V. K. Panchal

Hybrid bio-inspired techniques for land cover feature extraction: A remote sensing perspective

Recent advances in the theoretical and practical implementations of biogeography have led to the exploration of new bio-inspired techniques which can prove to be the building blocks of hybrid bio-inspired techniques. This aspect was discovered while considering the exploration of bio-inspired intelligence for developing generic optimization algorithms that can be adapted for performing the given land cover feature extraction task at hand. Certain bio-inspired techniques when integrated with the existing optimization techniques can drastically improve their optimization capability hence leading to better feature extraction. In this paper, we propose a generic architectural framework of a hybrid biologically inspired technique that is characterized by its capability to adapt according to the database of expert knowledge for a more efficient, focused and refined feature extraction. Since our hybrid feature extractor possesses intelligence for selective cluster identification for application of either of the constituent techniques which is in turn based on an inefficiency analysis, we term our classifier as the hybrid bio-inspired pattern analysis based intelligent classifier. Our hybrid classifier combines the strengths of the modified BBO Technique for land cover feature extraction with the Hybrid ACO2/PSO Technique for a more refined land cover feature extraction. The algorithm has been tested for for the remote sensing application of land cover feature extraction where we have applied it to the 7-Band carto-set satellite image of size 472x546 of the Alwar area in Rajasthan and gives far better feature extraction results than the original biogeography based land cover feature extractor [20] and the other soft computing techniques such as ACO, Hybrid PSO-ACO2, Hybrid ACO-BBO Classifier, Fuzzy sets, Rough-Fuzzy Tie up etc.. The 7-band Alwar Image is a benchmark image for testing the performance of a bio-inspired classifier on multi-spectral satellite images since this image is a complete image in the sense that it contains all the land cover features that we need to extract and hence land cover feature extraction results are demonstrated and compared using this image as the standard image.

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.

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.

Resolving mixed pixels by hybridization of biogeography based optimization and ant colony optimization

Recent advances in remote sensing techniques made research possible in those areas where human hands are inaccessible. Digital Imagery brings the virtual image of a desired location, which requires some pre-processing to bring the view to an optimal level. Accuracy level in image classification is assumed on the categorization of the pixel into one of the several land cover classes. When the recognition of pixel accounts for two different classes at the same time, the resulting pixel is categorized as a mixed pixel. This paper proposes a novel approach by clustering the dataset of mixed pixel and thereafter implementing fusion of Ant Colony Optimization (ACO) and Biogeography Based Optimization (BBO) thereby resolving the problem of mixed pixels.

Taxonomy of nature inspired computational intelligence: A remote sensing perspective

The concepts in geospatial sciences are generally vague, ambiguous and imprecise. Also, a combination of spectral, spatial and radiometric resolution of space-borne sensors presents a selective and incomplete look of the geospatial feature/object under its view from the space. Recently, the nature inspired computational intelligence (CI) techniques have emerged as an efficient mechanism to handle diverse uncertainty characteristics. This paper proposes that the human-mind model based computational intelligence techniques, the artificial immune system based computational intelligence techniques; the swarm intelligence based computational intelligence techniques and the emerging geo-sciences based intelligent techniques can be considered as the four pillars of nature inspired CI techniques and hence redefines and extends the taxonomy of nature inspired CI. Researchers have shown keen interest on the applications of natural computing in divergent domains. Scanty references are available on the applications of nature inspired computing in the area of remote sensing. We hence also propose the taxonomy of the most recent nature inspired CI techniques that have been adapted till date for geo-spatial feature extraction and analyze their performances. We also construct a technology timeline of these recent nature inspired CI techniques.

