Design and Implementation - Segmentation of Hand Bone for Bone Age Assessment - page 53

The proposed MBOBHE can be perceived as simultaneous accomplishments
of three objectives functions using Pareto optimization of bi-modal model [ 1 ].
The aim is to improve the contrast without producing the mean-shift and detail
loss artifacts. In terms of Pareto optimization, this improvement without sacrific-
ing the others is regarded as Pareto-efficiency or Pareto-optimality, whereas the
contrast, brightness mean preservation and details preservations are regarded
as Pareto criteria. MBOBHE is different from the previous MMBEBHE that
performs only a single objective function optimization of bi-modal model as
in [ 2 ]. The main motivation of proposing MBOBHE is to design a HE that can

Input
Radiograph

Bi-Histogram
Equalization

Resultant
Radiograph

Relative Brightness
Difference (RBD)

Relative Contrast
Difference (RCD)

Average Structural
Difference (ASD)

Beta based
Brightness
Preservation Score
(BPS)
Computation

Beta based Optimum
Contrast Score
(OCS) Computation

Beta based Detail
Preservation Score
(DPS) computation

Optimization using
Aggregated Objective
Regularization Function

Result

End

Fig. 3.2 The flowchart of the proposed MBOBHE