The requirement for scalable operators in image processing has emerged in recent years as research in the field of computer vision has shown that, typically, a feature in an image may exist significantly over a specific range of scales, with the detected strength of a featuredepending on the scale at which the appropriate feature detection operator is applied. Recent research in computer vision has focussed on the use of range images to provide an almost 3-dimensional description of a scene. Feature-driven segmentation of range images has been primarily used for 3D object recognition, and hence the accuracy of the detected features has become a prominent issue. Feature extraction on range images has proven to be a more complex problem than on intensity images due to both the irregular distribution of range image data and the nature of the features that are present in range images. This paper presents a design procedure for scalable second order derivative operators that can be used directly on irregularly distributed and sparse data through the use of the finite element framework; such operator are hence appropriate for direct use on range image data without the requirement of data pre-processing.
|Title of host publication||Unknown Host Publication|
|Number of pages||8|
|Publication status||Published - 2005|
|Event||The Irish Machine Vision and Image Processing Conference - Queens University Belfast|
Duration: 1 Jan 2005 → …
|Conference||The Irish Machine Vision and Image Processing Conference|
|Period||1/01/05 → …|