Is the analysis of wavelet decomposed sub-bands of a given image called multi-scale analysis? If so, then what about analysis on multiple filtered images of the same image using different filters? Is it also a multi-scale analysis? Similarly, when an image is decomposed into binary bit planes, does the analysis of those planes make it a multi-scale analysis? Please suggest.
It just means data complexity or high conventionality of data or nonlinear behavior of the given database. Remedy is to learn conventionality reduction (PCA) or convolution neural network for data processing or data analysis.
The most famous algorithms using these techniques are SIFT (scale-invariant feature transform) and SURF (Speeded up Robust Features). In both cases; the main idea is to extract the key-points (that are describing the most important features extracted from the same filtered image for different scales). Later, the key-points based matching procedure should be achieved. Building a pyramid of different scales filtered image, represent the key-solution to solve the classical problem of multiscale variant matching.
Is the analysis of wavelet decomposed sub-bands of a given image called multi-scale analysis? If so, then what about analysis on multiple filtered images of the same image using different filters? Is it also a multi-scale analysis? Similarly, when an image is decomposed into binary bit planes, does the analysis of those planes make it a multi-scale analysis? Please suggest.
- Badges
- Science topic
- Similar topics
- Computer Science
- Images