Text/non-text classification of connected components in document images

Text segmentation is an important problem in document analysis related applications. We address the problem of classifying connected components of a document image as text or non-text. Inspired from previous works in the literature, besides common size and shape related features extracted from the components, we also consider component images, without and with context information, as inputs of the classifiers. Multi-layer perceptrons and convolutional neural networks are used to classify the components. High precision and recall is obtained with respect to both text and non-text components.