Categorization And Translation Operating System’s Assistance in Explication of Different Bangladeshi Accents (Published)
National language of Bangladesh is Bengali and it’s also the official language used frequently. Our paper’s focal point was to categorize and differentiate West Bangla language or Bangladeshi Bangla accent in a Bengali sentence. We first amassed text from literature files. Then converted text sentence data to numeric data by using TF-IDF. After PCA application by MATLAB, final data set was being obtained. Our strategy for future will assist in developing an automatic software that detects if a sentence has been written in West Bangla or Bangladeshi Bangla and then it will do translation from one to another form. Differences between both Bangladeshi accents is already so minimum that only native speaker can identify them distinctively. There was no data available previously for this study. This work denoted that as if languages seems to be same but are unique and different in their own way and depicts the identity of two geographically separated regions. The major output of this work paid heed on identification of the form of language frequently used today. Many other studies could be conducted, based on the results of our study, on the effects of Sanskrit and Foreign literature
Keywords: Bangladeshi Bangla, Inverse Data Frequency, Linear SVM, Principal Component Analysis, Python, Term Frequency, West Bangla
A Novel Method Of Average Filtering For Removing Noise And Face Recognition (Published)
Face recognition is new and difficult which requires great effort and determination due to the Wide variety of faces, complexity of noises and image backgrounds. In this paper, we propose an Average Filtering based novel method for face recognition in cluttered and noisy images. It is imperative that computational researchers know of the key findings from experimental studies of face recognition by human. These findings provide insights into the nature of starting symbol to begin that the human visual system relies upon for achieving its great deal of performance and serve as the building blocks for efforts to artificially emulate these abilities. In this paper, we are presenting what we believe are various basic results, with implications for the computational design systems. The aim of our proposed work of average filtering based method for face recognition is to improve the recognition accuracy. We use AT&T face database and experiments on it are performed to demonstrate the effectiveness of the proposed method.
Keywords: Average Filter, Eigenfaces, Face Recognition, Feature Extraction, Fisherfaces., Laplacianfaces, Linear Discriminant Analysis, Principal Component Analysis, Smooth Mean Filter