Preprocessing of colour images based on the Principal Components Analysis

PCA can be thought as technique that takes a collection of data and transforms it such that the new data has given statistical properties. The statistical properties are chosen such that the transformation highlights the importance of data elements. Thus, the transformed data can be used for classification by observing important components of the data. Data can also be reduced or compressed by eliminating (filtering out) the less important elements. In this paper we used PCA method to reduce RGB colour images to grey level in the preprocessing step. Two images were tested in the experiment. The first principal component was used as the grey level of the image in the experiment. The results show that the method is valid. The grey level can be used in the further processing and will be widely used in the image processing and computer vision and relatively field.