Variance ratio feature based image fuzzy region detection and fuzzy core calculation method
A technology of fuzzy area and calculation method, applied in the field of computer vision, to achieve the effect of simple program and easy implementation
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment Construction
[0054] Based on the proposed variance ratio feature, the present invention distinguishes blurred areas and clear areas of images by training an SVM classifier, thereby performing blurred area detection. And use the variance ratio to match the fuzzy kernel, and calculate the fuzzy kernel of the fuzzy area. In this process, it mainly includes two parts, namely, fuzzy area classification and fuzzy kernel calculation. The specific implementation method is as follows:
[0055] 1) SVM classifier for fuzzy area detection
[0056] 11) Select a training image set, one part is the image blurred by the blur kernel k, and the other part is the clear image.
[0057] 12) Block the training image and extract its variance ratio feature vector z k .
[0058] 13) Train a linear SVM classifier:
[0059] g k ( z k ) = w k ...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 