An Efficient Rain or Snow Removing in a Single Color Image using CNN |
Author(s): |
| R. Mahisha , IFET College of Engineering; K. Bhuvaneshwari, IFET College of Engineering |
Keywords: |
| Convolutional Neural Network, Residual Network, Negative Residual Network, Profound Detail Network |
Abstract |
|
A new profound system architecture for expelling precipitation streaks from singular pictures in light of the profound convolutional neural network (CNN) is proposed. Motivated by the profound residual system that rearranges the learning procedure by changing the mapping structure, a profound detail system to straightforwardly diminish the mapping range from contribution to yield, which influences the figuring out how to process less demanding is proposed. To additionally enhance the de-drizzled result, utilize from the earlier picture area information by concentrating on high recurrence detail amid preparing, which expels foundation impedance and spotlights the model on the structure of rain in pictures. |
Other Details |
|
Paper ID: IJSRDV6I20829 Published in: Volume : 6, Issue : 2 Publication Date: 01/05/2018 Page(s): 2217-2219 |
Article Preview |
|
|
|
|
