Image Analysis

is a well documented field of study with many algorithms available to perform specific tasks that can sharpen, smooth and emphasise the original image. These algorithms are widely available on the internet and give accurate and often impressive results. However it is not the case that these image analysis algorithms can simply be copied, modification is often required. Applying these algorithms to real life situations is much more difficult than having a basic understanding of the algorithms and often requires in-depth knowledge and experience.


Image analysis is a form of pattern recognition. Many of the algorithms used in pattern recognition are part of probability theory and involve either Bayesian learning or error minimisation as a means of converging to a correct result. In multivariate space both of these problems suffer from what is known as 'the curse of dimensionality'. The curse of dimensionality occurs when more variables are required to solve a problem and the amount of possible results grow uncontrollably, making the correct result more difficult to find.

To reduce the curse of dimensionality pre-processing of the data is often required. Feature extraction is an effective form of dimension reducing pre-processing and it is especially useful in image Analysis. Image analysis often requires the extraction of features whether it is the features of someone's face or the features of a product going past a production line. Using prior knowledge such as expected position, colour or intensity, areas of interest can be identified and isolated within the image. This is often a multi-step process running from a holistic to granular level increasing the intensity of processing as the area of interest size decreases.

Aplos Systems are experts in the fields of

Image Analysis

, with first degrees and masters in mathematics, algorithms and intelligent systems the team is an excellent choice for an image analysis project. Our capability to modify the algorithms to fit real-life situations and allow for tight time-restraints and high accuracy requirements is supplemented by our ability to provide it in a user friendly interface whether it's on a computer or through a browser.

Aplos systems can apply image analysis algorithms to all fields from face recognition to feature recognition and have a specialism with production line image analysis. We have created a production line system for one of the largest manufacturers of contact lenses, Coopervision. The project had tight time constraints and a need for high accuracy, which were just two of the requirements successfully met by Aplos Systems.