Susanta Chakraborty

Color Image Segmentation by NSGA-II Based ParaOptiMUSIG Activation Function

Based on different criteria any real life problem generates a set of alternative solutions instead of a single optimal solution. Color image segmentation by single objective based parallel optimized MUSIG (ParaOptiMUSIG) activation function may or may not render better solutions for different objective functions. To overcome this problem, a non-dominated sorting genetic algorithm-II (NSGA-II) based ParaOptiMUSIG activation function is proposed in this article to segment color images. Segmentation is achieved using optimized class responses from the image content with a parallel self organizing neural network (PSONN) architecture. Some standard objective functions which are used to assess the quality of the segmented images forms the NSGA-II based image segmentation method.

Redundancy Elimination in Video Summarization

Video summarization is a task which aims at presenting the contents of a video to the user in a succinct manner so as to reduce the retrieval and browsing time. At the same time sufficient coverage of the contents is to be ensured. A trade-off between conciseness and coverage has to be reached as these properties are conflicting to each other. Various feature descriptors have been developed which can be used for redundancy removal in the spatial and temporal domains. This chapter takes an insight into the various strategies for redundancy removal. A method for intra-shot and inter-shot redundancy removal for static video summarization is also presented. High values of precision and recall illustrate the efficacy of the proposed method on a dataset consisting of videos with varied characteristics.

Efficient two phase heuristic routing technique for digital microfluidic biochip

Digital microfluidic biochips (DMFBs) have emerged as an alternative to various in-vitro diagnostic tests and are expected to be closely coupled with cyber physical systems. Efficient-error-free-routing and cross-contamination minimisation are needed during bioassay operations on DMFB. This study proposes a two phase heuristic technique for routing droplets on a two-dimensional DMFB. Initially it attempts to route maximum number of nets in a concurrent fashion depending on the evaluated value of a proposed function named interfering index (IInet). Then exact routing is attempted based on tabulation minimisation process. Remaining nets having interfering index values higher than threshold will be routed considering various constraints in DMFB framework. In second phase another metric named routable ratio (RR) is proposed and depending on RR metric, the routing order among conflicting paths are prioritised to avoid deadlock from there onwards till the droplet reaches its target location. Finally we formulate droplet movement problem as satisfiability problems and solve with SAT based solver engine if higher number of overlapping (≥5) nets exist. Experimental results on benchmark suite I and III show our proposed technique significantly reduces latest arrival time, average assay execution time and number of used cells as compared with earlier methods.

Self-supervised Grey Level Image Segmentation Using Multi-Objective-Based Optimised MUSIG (OptiMUSIG) Activation Function

The multilevel greyscale image can efficiently be segmented by the OptiMUSIG activation function with the help of the multilayer self-organizing neural network (MLSONN) architecture.

Unsupervised Genetic Algorithm Based Automatic Image Segmentation and Data Clustering Technique Validated by Fuzzy Intercluster Hostility Index

The main criteria of the segmentation methods discussed in the previous chapters, is to supply the number of classes previously. So, anyone can not be assured that number of classes generate the best segmented outputs. This chapter tries to overcome that drawback. This approach is quite handsome to derive the exact number of classes from a large number of classes that will give the good segmented/cluster output.

Self-supervised Colour Image Segmentation Using Multiobjective Based Parallel Optimized MUSIG (ParaOptiMUSIG) Activation Function

In Chap. 4, it has been illustrated and proved that the colour images are efficiently segmented by the ParaOptiMUSIG [193, 258, 273] activation function in connection with the parallel self-organizing neural network (PSONN) [195, 196] architecture.

Self-supervised Grey Level Image Segmentation Using an Optimised MUSIG (OptiMUSIG) Activation Function

Different types of image segmentation methods, both supervised and unsupervised, as discussed in Chap. 2, have been applied over the years for the purpose of image segmentation and extraction.

Self-supervised Colour Image Segmentation Using Parallel OptiMUSIG (ParaOptiMUSIG) Activation Function

Colour image segmentation and analysis form a challenging proposition in the image processing arena owing to the nature and variety of data to be processed [258, 259].