Tension in Active Shapes
Further Details Contact: A Vinay 9030333433, 08772261612
Email: [email protected] | www.takeoffprojects.com
Abstract
The concept of tension is introduced in the framework of active contours with prior shape
information, and it is used to improve image segmentation. In particular, two properties of this
new quantity are shown: 1) high values of the tension correspond to undesired equilibrium points
of the cost function under minimization and 2) tension decreases if a curve is split into two or
more parts. Based on these ideas, a tree is generated whose nodes are different local minima of
the cost function. Deeper nodes in the tree are expected to correspond to lower values of the cost
function. In this way, the search for the global optimum is reduced to visiting and pruning a
binary tree. The proposed method has been applied to the problem of fish segmentation from low
quality underwater images. Qualitative and quantitative comparison with existing algorithms
based on the Euler–Lagrange diffusion equations shows the superiority of the proposed approach
in avoiding undesired local minima.
Further Details Contact: A Vinay 9030333433, 08772261612
Email: [email protected] | www.takeoffprojects.com
Existing Method
In many cases, as demonstrated by both qualitative and quantitative comparison with existing
techniques, the proposed method succeeds in segmenting the objects of interest from low quality
images, such as those encountered in the industry of fish farming.
Further Details Contact: A Vinay 9030333433, 08772261612
Email: [email protected] | www.takeoffprojects.com
Proposed Method
The proposed duplication algorithm shares some similarities with early split and merge
techniques which attempted to deal with topology changes in active contours. In contrast, the
approach proposed here aims at avoiding undesired local minima of the cost function, while
maintaining topology unchanged.
Merits
Complexity is high.
Demerits
Complexity is low.
Further Details Contact: A Vinay 9030333433, 08772261612
Email: [email protected] | www.takeoffprojects.com
Results
(a)
Example of a non perfect segmentation produced by the proposed technique due to
limitations of the cost function (35). (a) Input image. (b) Crop of the edge map, with many
false edges and missing contours. (c) Desired contours (dotted line), which result in higher
value of the cost function than and output of the proposed algorithm (solid line).
Further Details Contact: A Vinay 9030333433, 08772261612
Email: [email protected] | www.takeoffprojects.com