Current Status and Performance Analysis of Table Recognition in Document Images With Deep Neural Networks

The first phase of table recognition is to detect the tabular area in a document. Subsequently, structures are recognized second order extract information from respective cells. Table detection and structural pivotal problems domain understanding. However, analysis perplexing task due colossal amount diversity asymmetry tables. Therefore, it an active research document image analysis. Recent advances computing capabilities graphical processing units have enabled deep neural networks outperform traditional state-of-the-art machine learning methods. understanding has substantially benefited recent breakthroughs networks. there not been consolidated description methods for structure recognition. This review paper provides thorough modern methodologies that utilize Moreover, presents comprehensive current related challenges images. leading datasets their intricacies elaborated along with quantitative results. Furthermore, brief overview given regarding promising directions can further improve