Result and Discussion - Segmentation of Hand Bone for Bone Age Assessment - page 119

Fig. 4.6 Qualitative visual inspection on the differences between manual rigid adaptive schemes
on an arbitrary TR-2.1
adaptive scheme. To further justify the effect statistically encompassing all the
dataset, quantitative analysis is needed. As mentioned, we adopted the defined
metrics of with specified parameters as described in Sect. 4.3.2 to quantify the pre-
dicted improvement. The results are plotted as shown in Fig. 4.7 .
Figure 4.7 have shown obviously that the resultant segmentation accuracy of the
proposed automated fuzzy quadruple scheme outperforms the other rigid division
schemes in term s of correct labeling of edge pixels which are measured quantita-
tively by FOM . The expected value across all age groups of the proposed scheme
was 0.5721, followed by 0.3271 from rigid scheme of row = 4, column = 4,
0.2129 from rigid scheme of row = 2, column = 2, 0.2614 from rigid scheme of
row = 3, column = 3, and 0.1614 from rigid schem e of row = 1, column = 1. The
overall average of these rigid schemes in terms of FOM was 0.2407. Therefore,
in terms of percentage, if compared to conventional rigid schemes, the proposed
scheme improved by 118.85 %. This tremendous improvement over rigid scheme