ANALYSIS OF AUTHENTICITY OF HANDWRITTEN SIGNATURES IN DIGITAL DOCUMENTS

Abstract

The recognition of images, such as figures, people, words, characters, handwriting and signatures, made by man is an easy task, but it becomes very difficult to be performed by the machine with the desired success, despite all the technological development in the area of computing. This study aimed to analyze the authenticity of handwritten signatures in digital documents. For this, we adopted a literature review approach, seeking studies published in recent years. The methodology used was the critical analysis and synthesis of information found in academic databases, such as Scielo and Google Scholar, in addition to sites in the area of ​​law and legislation documents. During the research, we identified that the authenticity of signatures is essential to validate documents in different contexts, such as legal, financial and business. We also note that there are different methods of analysis, both online and offline, each with its specific advantages and challenges. The main results indicated that the automatic verification of handwritten signatures in digital documents still presents challenges, mainly due to intra and interindividual variation, quality of the scanned image and context of use. Furthermore, we note the need for advanced techniques such as machine learning and image processing to improve the accuracy and efficiency of these methods. Based on the results obtained, we conclude that the analysis of the authenticity of handwritten signatures in digital documents is a field in constant evolution, demanding more in-depth and empirical studies to validate the effectiveness of existing methods.