Samiksha Goel

Biogeography based land cover feature extraction

Recent developments in applied and heuristic optimization methods used for feature extraction from satellite images have been strongly influenced and inspired by natural and biological system. The findings of recent studies are showing strong evidence to the fact that some aspects of the biogeography can be applied to solve specific problems in science and engineering. The algorithm based on this paradigm, Biogeography Based Optimization is investigated in this paper by applying it to detect various features from multi-source satellite images. This algorithm is based on the geographical distribution of biological organisms. It is modeled after the immigration and emigration of species between islands in search of more friendly habitats. The original BBO algorithm does not have the inbuilt property of clustering. However, in this paper we have introduced a modified BBO algorithm. This recent and novel approach is used to make the clusters of different land cover features. The results indicate that highly accurate land-cover features can be extracted effectively when BBO is used, instead of other conventional classifiers.

Cuckoo Search Clustering Algorithm: A novel strategy of biomimicry

A novel, nature inspired, unsupervised classification method, based on the most recent metaheuristic algorithm, stirred by the breeding strategy of the parasitic bird, the cuckoo, is introduced in this paper. The proposed Cuckoo Search Clustering Algorithm (CSCA) yields good results on benchmark dataset. Inspired by the results, the proposed algorithm is validated on two real time remote sensing satellite- image datasets for extraction of the water body, which itself is a quite complex problem. The CSCA makes use of Davies-Bouldin index (DBI) as fitness function. Also a method for generation of new cuckoos used in this algorithm is introduced. The resulting algorithm is conceptually simpler, takes less parameter than other nature inspired algorithms, and, after some parameter tuning, yields very good results.

Extended Biogeography Based Optimization for Natural Terrain Feature Classification from Satellite Remote Sensing Images

Remote sensing image classification in recent years has been a proliferating area of global research for obtaining geo-spatial information from satellite data. In Biogeography Based Optimization (BBO), knowledge sharing between candidate problem solutions or habitats depends on the migration mechanisms of the ecosystem. In this paper an extension to Biogeography Based-Optimization is proposed for image classification by incorporating the non-linear migration model into the evolutionary process. It is observed in recent literature that sinusoidal migration curves better represent the natural migration phenomenon as compared to the existing approach of using linear curves. The motivation of this paper is to apply this realistic migration model in BBO, from the domain of natural computing, for natural terrain features classification. The adopted approach calculates the migration rate using Rank- based fitness criteria. The results indicate that highly accurate land-cover features are extracted using the extended BBO technique.

Development of Swarm Based Hybrid Algorithm for Identification of Natural Terrain Features

Swarm Intelligence techniques facilitate the configuration and collimation of the remarkable ability of a group members to reason and learn in an environment of uncertainty and imprecision from their peers by sharing information. This paper introduces a novel hybrid approach PSO-BBO that is tailored to perform classification. Biogeography-based optimization (BBO) is a recently developed heuristic algorithm, which proves to be a strong entrant in this area with the encouraging and consistent performance. But, as BBO lacks inbuilt property of clustering, it is hybridized with Particle Swarm Optimization (PSO), which is considered as a good clustering technique. We have successfully applied this hybrid algorithm for classifying diversified land cover areas in a multispectral remote sensing satellite image. The results illustrate that the proposed approach is very efficient and highly accurate land cover features can be extracted by using this method. Also, this technique can easily be extended for other global optimization problems.

Novel approaches for classification based on Cuckoo Search Strategy

This paper introduces a novel bio inspired clustering algorithm called Cuckoo Search Clustering Algorithm CSCA. This algorithm is based on the recently proposed Cuckoo Search Optimization technique which mimics the breeding strategy of the parasitic bird-cuckoo. The algorithm is further extended to a classification method, Biogeography Based Cuckoo Search Classification Algorithm BCSCA, which is a hybrid approach of the two nature inspired metaheuristic techniques. The proposed algorithms are validated with real time remote sensing satellite image datasets. The CSCA was first tested with benchmark dataset, which yields good results. Inspired by the results, it was applied on two real time remote sensing satellite image datasets for extraction of the water body, which itself is a quite complex problem. A new method for the generation of new cuckoos has been proposed, which is used in the algorithms. The resulting algorithm is conceptually simpler, takes less parameter than other nature inspired algorithms, and, after some parameter tuning, yields very good results. The extended algorithm BCSCA is also tested on the same satellite image for identifying different land covers by classifying the image in various classes. The algorithm was successful in classifying other land cover regions like rocky, barren, urban and vegetation. We strongly feel that results can be further improved by finer tuning of the parameters. Both the algorithms use Davies-Bouldin index DBI as fitness function. Further exploration of suggested algorithms CSCA and BCSCA may prove them to be strong entrants in the pool of nature inspired techniques.

