Shredded paper splicing restoration method and system based on semantic error correction and association

The invention relates to a shredded paper splicing restoration method and system based on semantic error correction and association. The method comprises the following steps: S1, preprocessing a shredded paper image; s2, acquiring N shredded paper images, and performing maximum width filling on characters of the ith row of each shredded paper image; s3, according to the line width of the ith line of the shredded paper image, performing vertical centering alignment on the ith line of the shredded paper image; and S4, sorting the N shredded paper images according to the edge features of the ith row between the shredded paper images to obtain a first restored image. According to the method, maximum width filling is carried out on character rows of each shredded paper image, the character rows are aligned, then a correlation degree matrix is established according to edge features between the rows of the shredded paper images, two shredded paper images with the highest shredded paper edge correlation are found out