Efficient brightness adaptive deep-sea image stitching using biorthogonal multi-wavelet transform and harris algorithm
V Mohana Maniganda Babu, T. Santha · 2017
The main construct behind Deep-Sea image stitching is a combined work of multiple underwater camera captured pictures with overlaying fields of view to produce a high-resolution and panoramic image registration. In image stitching many overlapping pictures are assembled so as to represent one bird's-eye image view. During this research, a replacement feature-based approach is proposed for automatic Deep-Sea image registration and stitching. The projected methodology is enforced on real complicated undersea pictures which rely on four main steps. First, the Harris Algorithm is employed to extract the feature points within the reference and detected pictures. Second, feature matching is applied for treating the geometer difference of the signature vectors are achieved by Biorthogonal multi wavelet Transformation. Third, transformation factors are achieved by treating the least-square rule supported by general wavelet transformation. Finally, the Picture resampling and transformation are executed for the treatment of additive interpolation which induces the image registration and image stitching.