PROJECT TITLE :
2D Orthogonal Locality Preserving Projection for Image Denoising
Sparse representations using remodel-domain techniques are widely used for higher interpretation of the raw data. Orthogonal locality preserving projection (OLPP) is a linear technique that tries to preserve native structure of data in the transform domain likewise. Vectorized nature of OLPP requires high-dimensional data to be converted to vector format, hence could lose spatial neighborhood information of raw knowledge. On the other hand, processing 2D data directly, not only preserves spatial data, but also improves the computational efficiency significantly. The 2D OLPP is expected to learn the transformation from 2D knowledge itself. This paper derives mathematical foundation for 2D OLPP. The proposed technique is employed for image denoising task. Recent state-of-the-art approaches for image denoising work on 2 major hypotheses, i.e., non-local self-similarity and sparse linear approximations of the info. Locality preserving nature of the proposed approach automatically takes care of self-similarity gift in the image while inferring sparse basis. A international basis is adequate for the complete image. The proposed approach outperforms many state-of-the-art image denoising approaches for grey-scale, color, and texture pictures.
Did you like this research project?
To get this research project Guidelines, Training and Code... Click Here