Offline handwriting individual recognition system and method based on three-dimensional dynamic features

An offline handwriting individual recognition system based on three-dimensional dynamic features, comprising:

• a data acquisition module, configured to scan obtained suspicious handwriting to obtain a first white light image and a first three-dimensional image corresponding to the suspicious handwriting, and configured to scan a plurality of obtained sample handwritings to obtain a plurality of second white light images and a plurality of second three-dimensional images corresponding to each of the plurality of sample handwritings;
• a data pre-processing module, connected to the data acquisition module, configured to pre-process the first white light image to obtain a first pre-processed image corresponding to the suspicious handwriting, and configured to pre-process each of the plurality of second white light images to obtain second pre-processed images corresponding to each of the plurality of sample handwritings;
• a skeleton extraction module, connected to the data pre-processing module, configured to extract the first pre-processed image to obtain a first skeleton image corresponding to the suspicious handwriting, and configured to extract each of the plurality of second pre-processed images to obtain second skeleton images corresponding to each of the plurality of sample handwritings;
• wherein a line width of the first skeleton image and of each of the second skeleton images is a single pixel;
• a stroke order recognition module, connected to the skeleton extraction module, configured to perform handwriting tracking according to the first skeleton image to obtain a first writing trajectory corresponding to the suspicious handwriting, and configured to perform handwriting tracking according to each of the second skeleton images to obtain second writing trajectories corresponding to each of the plurality of sample handwritings;
• a first feature extraction module, connected to the data acquisition module and the stroke order recognition module, respectively, configured to extract features from corresponding first white light image according to the first writing trajectory to obtain a first dynamic feature of each pixel in the first writing trajectory, and configured to extract features from each corresponding second white light image according to the each of the second writing trajectories to obtain a second dynamic feature of each pixel in each of the second writing trajectories;
• a second feature extraction module, connected to the data acquisition module and the stroke order recognition module, respectively, configured to extract a first three-dimensional feature of each pixel in the first writing trajectory from a corresponding first three-dimensional image according to the first writing trajectory, and configured to extract a second three-dimensional feature of each pixel in each of the second writing trajectories from each of the corresponding second three-dimensional images according to the each of the second writing trajectories;
• a first data processing module, connected to the first feature extraction module and the second feature extraction module, respectively, configured to obtain a correlation coefficient between the suspicious handwriting and each of the plurality of sample handwritings, according to the first dynamic feature and the first three-dimensional feature corresponding to the suspicious handwriting, and the second dynamic feature and the second three-dimensional feature corresponding to the each of the plurality of sample handwritings;
• and
• a second data processing module, connected to the first data processing module, configured to process the suspicious handwriting, according to a first probability density distribution data and a second probability density distribution data obtained in advance by using statistics and each of the correlation coefficients, to obtain an individual recognition result of the suspicious handwriting.