A Novel Multi-focus Image Capture and Fusion System for Macro Photography - Advances in Image and Graphics Technologies - page 26

Table 1. Quantitative evaluation for the first data set

Standard deviation Information entropy Average gradient

55.4012 4.8904 0.0002767

55.6480 4.9104 0.0003714

55.3374 5.0465 0.0005607

The second data set for evaluation consists of 81 images for a bee. Six images of
them are showed in figure 5. Every image in the data set has different focus region.
The fusion results generated by three methods are shown in figure 6.

Fig. 5. Six images of the second data set

Fig. 6. Fusion results generated by different methods

Table 2 presents a quantitative comparison of various multi-focus fusion methods
for this data set. The evaluation order is figure 6(a), 6(b) and 6(c). These fusion me-
thods consists of average based method, wavelet based method and pyramid based
method in terms of standard deviation, information entropy and average gradient.
From table 2, pyramid based method has maximum values in all three columns, which
indicates that it can generate best fusion result.