Application image processing to predict personality based on structure of handwriting and signature

Abstract

Handwriting stroke reflects how the author faced his world and the emotional honesty. By examining all elements of handwriting and interpreting them separately or integrated, we could generate a sketch of the writer's character traits, emotional disposition and social style using standard of graphology. As image, the analysis of graphology is divided into two approaches that graphics features and segmentation digit each character. In this research, using graphical approach based on a combination of signature and handwriting to predict the more personality using structure algorithms and multiple artificial neural networks (ANN). The image in A4 paper split into two areas: Signature area which nine features and handwriting based on five features. Each area had pre-processing performed to improve the recognition accuracy. Signature area is classified using ANN based on five features and using multi structure algorithms based on four features. While the handwriting area is classified using multi structure algorithm based on four features (margins, spacing between words and lines, and zone domination) and using ANN after hill valley extraction based on baseline features. Eight features are processed using multi-structure algorithms that provide 87-100% accuracy. In the meantime, six features are classified using an ANN which result an accuracy of 52-100%. It used 100 sets of data testing after training using back propagation with 25-75 data. The system has been implemented with the software so that it can be used for classification of personality from handwriting scanned automatically.

Writer Identification on Multi-Script Handwritten Using Optimum Features

“…This is demonstrated by Figure 2 The proposed method concentrates on the important writing areas that provides the most important information on the basis of the foundation of this study and the theoretical work contributed by the handwriting analysis experts which contends that ending-strokes is deemed to be the most significant handwriting parts as they can contribute in predicting personality traits. Figures 2, 3 support this theory of the handwriting experts [19,20] which serve the main concern of this study. The significant of the proposed method is to characterize the handwriting based on the Ending strokes not like previous studies which are employed all the handwriting text to represent the writer.…”

Revisiting the Quiet-Life Hypothesis in the Banking Sector: Do CEOs’ Personalities Matter?

“…Their findings imply the potential of using machine learning and artificial intelligence to identify the BFPM dimensions of an individual by using different types of handwriting, such as signatures. According to graphology theory, a signature may determine traits, attitudes, qualities, postures, or sentiments (Djamal et al 2013b); thus, the structural signature reflects personality category (Lokhande and Gawali 2017). Le et al (2023) first attempted to classify CEOs' personalities through their signatures in a banking study.…”

“…The psychology literature suggests that personality and value constructs can be accurately and reliably measured by the Big Five personality model (BFPM) (John et al 2008). The five dimensions of the BFPM (extraversion, agreeableness, conscientiousness, neuroticism, and openness to experience) can be predicted through the analysis of human facial images (Jones et al 2012;Hu et al 2017) or graphology analysis (e.g., handwriting and signatures) (Djamal et al 2013a;Kedar and Bormane 2015;Lokhande and Gawali 2017;Maliki and Sidik 2020). Using these new techniques, Wang and Chen (2020) showed that cost efficiency and profitability are positively associated with CEO extraversion, agreeableness, and emotional stability and negatively related to CEO conscientiousness.…”

Human Behavior Identification Based on Graphology Using Artificial Neural Network

“…They collected samples from 100 different writers and used K-NN classification schemes. Djamal et al [5] have extensively worked on signatures. They have proposed a method to identify the writer's behavior by analyzing his or her signature.…”