Dynamic Model of Blended Biogeography Based Optimization for Land Cover Feature Extraction

This paper presents a dynamic model of the blended biogeography based optimization (BBO) for land cover feature extraction. In the blended BBO, the habitats represent the candidate problem solutions and the species migration represents the sharing of features (SIVs) between candidate solutions according to the fitness of the habitats which is called their HSI value [9]. However, it is assumed that these SIVs i.e. the number of solution features, remain constant for every habitat [10] and the HSI for each habitat depends only on the immigration and the emigration rates of species [9]. This paper extends the blended BBO by considering the fact that the no. of SIVs or the decision variables may not remain constant for all candidate solutions (habitats) that are part of the Universal habitat. Since the characteristics of each habitat vary greatly hence, comparing all the habitats using the same set of SIVs may be misleading and also may not lead to an optimum solution. Hence, in our dynamic model, we consider the fact that HSI of a solution is affected by factors other than migration of SIVs i.e. solution features, also. These other factors can be modeled as several definitions of HSI of a habitat, each definition based on a different set of SIVs which simulates the effect of these other factors. We demonstrate the performance of the proposed model by running it on the real world problem of land cover feature extraction in a multi-spectral satellite image.

Performance governing factors of biogeography based land cover feature extraction: An analytical study

In recent years, nature inspired remote sensing image classification has become a global research area for acquiring the geo-spatial information from satellite data. 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. This paper is an analytical study and a performance based characterization of the most recent nature inspired image classification technique i.e. Biogeography based Optimization (BBO) that has been used for focused land cover feature extraction [6]. The paper explores the behavior of BBO over different terrain features of a multi-spectral satellite image and establishes the fact that the classification efficiency of BBO for a given land cover feature is proportional to the degree of disorder of the Digital number (DN) values of the pixels comprising that land cover feature when viewed in any of the bands of the multi-spectral satellite image. More precisely, the classification efficiency of BBO on a terrain feature is inversely proportional to the entropy for that feature when viewed in any of the bands of the multi-spectral satellite image. For verification, we calculated the entropies for each of the land cover feature in two bands and found the same results in both the bands, which proves our proposed concept. The dataset on which the proposed concept is demonstrated is the 7-band cartoset satellite image of size 472 × 576 pixels of the Alwar region in Rajasthan. The results indicate that BBO is able to classify the homogeneous regions i.e. the regions with the lower entropy, more efficiently than the regions which show a greater degree of heterogeneity, i.e. higher entropy.

A novel approach based on nature inspired intelligence for face feature extraction and recognition

Feature Extraction and Feature Selection are the most important steps that can affect the performance of a Face Recognition system. In this research we have used a meta-heuristic approach to solve the problem of face recognition. The proposed algorithm is applied to features extracted using discrete wavelets and feature selection is done using Artificial Bee Colony (ABC) optimization where evolution is driven by fitness function defined in terms of correlation value for pattern recognition. Finally Ant Colony optimization (ACO) technique is used for face recognition by measuring the distances between the selected features. Experimental results show that the algorithm was found to generate encouraging recognition results with the minimal set of selected features.

Applying Case Based Reasoning in Cuckoo Search for the Expedition of Groundwater Exploration

A new metaheuristic algorithm named Cuckoo search came up in the recent years. Though lots of metaheuristic algorithms exist but the main advantage of this algorithm is that its search space is extensive in nature. Due to its new arrival it hardly has any footprint in any application. Thus we have adapted this new nature inspired algorithm in our application. The main objective of our application is to find out the potentiality of groundwater in any area as queried by the user. To detect the presence of groundwater, we not only applied Cuckoo search but also case based reasoning. Thus our paper tries to integrate the above mentioned techniques. Our expert provided us with different geographical attributes such as landform, soil, lineament, geology, landuse, and slope. Depending on these attribute values, our proposed algorithm finds out the intensity of groundwater in such areas. Basically the intensity values are narrowed down to high, moderate and low. Thus, once a problem case is given by the user, the case based reasoning uses the K nearest neighbor algorithm to find out the best possible match. After the best possible match is obtained, we apply some propositional logic conditions. The need of propositional logic arises because we have observed a lot of varieties in our case base. Thus to maintain consistency in our output propositional logic is required. Our algorithm achieved 99 % efficiency. Thus we can use our proposed work in any real life problem where groundwater detection is necessary. In our application, cases are basically nests.

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.