Image Pattern Recognition Using Modified Conjugate Directional Filtering (MCDF)
Universiti Sains Malaysia
Faculty of Computing, Health and Science
School of Computer and Information Science
In image processing, some special patterns that are primarily consisted of linear components or linear boundaries are required to be recognised. These patterns sometimes are very weak and hidden in other predominate patterns. Traditionally edge detection approaches are widely used to recognise such patterns. However, the results from those edge detection-based approaches are not always satisfactory. We recently proposed a method named the modified conjugate directional filtering (MCDF) to carry out the enhancement of conjugate linear features. This method is adaptable and can be used in recognising the linear feature associated patterns. In this paper we present four examples (two synthetic models and two real images) in using the MCDF to recognise such patterns in images.