Association rule analysis using biogeography based optimization

In recent years, data mining has become a global research area for acquiring interesting relationships hidden in large data sets. Data Mining has been used in various application domains such as market basket data, bioinformatics, medical diagnosis, web mining and scientific data analysis. In this paper, we have tried to optimize the rules generated by Association Rule Mining using Biogeography Based Optimization(BBO). BBO has a way of sharing information between solutions depending on the migration mechanisms. The motivation of this paper is to use the feature of BBO for finding more accurate results.

Study on the behaviour of BBO over natural terrain features

In recent years the remote sensing image classification has become a global research area for acquiring the geo-spatial information from satellite data. In this paper we have tried to explore the behavior of Biogeography-Based Optimization (BBO) over different terrain features of a satellite image. The findings of recent studies are showing strong evidence to the fact that various classifiers perform differently when applied to images having different natural terrain features. BBO has a way of sharing information between solutions depending on the migration mechanisms of ecosystem. The motivation of this paper is to use this feature of BBO for finding more accurate results. The approach followed involves making the clusters of different land cover features. The results indicate that highly accurate Homogeneous land-cover features are extracted when BBO is used, instead of other conventional classifiers.

Multiobjective Cuckoo Search for Anticipating the Enemy's Movements in the Battleground

Since ages nations have been trying to improve their military effectiveness by adopting various measures. Having an anticipatory system, which can not only accurately predict the most probable location for the enemys base station but also finds the best route to that point, will lead to improved military operations. This paper aims to propose an integrated framework for developing an efficient anticipatory system. In the first phase of the framework, it proposes Anticipatory Multi objective Cuckoo Search (AMOCS) algorithm to identify the best probable location for deployment of enemy forces. For the second phase a hybrid CS-ACO algorithm is developed for obtaining the most suitable path to the location identified in the first phase. To test the proposed system, satellite image of regions of different terrain types namely plain/desert and mountainous respectively, are chosen. Experimental results demonstrate that the system makes accurate predictions.

A Hybrid Algorithm for Satellite Image Classification

Remote sensing is the most relevant science that permits us to acquire information about the surface of the land, without having actual contact with the area being observed. Amongst the multiple uses of remote sensing, one of the most important has been its use in solving the problem of land cover mapping. Multi spectral classification of remotely sensed data has been widely used to generate thematic Land-Use/Land-Cover maps. Two of the extensively used algorithms for image classification are Self Organizing Feature Maps (SOFM) and Ant Colony Optimization. Although both are bio-inspired optimization techniques, however combining them is a challenging task, especially in the field of remote sensing. In this paper, we have used a Self Organizing Ant Algorithm for Classification of remotely sensed data. Also, we have suggested a new reinforcement factor for the pheromone updation. A test of algorithm is conducted by classifying a high resolution, multi-spectral satellite image of Alwar Region. Results obtained are encouraging.

Performance Analysis of Bio-Inspired Techniques

Increasing popularity of nature inspired meta-heuristics and novel additions in the pool of these techniques at a rapid pace results in a need to categorize and explore these meta-heuristics from different point of views. This paper attempts to compare three broad categories of bio-inspired techniques- Swarm Intelligence methods, Evolutionary techniques and Ecology based approaches via three renowned algorithms, Ant Colony Optimization (ACO), Genetic Algorithm (GA) and Biogeography Based Optimization (BBO) that fall under three categories respectively, based on few varied characteristics. Six benchmark functions are considered for comparison. The paper further suggests a taxonomy of nature inspired methods based on the source of inspiration.