Object and Texture Classification Using Higher Order Statistics
IEEE Transactions on Pattern Analysis and Machine Intelligence
Accurate Recovery of Three-Dimensional Shape from Image Focus
IEEE Transactions on Pattern Analysis and Machine Intelligence
Digital Image Processing
IEEE Transactions on Pattern Analysis and Machine Intelligence
Multifocus image fusion using artificial neural networks
Pattern Recognition Letters
Analysis and application of autofocusing and three-dimensional shape recovery techniques based on image focus and defocus
Evaluation of focus measures in multi-focus image fusion
Pattern Recognition Letters
Pattern selective image fusion for multi-focus image reconstruction
CAIP'05 Proceedings of the 11th international conference on Computer Analysis of Images and Patterns
Fast algorithms for discrete and continuous wavelet transforms
IEEE Transactions on Information Theory - Part 2
Segmenting a low-depth-of-field image using morphological filters and region merging
IEEE Transactions on Image Processing
3D shape recovery from image focus using kernel regression in eigenspace
Image and Vision Computing
The direct use of curvelets in multifocus fusion
ICIP'09 Proceedings of the 16th IEEE international conference on Image processing
Shape from focus using fast discrete curvelet transform
Pattern Recognition
The complex bidimensional empirical mode decomposition
Signal Processing
Hi-index | 0.00 |
This paper presented a simple algorithm for fusion of multi-focus color images. The algorithm was developed in HSI space where the intensity component was not very sensitive to noise. First, an initial decision map was generated by computing the classical SML (sum-modified-laplacian) as the pixels' focus measure using the intensity component; second, by subtracting one of multi-focus images from another, a coarse edge map was obtained to refine the initial decision map; third, a fast region-filling method was used to build the final decision map; fourth, a soft fusion strategy was applied to the transition zone between focused and defocused regions and a nice-looking fused image was generated. Experiments showed that the algorithm was effective and the results were acceptable.