Lobachev M. V., Purish S. V. / Herald of Advanced Information Technology 2023; Vol.6 No.3: 263–277 ISSN 2663-0176 (Print) ISSN 2663-773123 (Online) Information technology in socio-economic, organizational and technical systems 263 DOI: https://doi.org/10.15276/hait.06.2023.18 UDC 004.93.1 Machine learning models and methods for human gait recognition Mykhaylo V. Lobachev 1) ORCID: 0000-0002-4859-304X; lobachevmv@gmail.com. Scopus Author ID: 36845971100 Sergiy V. Purish 1) ORCID: 0009-0009-0346-842X; spurish@gmail.com 1) Odessa Polytechnic National University, 1, Shevchenko Ave. Odessa, 65044, Ukraine ABSTRACT The paper explores the challenge of human identification through gait recognition within biometric identification systems. It outlines the essential criteria for human biometric features, discusses primary biometric characteristics, and their application in biometric identification systems. The paper also examines the feasibility of utilizing gait as a biometric identifier, emphasizing its advantages, such as not requiring the upfront provision of personal biometric information and specialized equipment. The authors conduct an analysis of existing scientific literature in the field of gait recognition, categorizing gait recognition methods into template-based and non-template-based approaches. Throughout their research, they identify the key issues and challenges that researchers face in this domain, along with the prevailing trends in human gait recognition within biometric identification systems. Additionally, the paper introduces a method for person identification based on gait, utilizing the Histogram of Oriented Gradients and the Sum Variance Haralick texture features. It involves transforming input video into a series of images depicting the gait silhouette, creating a Gait Energy Image (GEI) by combining these gait silhouettes throughout a gait cycle, and translating the GEI into the Gait Gradient Magnitude Image (GGMI). The subsequent step involves extracting recommended gait characteristics from the GGMIs of participants included in a dataset. To preprocess the collected characteristics, Principal Component Analysis (PCA) is applied, reducing the dimensions that may negatively impact classification robustness, thereby enhancing overall performance. In the final step, a K-Nearest Neighbors (KNN) classifier is employed to categorize the characteristics obtained from a specific dataset. The proposed novel feature vector in the paper demonstrates increased reliability and effectively captures spatial variations in gait patterns. Notably, it reduces the dimensionality of the feature vector from 3780×1 to 63×1, resulting in decreased computational complexity in the gait recognition system. Experimental evaluations on the CASIA A and CASIA B datasets reveal that the proposed approach outperforms other HOG-based methods in most scenarios, with the exception of situations involving frontal images. Keywords: Gait recognition; histogram of oriented gradients; haralick texture features; principal component analysis; classification; gait patterns; computer vision For citation:. Lobachev M. V., Purish S. V. “Machine learning models and methods for human gait recognition”. Herald of Advanced Information Technology. 2023; Vol. 6 No. 3: 263–277. DOI: https://doi.org/10.15276/hait.06.2023.18 INTRODUCTION, FORMULATION OF THE PROBLEM Biometry refers to a distinct attribute or personal feature that endures consistently and has the capacity to uniquely identify an individual. This technology finds use in several domains, including but not limited to forensics, access control, workforce monitoring, and shoplifter detection [1]. Historically, biometric modalities such as fingerprint, face, and iris have been widely used [2, 3], [4]. Nevertheless, there are several limitations that restrict its use in certain circumstances. One limitation is that many biometric methods need the active participation and collaboration of the individual being recognized, which may not be ideal © Lobachev M., Purish S., 2023 in surveillance contexts. Furthermore, it is important to note that the extraction of these biometrics is not feasible by remote means, since the information pertaining to these biometrics is either inaccessible or significantly compromised when obtained from a distance [5]. An alternate option arises in the form of gait, which involves a multifaceted array of biomechanical dynamics that are coordinated by the central nervous system and occur entirely at the subconscious level [6]. The research results from neurophysiology and psychology provide support for the notion that gait has distinct characteristics that may be used to differentiate and identify individuals based on their walking patterns [7, 8], [9]. It is noteworthy that individuals possess the capacity to perceive a person whom they are acquainted with from afar, even This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/deed.uk)