PROJECT TITLE :

Compressive Color Pattern Detection Using Partial Orthogonal Circulant Sensing Matrix

ABSTRACT:

To get acceptable signal reconstruction quality with compressive sensing, it's important to create a sensing matrix that's both random enough and has some desirable qualities, such as orthogonality or circulancy. It is common practise to first produce an orthogonal circulant matrix and then pick only a few rows to use as sensing matrices. An orthogonal circulant sensing matrices that generates a circulant matrix where only a specific subset of its rows are orthogonal has been proposed in this research. _ If we do it this way, the generation process is less constrained and we still get the desirable qualities in sensing matrices. In the context of signal reconstruction, the suggested partial shift-orthogonal sensing matrix is put up against random and learned sensing matrices. This sensing matrix is pattern-dependent, hence it is efficient at detecting colour patterns and edges from the measurements of a colour image..