Child Emotion Recognition via Custom Lightweight CNN Architecture

Muhammad Hussain, Hussain Al-Aqrabi

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review

2 Citations (Scopus)


Advancements in e-learning coupled with the recent pandemic has resulted in a paradigm shift when it comes to remote-based education. Whilst this has its benefits the most important being the continuation of the curriculum there are various risks posed to younger learners. Their exposure to a wide variety of content including that with malicious intent poses a serious risk to the emotional well being of children and is a concern for parents. This chapter presents online child emotion recognition framework enabling parents to monitor their childs well-being whilst engaging with online content. The framework is a result of breakthroughs in deep learning for facilitating the development of lightweight convolutional neural networks that can be deployed without the need for specific hardware requiring Graphical Processing Units. The chapter also covers the security concerns presenting an offline rather than a cloud-based inference mechanism.

Original languageEnglish
Title of host publicationKids Cybersecurity Using Computational Intelligence Techniques
EditorsWael M. S. Yafooz, Hussain Al-Aqrabi, Arafat Al-Dhaqm, Abdelhamid Emara
PublisherSpringer, Cham
Number of pages10
Volume1080 SCI
ISBN (Electronic)9783031211997
ISBN (Print)9783031211980, 9783031212017
Publication statusPublished - 21 Feb 2023

Publication series

NameStudies in Computational Intelligence
PublisherSpringer Cham
Volume1080 SCI
ISSN (Print)1860-949X
ISSN (Electronic)1860-9503


Dive into the research topics of 'Child Emotion Recognition via Custom Lightweight CNN Architecture'. Together they form a unique fingerprint.

Cite this