Multi structuring element top-hat transform to detect linear features

Multi structuring element top-hat transform is proposed in this paper to improve the performance of top-hat transform. The desired image feature is treated as a set. And, the set is divided into different subsets. Then, multi structuring elements corresponding to different subsets are constructed. After that, top-hat transform is performed by using the constructed structuring elements, and the results are combined and processed to reconstruct the desired image feature. To verify the efficiency of multi structuring element top-hat transform, the application of linear feature detection is discussed. The analysis and experimental results of multi structuring element top-hat transform show that, because of the efficient performance of multi structuring element top-hat transform, the linear features of different real images in different applications can be efficiently detected. Moreover, multi structuring element top-hat transform can be widely used in different applications.