Outliers are important features that are of special interest to image analysts in their work. The objective of this paper is to show how several statistical techniques with different theoretical foundations can be successfully applied complementarily to detect anomalies in hyperspectral imageries. The methodology is shown in airborne hyperspectral imagery with 60 bands. The visual inspection of the last components of Principal Component Analysis (PCA), together with the analysis of the images provided by the Reed and Xiaoli Yu algorithm and projection pursuit algorithm, allows clear extraction of most of the anomalies, such as synthetic material of tennis court floors or metallic roofs of buildings. A discussion and comparison of the three methods is given.