Search

In this paper, the authors have analysed the signature of extracting water body from SAR intensity and coherence images with morphological segmentation, and presented a solution to water body mapping in the absence of ancillary information from air or ground. The use of morphological image reconstruction algorithms is a great improvement for mapping water with SAR imagery. Especially, the innovative decision fusion method on integrating intermediate extracted data from intensity and coherence images is the first of its kind in water body mapping. The image reconstruction from fusion marker and mask has shown its promise in removing non-water related features/patterns. A case study is given to demonstrate the efficiency and accuracy of this proposed methodology in removing vegetation cover and man-made features. Thus, the advantage of SAR imagery available unlimited by weather condition can be fully exploited. Furthermore, due to the reduced computing time, this approach for water body mapping can assist authorities to respond promptly to emergencies such as major floods. [ABSTRACT FROM PUBLISHER]

In the content based image retrieval is an important to retrieve all closely matching images. Human brain MR Image retrieval for similar MR Images is an important application in Radiology field of medical diagnostics. Morphological segmentation is a promising method for highlighting and extraction of the region based features of human brain T2 - weighted MR Images. Fuzzy representation of these features and its use in retrieval of brain MRI is demonstrated in our previous work. Unified Fuzzy Matching based fuzzy feature matching is used to assess the image-to-image similarity of MRI in it. The pattern recognition and the identification of similarity of Brain MRI is a complex problem. In addition to the general region based features, additional chain code based features are proposed in this paper. Use of integration of region based features and the chain code based features is demonstrated in this work. This application of Brain MRI retrieval is found to give better retrieval results in terms of precision and average rank. [ABSTRACT FROM PUBLISHER]

Data sparseness is a well-known problem for statistical machine translation (SMT) when morphologically rich and highly inflected languages are involved. This problem become worse in resource-scarce scenarios where sufficient parallel corpora are not available for model training. Recent research has shown that morphological segmentation can be employed on either side of the translation pair to reduce data sparsity. In this work, we consider a highly inflected Indian language as the source-side of the translation pair, Bengali. This paper presents study of morphological segmentation in SMT with a less explored translation pair, Bengali-to-English. We worked with a tiny training set available for this language-pair. We employ a simple suffix-stripping method for lemmatizing inflected Bengali words. We show that our morphological suffix separation process significantly reduces data sparseness. We also show that an SMT model trained on suffix-stripped (source) training data significantly outperforms the state-of-the-art phrase-based SMT (PB-SMT) baseline. [ABSTRACT FROM PUBLISHER]