A Method for Identifying Intersecting Stroke Order Using Fast 3D Raman Spectral Reconstruction

Abstract

The invention discloses a method for identifying a crossstroke order by using a fast three-dimensional Raman spectroscopy reconstruction technology. The method comprises the following steps: S1, constructing a Raman spectroscopy database; S2, selecting a region to be identified from handwriting to be identified, and constructing an XYZ coordinate system by using the plane where the region to be identified as an XY plane; S3, performing white light reconstruction to obtain a plurality of white light photos; S4, uniformly selecting a plurality of coordinate points in the XY plane corresponding to the region to be identified, and sequentially comparing the coordinate points in the white light photos to obtain maximum gray values corresponding to the respective coordinate points, and recordingthree-dimensional coordinates corresponding to the maximum gray value points respectively; S5, obtaining a three-dimensional Raman spectroscopy data set of the region to be identified; S6, performingdata pre-processing on the three-dimensional Raman spectroscopy data set; S7, selecting at least one Raman fingerprint peak from the 3D Raman data set obtained in the previous step, and performing Raman reconstruction to obtain a Raman reconstruction image; and S8, identifying the ink writing order in the handwriting to be identified according to the characteristic peaks in the Raman reconstruction image.

Description

technical field [0001] The invention relates to the field of identification of stroke writing sequence, in particular to a method for identifying the sequence of crossed strokes using fast three-dimensional Raman spectrum reconstruction technology. Background technique [0002] In the field of handwriting identification, when it comes to changing text, adding numbers, tampering and altering, etc., the authenticator usually needs to verify the key cross-changes in the material, so as to determine whether the material is written by a certain person. In addition, judging the handwriting characteristics and writing habits of the writer also needs to study the writing order of the cross strokes. [0003] Traditional handwriting identification methods mostly use two-dimensional plane imaging mode, and the imaging effect is easily affected by factors such as ink component differences, distribution state, stroke continuity, convergence phenomenon, etc., and the layering efficiency i...

AI Technical Summary

Problems solved by technology

The disadvantage of this method is that it is subject to the subjective influence of the appraiser, especially when the phenomenon is not obvious, and the judgment accuracy is low

[0005] (2) The transfer extraction method obtains the edge shape of the ballpoint pen mark by means of transfer, and converts the direct observation of the shape of the pen mark into an indirect observation, which can effectively avoid the subjective influence of the observer. Its defect lies in the transfer of the pen mark High extraction capacity and operating technical requirements

[0009] (1) Study the writing order of cross strokes by detecting the ink components of different layers at the same detection point. However, since the two inks are mixed with each other in most cases, this method is difficult to apply

[0010] (2) Raman surface scanning imaging technology is used to detect cross strokes, and the order of strokes is judged by color matching imaging. However, it is limited to two-dimensional mode, and no accurate conclusion can be drawn for the situation of "network" cross images

Images