Methods of Extracting Key Handwriting Characteristics for Building a Psychological Profile

Modern methods of extracting key handwriting characteristics to build a psychological profile of a person are considered. The main focus is on the stages of handwriting processing, including data preprocessing, feature extraction using computer vision methods and the use of machine learning models. The geometric, dynamic, and frequency characteristics of handwriting are analyzed, which can be used to interpret the psychological characteristics of the author of the text. Approaches to the development of hybrid algorithms combining convolutional and recurrent neural networks are described, which makes it possible to achieve high accuracy and reliability of analysis. The practical significance of the work lies in the creation of a methodology applicable in psychology, medicine, criminology and education. The results obtained demonstrate the possibilities of automating handwriting analysis and the prospects of using intelligent systems for psychological diagnostics