Page Segmentation for Content Sequence

This paper describes a technique for separate document images. There are 2 steps in this proposed. Firstly, this paper proposes a method for calculate the skew angle of scanned direction. Next, this paper proposes the technique for page segmentation by using a recursive x-y cut technique. The advantage of this technique is segmented boxes ordered by content sequence. There are useful for the OCR process because the output of OCR process is determined the human reading order. The experimental results shown that this proposed are suitable for 300 dpi resolution and text document images