Image formation and processing methods

Main results of different objects studying using scanning probe microscopy methods are 3-dimensional images of these objects usually. And how well they will be interpreted depends on skillfulness of a researcher. Taking this into account, is important to mention, that still there are some traditional approaches to image formation worth knowing during image analysis.

Scanning probe microscope appeared at the moment when computers were developing intensively. Digital methods of information storage, that was developed for computers, was used in this kind of microscopes to store 3-dimensional images. It gives significantly more convenient image processing, but photographic quality, common for electronic microscopy methods was lost.

Information obtained with scanning probe microscope is represented in a computer as numbers in 2 dimensional matrix. Each number in this matrix, depending on scanning mode, can be tunneling current value, or cantilever deflection value, or a value of a more complex function. If you show this matrix to someone, he hardly can get any meaningful information about the studied surface. So, the first problem arises -- how to transform the numbers to more convenient form. It can be done by following routine.

Numbers in the matrix are in some range, there are maximum and minimum values. This numeric range is possible to put in accordance with some color palette. Thus, every matrix value is represented by point of some colour on a rectangular image. Column and row, that contain this value, becomes point coordinates. As a result we have a picture, on which surface height represented by colour, just like on a map. Usually maps has dozens of colours but our picture has thousands of colours. Points with small height difference usually represented by similar colours for convenience.

It can happen (and usually it happens) that initial data range is wider than number of accessible colours. In this case some data can be lost, and this problem cannot be solved by increasing number of colours due to limitations of human's eye abilities. So, additional data processing is needed, and this processing should be different depending on tasks to be solved. Somebody needs see the whole picture, the other needs to see some details. Different methods are used to achieve this variety in data representation, and the rest of the chapter will describe this methods. Images on which point height represents by a colour will be considered as an example.