Human vs machine learning in face recognition - a case study from the travel industry

Aug 3, 2026

Upload by Riley

Research evaluates whether a machine learning face-recognition simulation can substitute for human face-recognition ability under travel-industry-like conditions. Human performance is measured via a survey requiring identification of faces with similar appearance plus hair/makeup disguises, partial facial regions, and dark lighting. Machine learning performance uses HOG features combined with an SVM and is tested on two datasets: Extended Yale B (EYB) for dark-light challenges and the Extended Makeup Face Dataset (EMFD) for makeup disguise. Results show up to 95.4% accuracy for machines in dark lighting and 70.8% with makeup disguises, while humans reach 48% in dark lighting and rise to 94–96% after contrast adjustment but only 36–37% under makeup disguises.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages12

Reading Time30 min

AccessFree to download

Introduction

Travel industry applications of facial recognition

Challenges in real-world deployment

Related work and contrasting methods

Human vs machine learning in face recognition: a case study from the travel industry Regina Lionnie1*, Vidya Hermanto2 1Department of Electrical Engineering, Faculty of Engineering, Universitas Mercu Buana, Indonesia 2PT. Panorama JTB Tours Indonesia , Indonesia | Abstract
This research was conducted to help answer whether a machine learning simulation can replace the human ability to recognize human faces, especially under challenges under travel industry requirements. The human ability to recognize faces was evaluated using a series of questions in a survey. The questions challenged the human respondents to recognize faces under similar looks, with hair and makeup disguises, only part of the facial area, and under dark lighting conditions. At the same time, a histogram of oriented gradient (HoG) combined with a support vector machine (SVM) was built for machine learning simulations. The machine learning was evaluated using two datasets, i.e., the Extended Yale B (EYB) Face dataset for challenge under dark lighting conditions and The Extended Makeup Face Dataset (EMFD) for challenge using face with makeup disguise. The results showed that machine learning simulation of the face recognition system yielded accuracy as high as 95.4% under dark lighting conditions and 70. 8% under facial makeup disguise. On the contrary, only 48% of respondents accurately recognized human faces in dark lighting. The number was increased to 94-96% when the face images were adjusted first with the contrast adjustment method. However, only 36-37% of respondents accurately recognized human faces under face makeup disguise.
This is an open access article under the CC BY-SA license
| Keywords:
Face recognition;
Histogram of oriented gradient; Human vs machine learning; Machine learning;
Travel industry;
Article History:
Received: June 10, 2024
Revised: August 15, 2024
Accepted: September 2, 2024
Published: January 5, 2025
Corresponding Author:
Regina Lionnie
Department of Electrical Engineering, Universitas Mercu Buana, Jakarta, Indonesia Email:
[regina.lionnie@mercubuana.ac.id](regina.lionnie@mercubuana.ac.id) | | --- | --- | INTRODUCTION Facial recognition technology, a type of biometric artificial intelligence, can identify or verify individuals using only their facial biometric data. It typically compares digital facial images to those stored in a database, matching facial features or skin textures. This technology is widely used across different fields; for instance, Facebook employs it to identify faces in digital images, and Apple’s Face ID system authenticates user identity to prevent unauthorized access [1] . Additionally, it is frequently utilized in security , such as smart homes and automation lock systems [2][3] . Facial recognition technology is also increasingly being explored and utilized in travel and tourism. This technology is beneficial as tourism companies must deal with many tourists and customers, so any technology that can help speed up the process will greatly benefit. Additionally, security is a top concern in airportsand hotels , and facial recognition can be used to identify people more quickly, give certain people access to places, and prevent others from entering. Moreover, instantly recognizing faces can also improve customer experience through better personalization. One of the most obvious ways the travel industry uses facial recognition technology is to increase customer personalization. Matching faces in real-world environments with faces in databases, hotels , and other companies can quickly identify people and tailor their services. Hotels, for example, can offer guests the option to provide a photo of themselves during the booking process. When hotel cameras identify their faces upon arrival, hotel staff can greet them by name and use their booking information to ensure they receive specific services. It can also identify guests who have stayed at the hotel

FAQs

What problem does the study address in face recognition for the travel industry?

It examines whether machine learning simulations can replace human face recognition when facing travel-related challenges like dark lighting and disguise conditions.

How is human face-recognition performance evaluated?

Through a survey in which respondents identify faces under similar-looking conditions, including hair/makeup disguises, partial facial areas, and dark lighting.

How does the machine learning system work and what datasets are used?

It uses HOG features with an SVM classifier, evaluated on EYB for dark lighting and EMFD for makeup-disguise challenges.

Aug 3, 2026

Upload by Riley