Rajib Das

Estimating Spatial Variation of River Discharge in Face of Desertification Induced Uncertainty

Climate change and global warming along with wide scale forest degradation have induced desertification in different parts of the world including India. The problem of desertification includes excess runoff, soil erosion, etc., which ultimately leads to catchment degradation. A study was performed to analyze the impact of desertification on river discharge. River Ajay, a small tributary of river Bhagirathi in the west of West Bengal was chosen as the study area due to the semideserted condition of the catchment. DIStributed COupled RATional Model (DISCORAT) where Orange County rational method (Rational OC) and MODified RATional (MODRAT) were coupled to estimate river runoff due to desertification-induced uncertainty. The desertification-induced uncertainty was generated by three scenarios where two scenarios represent extreme desertification (Actual-50%) and semi-desertification (Actual-5%). The input variables were modified according to the generated scenarios and applied to DISCORAT model for estimation of stream flow. As the catchment was divided into 16 15/15 grids and contribution of each grid was included in the estimation, the predicted stream flow for the desertification scenarios would give a distributed variation of stream flow and impact of desertification for each grid could be observed from the estimated stream flow at the grids. Cumulatively, a continuous variation of stream flow due to desertification could be generated and analysis could be made about the impact of desertification on stream flow. According to the results, reduction of stream flow was observed due to desertification and the relationship between desertification and reduction of stream flow was found to be inversely proportional, that is, more intense desertification would imply more reduction of stream flow except in the outlet of the river basin where an opposite relationship was observed between desertification and stream flow. A reason for this estimation could be contributed to the reduction of rainfall as considered in the scenarios of desertification. The reversal of relationship at the outlet could be because of runoff-rainfall ration, which was considered to be well above 150% in Actual-50% scenario of desertification.

Fuzzy evaluated quantum cellular automata approach for watershed image analysis

Fuzzy approaches in a low-level image processing method to partition the homogeneous regions are important challenges in image segmentation. The analysis of the fuzziness in data produces comparable or improved solutions compared with the respective crisp approaches. The novel approach proposed in this chapter has been found to enhance the functionality of the fuzzy rule base and thus enhance the established potentiality of new fuzzy-based segmentation domain with the help of partitioned quantum cellular automata. Image segmentation among overlapping land cover areas on satellite images is a very crucial problem. To detect the belongingness is an important problem for mixed-pixel classification. This new approach to pixel classification is a hybrid method of fuzzy c-means and partitioned quantum cellular automata methods. This new unsupervised method is able to detect clusters using a two-dimensional partitioned cellular automaton model based on fuzzy segmentations. This method detects the overlapping areas in satellite images by analyzing uncertainties from fuzzy set membership parameters. As a discrete, dynamical system, a cellular automaton explores uniformly interconnected cells with states. In the second phase of our method, we use a two-dimensional partitioned quantum cellular automaton to prioritize allocations of mixed pixels among overlapping land cover areas. We tested our method on the Tilaiya Reservoir catchment area of the Barakar River for the first time. The clustered regions are compared with well-known fuzzy C-means and K-means methods and also with the ground truth information. The results show the superiority of our new method.

Automatic mixed pixel detection using a new hybrid Cellular automata approach on satellite image

Mixed-pixels classification in land-cover regions is a challenging task in remote sensing imagery. To classify mixed-pixels, vagueness is always the main characteristic by handling uncertainty. We propose a hybrid approach for pixel classification using Rough sets and Cellular automata models to solve this problem. Multiple belongingness and vagueness among data can be handled efficiently using Rough set theory and is appropriate for detecting arbitrarily-shaped clusters in satellite images. We propose a rough-set based automatic heuristically decision-rule generation algorithm to obtain initial set of clusters. As a discrete, dynamical system, cellular automaton comprises of uniformly interconnected cells with states. In the second phase of our method, we utilize a 2-dimensional cellular automaton to prioritize allocations of mixed pixels among overlapping land cover regions. We experiment our algorithm on Ajoy river catchment area. The segmented regions are compared with well-known FCM and K-Means methods and the ground truth knowledge, which shows superiority of our new approach.

Hybrid Rough-PSO Approach in Remote Sensing Imagery Analysis

Pixel classification among overlapping land cover regions in remote sensing imagery is a very challenging task. Detection of uncertainty and vagueness are always the key features for classifying mixed pixels. This paper proposes an approach for pixel classification using a hybrid approach of rough set theory and particle swarm optimization methods. Rough set theory deals with incompleteness and vagueness among data, which property may be utilized for detecting arbitrarily shaped and sized clusters in satellite images. To enable fast automatic clustering of multispectral remote sensing imagery, in this article, we propose a rough set-based heuristical decision rule generation algorithm. For rough-set-theoretic decision rule generation, each cluster is classified using heuristically searched optimal reducts to overcome overlapping cluster problem. This proposed unsupervised algorithm is able to identify clusters utilizing particle swarm optimization based on rough set generated membership values. This approach addresses the overlapping regions in remote sensing images by uncertainties using rough set generated membership values. Particle swarm optimization is a population-based stochastic optimization technique, inspired from the social behavior of bird flock. Therefore, to predict pixel classification of remote sensing imagery, we propose a particle swarm optimization-based membership correction approach over rough set-based initial decision rule generation. We demonstrate our algorithm for segmenting a LANDSAT image of the catchment area of Ajoy River. The newly developed algorithm is compared with fuzzy C-means and K-means algorithms. The new algorithm generated clustered regions are verified with the available ground truth knowledge. The validity analysis is performed to demonstrate the superior performance of our new algorithms with K-means and fuzzy C-means algorithms.

Tilaiya reservoir catchment segmentation using hybrid soft cellular approach

Image segmentation among overlapping land cover areas in satellite images is a very crucial task. Detection of belongingness is the important problem for classifying mixed pixels. This paper proposes an approach for pixel classification using a hybrid approach of Fuzzy C-Means and Cellular automata methods. This new unsupervised method is able to detect clusters using 2-Dimensional Cellular Automata model based on fuzzy segmentations. This approach detects the overlapping regions in remote sensing images by uncertainties using fuzzy set membership values. As a discrete and dynamical system, cellular automaton explores the uniformly interconnected cells with states. In the second phase of our method, we utilize a 2-dimensional cellular automaton to prioritize allocations of mixed pixels among overlapping land cover areas. We experiment our method on Tilaiya Reservoir Catchment on Barakar river for the first time. The clustered regions are compared with well-known FCM and K-Means methods and also with the ground truth knowledge. The results show the superiority of our new method. (Less)

Shannon entropy based fuzzy distance norm for pixel classification in remote sensing imagery

Pixel classification of mixed pixels in overlapping regions of remote sensing images is a very challenging task. Efficiency and detection of uncertainty are always the key ingredients for this task. This paper proposes an approach for pixel classification using Shannons entropy-based fuzzy distance norm. Unsupervised clustering is used to group the objects based on some similarity or dissimilarity. The proposed algorithm is able to identify clusters comparing fuzzy membership values based on Shannons entropy evaluation. This new normalized definition of the distance also satisfies separability, symmetric and triangular inequality conditions for a distance metric. This approach addresses the overlapping regions in remote sensing images by uncertainties using fuzzy set membership values. Shannon entropy further introduces belongingness and non-belongingness to one cluster within the distance measure. We demonstrate our algorithm for segmenting a LANDSAT image of Shanghai. The newly developed algorithm is compared with FCM and K-Means algorithms. The new algorithm generated clustered regions are verified with on hand ground truth facts. The validity and statistical analysis are carried out to demonstrate the superior performance of our new algorithms with K-Means and FCM algorithms.