette tape – the more the volume is turned up (the higher the ISO) the greater the Tape hiss (digital noise).
To make matters worse, the public demand more and more megapixels (thinking, incorrectly that more = better) More pixels for the same size of sensor means that each individual one is obviously smaller. For example a sensor that has 12 megapixels will have individual photosites half the size of a 6 MP sensor of the same size. The state of the art performance currently is the Nikon D3 it has a full frame sensor and “only” 12 megapixels. You can imagine then how optimistic a 1/(1.8)” sensor with 12 Megapixels is then? A triumph of marketing over engineering!
The dynamic range – or difference between the ability of a camera to distinguish the brightest to the darkest shades – reduces with sensor size.

Some solutions

Choosing a camera and setup

If you buying a low cost Point and Shoot, there are a limited number that are designed for image quality over features. See separate article.
If a more expensive one make sure it has RAW. But these usually cost as much as a basic DSLR, so these only make sense as second cameras to a DSLR main camera.
Don’t be seduced by the megapixel Myth. A 6MP Fuji F31FD from several years ago produces better images than it’s replacement the 12MP Fuji F50, the current model – the Fuji f200EXR is better in bright light, worse in low light. Consequently the F31 is now trading on ebay for 30% more than the price of a new F200EXR. Why? The sensors are the same size but the individual photosites on the F31 are twice as large – so less noise. 12 Megapixels also puts huge emphasis on the quality of the lens – it simply isn’t good enough to gain much extra resolution. Scaling up the 6 megapixel image in photoshop to 12 will yield a better result particularly at 400 ISO and above. But in reality you wouldn’t need to do that – see point 9 below.
If you have any control over parameters like sharpening and contrast, put all these onto their lowest settings. Set the camera to (in order good first RAW, TIFF, JPEG Large-Fine). Almost all P&S cameras do not shoot RAW – but if yours does, use it!

Shooting the Image

Shoot in good light to ensure that you can use low ISO. If you cant control the light use a tripod. I carry a tiny Manfrotto mini pocket tripod.
To cope with the issue of dynamic range, eg. Bright sky / dark foreground, use a graduated filter. Cokin make a kit specially for compacts that hold the filter using a clamp on the tripod bush.
If you have manual controls – or at least exposure compensation, shoot two or three images at different exposures and combine. Several P&S JPEG images can be successfully combined to produce an HDR image using Photoshop or any of the specialist HDR packages like Photomatix. A bargain product I came across to combine several images without complication of HDR, called Bracketeer, cost $20. Combines images well if they are aligned (ie. If you use a tripod)
Get the white balance right (if you have the ability to alter it). Changing white balance for a JPEG afterwards is very destructive and could lead to posterisation.

Editing the Image

Print small images, certainly no larger than A4. I used a Canon 10D 6MP and D30 3MP DSLR for my ARPS Panel, all printed at A4. A P&S cannot hope to match the quality of even a fairly old DSLR camera unless the prints are small. A5 from a P&S should look great. Low ISO shots could look good at A4. A3 is a problem for compacts, despite what the manufacturers say. A3 may well look fine on your wall at home, but compared to a print from a DSLR next to it in competition it will stick out.

Cleaning up

My Boscastle flood pictures were taken with a 4 MP Canon Ixus.

The images have appeared in books and magazines, at up to A4. These were shot as JPEGs, but I have spent time cleaning them up. Here’s how.

Edit images in 16 Bit. Yes I know JPEGs are only 8 but when making alterations eg. Removing a blotchy sky, the interim colours that 8 bit graduations don’t have are needed to smooth them out.
Noise reduction techniques,
1. Colour noise can be removed in Photoshop in 15 seconds by switching from RGB to LAB and blurring the two colour channels.

LAB stands for Luminance, A and B colour channels. In Photoshop>Image>Mode>LAB. By clicking on the “eyes” next to the channels you can switch of the L and leave either A or B. Gaussian or smart blur both A and B. (Don’t worry this doesn’t blur the image!) While you are there sharpen the Luminance channel if your image needs it. You can then switch back to RGB.

Random individual pixels can be eliminated using the Dust and Scratches filter, keep the radius to 1 – 2 or too much detail will be lost.
The best solution is to obtain a dedicated noise reduction plug-in for photoshop such as Neat Image or Define. (But this costs)
JPEG artefacts
1. Strange areas of visible usually black and white compression lines, made worse by sharpening. I tend to remove these using the clone tool to cover them up if they are really noticeable. This can take a while!
2. Chromatic Aberrations – purple Fringing
  1. All cameras suffer from this to a greater or lesser extent. The issue is – will it be noticed in a print? Lightroom and Camera RAW can now open JPEG images and the filters to remove these fringes that appear in high contrast areas can be used. Alternatively it is possible to use the sponge tool to de-saturate the colour fringes.
  2. Blotchy skies
    1. A common JPEG issue. Simply select the sky as accurately as you can feather the edge and use a little Gaussian blur eg. 2 – 4.

As you have probably worked out by now it is MUCH better to shot RAW, so buy a camera with that capability. There are only a few to choose from, so see my article on buying a compact camera. (Coming soon!)

To sum up. You can shoot great pictures with a point and shoot camera, and prepare them for printing with a little preplanning and care over lighting exposure, settings and some useful tricks in photoshop. The most important advice of all though is SMALL IS BEAUTIFUL with regard to prints!