Indian languages are free order word languages where the order of the word does not matter and this makes Indian languages difficult to process. For processing a language, the first step is a morphological analysis of the tokens/words of the language, then Part-of-Speech is assigned using different approaches and finally, a parser is made in the specified language. The parser specifies the syntactic and grammatical relationships. In Hindi and other Indian languages, there is a requirement for two-phase parsing where first local dependencies are handled and then long-distance dependencies are handled. A shallow parser or local word grouper is the first stage of parsing Indian Languages. They divide the sentence into chunks or groups based on their morphological information.  

In this research, one of the tools in “Indian Natural Language Processing”, a local word grouper for Hindi is evolved and analyzed. It includes the formulation of hand-crafted rules for major and minor phrases or groups. The major phrase includes the verb group (VG), noun group (NG), adjective group (ADG), adverb group (AVG), quantifier group (QG), and conjunctive group (CG). The minor phrase includes the postpositions (PPG), symbols (SYG), negations (NG), particles (RPG), and punctuations (PUNG) after finding the major phrase groups. 

Murals are an integral part of our heritage and culture. So there is a need to preserve them for coming generations. The major causes for its deterioration can be odd weather conditions, dust, smoke, etc. It can be affected by cracks or decolorization. Due to the irreversibility of chemical restoration of murals, we can go for digital restoration of murals. The appearance of cracks on paintings deteriorates the perceived image quality. However, one can use digital image processing techniques to detect and eliminate the cracks in digitized paintings.  In this research, we reviewed the techniques of digital restoration of murals including crack detection and filling, and color restoration.

We have designed and implemented an integrated system to achieve the digital restoration of old wall paintings and murals. By leveraging morphological feature detection, graph cuts, inpainting, template matching, and image stitching methods, and by suggesting appropriate algorithms for crack detection, we can satisfactorily restore missing wall painting areas of various sizes and shapes automatically. Integrating local and non-local information yields interesting interpretations of the missing areas while avoiding the overhead of machine learning.

Fuzzy classification of remote sensing images allows the characterization and classification of land covers with improved robustness and accuracy. Coarser resolution images contain mixed pixels as well as non-linearly separable data. The presence of these mixed pixels and non-linear data deteriorates the classification accuracy and computational complexity. Kernels were used for clustering and classification problems based on the similarity between any two samples and these samples are implicitly mapped to a feature space where they are linearly separable. In this research, Kernel-based fuzzy clustering has been used to handle both the problem of non-linearity and mixed pixels. A supervised Kernel-based Fuzzy c-Means classifier has been used to improve the performance of the FCM classification technique. Nine kernel functions are incorporated into the objective function of the FCM classifier. As a result, the effects of different kernel functions can be visualized in generated fraction images. The best single kernels are selected by optimizing the weight constant which controls the degree of fuzziness using an entropy and mean membership difference calculation. These are combined to study the effect of composite kernels which includes both the spatial and spectral properties.