Table structure recognition based on textblock arrangement andruled line position

摘要

Describes a new method to recognize table structures from documentimages. Each cell of a table is arranged regularly in two dimensions andis represented by a row-column pair. Even in the absence of ruled lines,its coordinates are explicitly found. Thus, it is assumed that anarrangement of textblocks defines the table structure, which is anarrangement of rows and columns, and ruled lines make clear theirrelationship. This process is composed of two procedures: the expansionof cell bounding boxes and the assignment of row-column numbers to eachedge. It is shown that the method can be applied to partially ruledtables with some experimental results