DESIGNING OF NEW PATTERN CLASSIFIER BASED ON MORPHOLOGICAL PARAMETER
Tripty Singh · Graduate Research in Engineering and Technology · 2013
Face and text recognition system should be able to automatically detect a face and text in any sample video or images. This involves extraction and analysis of its features. Pattern Classifier system recognizes face and text, regardless of lighting, ageing, occlusion, expression, illumination and pose. Morphological feature based on thresholding of image and gray level components analysis are used for linear discriminant analysis. These are than tested and compared for the template of face and text recognition of facial and textual images database. Present paper discusses designing of new pattern classifier based on morphological parameter. Present research used standard face 95 database, local database, and text databases. The performance of new pattern classifier based on morphological parameter is found to be 100%.Although performance of this classifier is highly dependent on the selection of parameters for thresholding and evaluation.