Efficient Identification of Arbitrary Color Filter Array Images Based on the Neural Network Prediction
| Autor: | Jin-Wang Huang, 黃進旺 |
|---|---|
| Rok vydání: | 2016 |
| Druh dokumentu: | 學位論文 ; thesis |
| Popis: | 104 This study presents a back propagation network–based method to identify the input CFA pattern. In this thesis, we can tackle 14 kinds popular CFAs. In the proposed learning-based CFA-recognition method, some statistical estimates and efficient sampling technique are employed. Based on Kodak and Imax test image sets, experimental results on nntool in Matlab show that the recognition accuracy is 100% although the execution time requirement has the room to be improved. |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|