The question of this post is subjective since “Quality” can mean something different to everyone. When looking at clothes for example, quality is easily understood; workmanship, materials, stitching, cut and shape. All items of clothing can be of a good quality but not fashionable (ie good). The same is with photography, image quality helps make a good image, but just because a photograph has good image quality doesn’t mean that it is a good image.
To understand image quality we don’t need to look at the image but what the image is made of. Its pixels. When discussing pixels we can start to judge quality by the images resolution, dynamic range of the pixels and the noise and the sharpness of the pixels
The first measure of quality is the resolution of the image, the number of pixels to the size of an image. If we consider printing as the highest form of a photographic image, we can presume that the minimum resolution need for a good quality image to be 240PPI. Since image resolution is linked to its output size, the quality of the image can be improved through upsampling. The galaxy S3 camera has a resolution of 72PPI, when using Instagram the camera produces 34”x34” size iamges, by changing the resolution to 300PPI the image size shrinks to 8”x8”. This example demonstrates that even low quality images can be made to be a higher quality image but at the sacrifice of the image size.
The dynamic range of the pixels in an image is also important to the quality of the image. Dynamic range is the range of light detail in an image. In a single exposure the dynamic range is measured in stops. For example colour film has a dynamic range of about 5 stops similar to the human eye. A digital SLR has a dynamic range of between 8 and 10 stops with some having as much as 14 which is the same as black and white film.
This is not to say that an accurately exposed image will result in a good quality image because of its dynamic range. Even a well exposed image, due the subject and the scene, may need a wider dynamic range than a single exposure may possess. Taking this into consideration the judgement of quality based on dynamic range is dependent on the image itself. This can also mean that too much dynamic range can also lessen the quality of the image.
A further important quality of an image is sharpness of and noise.
All digital capture technology has some noise as a by-product.
Image noise appears as artefacts, sometimes as an odd colour pixel but also as a grainy effect. The smaller the sensor the more prone it is to capturing noise. Camera manufacturers are in an ISO race at the moment, creating higher and higher ISOs that have less noise. The lower the ISO setting on the camera less noise should be captured. Noise reduces the quality of the image as it distracts the viewer (although for aesthetic reasons noise may actually add an extra level of depth to the reading of an image).
The sharpness of the image can also be an indication of quality. Something in the image needs to be sharp with this usually being the subject. With sharpness we need to balance how sharp the image is with its composition. Focusing can be used creatively to force a viewer’s eye around an image. Certain types of images should have certain elements which are sharp, for example in a portrait should have the eyes really sharp .
The quality of an image can be (and in my opinion) should be measured by these three things. Though just because an image has these things doesn’t mean that it is a good image. Each needs to be balanced and work with an image. Of course at times we may accept an image that could be seen as having less quality than another though actually has more because of the subject matter. This means that the subject of an image is a final determining factor in quality.
For each of these measures of quality there are things we can do to improve the quality of an image, some of them when taking the picture with others being done in post production and some a mixture of both.
We can increase the quality of the resolution through up sampling. When taking a picture with a low resolution camera you can start the process of creating a higher quality. You do this by taking multiple images and in post use a process called stitching to create an image with a better resolution. There is also a process of creating a super resolution image via stacking (Taking images at the same time with the same exposure and in post placing them on top of each other adding more information to the image).
The dynamic range is something that I believe can be slightly widened with a single image due to the dynamic range in the image, though the best way to create an image with a very wide dynamic range is to create a HDR Image. HDR Photography takes several images of different exposures (therefore different dynamic ranges than a single image) and tonally maps the images to create a new single image. HDR is a double edged sword, if done right the image will look brilliant but done wrong a poor messy image is created. HDR is also not great for all subjects and situations.
Focus stacking is one way with macro photography to improve the sharpness and depth of field of an image. This process needs some planning; you take images of the same subject with a slightly different focusing point, processing them together in a stack the sharpness elements are selected from each frame.
You can also use stacking to remove moving objects from a scene as well as reduce the noise of and image as well, these processes need to start when you take the picture and finish in the digital darkroom.
In future posts I am going to be posting my workflows of methods to improve image quality.
Jump to the following posts to read about;
If you enjoyed the post please remember to like and share.
If you wish to get notifications when I post on my blog, you can follow me on Twitter@apertureF64, on Facebook.com/aperturesixtyfour or alternatively be emailed by subscribing below.
All images are the Copyright of Benjamin Rowe , ALL RIGHTS Reserved unless credited to another photographer.
For more information please read my Copyright Statement