Adjustment method of a machine-readable form model and a filled form scanned image thereof in the presence of distortion

An adjustment method of a machine-readable form model and a filled form scanned image thereof in the presence of distortion, comprising

• form of model in electronic form describing spatial correlation and parametrical properties of the form objects free of distortion;
• form of image in electronic form obtained by means of optical input from a paper medium;
• said form of model comprising at least
• reference points spatial characteristics information,
• data input fields' spatial or parametrical characteristics information,
• other supplemental information;
• said form of image in the form of bit-mapped image, comprising at least
• reference points images,
• data input fields images,
• inputted data images into the input fields,
• distortion, caused by an external devices, on form transferring from electronic into paper form or vice versa;
• said method further comprising the following steps:
• assigning either form of image or form of model as changeable form, that would be subjected to changes, with the simultaneous assignment of the other said form as non-changeable, not subjected to changes,
• marking out on the non-changeable form one or more non-overlapped regions, disposed within the distortion expansion, the size of the said region to be greater then the covered distortion area,
• marking on the changeable form regions corresponding to those on the non-changeable form,
• dividing of all possible set of objects, that can be met on the form, into several types differing in reliability level of identification,
• assigning each object image, present on the form of image, a reliability level of identification in accordance with its type,
• for each selected region, and for each type of object, beginning from the highest level of identification reliability, step by step decreasing the said level to the lowest one, performing the following steps:
• setting the correspondence between identical objects of image form and model form,
• computing a partial distortion correction factors of the spatial parameters for each object present on the changeable form,
• computing a consolidated distortion correction factor of the spatial parameters, on the base of the said partial distortion correction factors for each object, taking into account their identification reliability levels,
• spatial parameters correction for objects of changeable form, taking into account said consolidated factor;
• performing the subsequent form image processing.