Child face detection on front passenger seat through deep learning

Abstract

Objective

One of the main causes of death worldwide among young people are car crashes, and most of these fatalities occur to children who are seated in the front passenger seat and who, at the time of an accident, receive a direct impact from the airbags, which is lethal for children under 13 years of age. The present study seeks to raise awareness of this risk by interior monitoring with a child face detection system that serves to alert the driver that the child should not be sitting in the front passenger seat.

Methods

The system incorporates processing of data collected, elements of deep learning such as transfer learning, fine-tunning and facial detection to identify the presence of children in a robust way, which was achieved by training with a dataset generated from scratch for this specific purpose. The MobileNetV2 architecture was used based on the good performance shown when compared with the Inception architecture for this task; and its low computational cost, which facilitates implementing the final model on a Raspberry Pi 4B.

Results

The resulting image dataset consisted of 102 empty seats, 71 children (0-13 years), and 96 adults (14-75 years). From the data augmentation, there were 2,496 images for adults and 2,310 for children. The classification of faces without sliding window gave a result of 98% accuracy and 100% precision. Finally, using the proposed methodology, it was possible to detect children in the front passenger seat in real time, with a delay of 1 s per decision and sliding window criterion, reaching an accuracy of 100%.

Conclusions

Although our 100% accuracy in an experimental environment is somewhat idealized in that the sensor was not blocked by direct sunlight, nor was it partially or completely covered by dirt or other debris common in vehicles transporting children. The present study showed that is possible the implementation of a robust noninvasive classification system made on Raspberry Pi 4 Model B in any automobile for the detection of a child in the front seat through deep learning methods such as Deep CNN.

Keywords:

• Child safety
• deep learning
• face detection
• passenger seat
• vehicle safety

Acknowledgements

The authors would like to express their sincere appreciation and gratitude for the participation of our research volunteers, and to the universities: Universidad de Monterrey (UDEM) and Universidad Autónoma de Zacatecas (UAZ) for their support and guidance along this project. A.I.T.-C. thanks the Consejo Nacional de Humanidades, Ciencias y Tecnologías (CONAHCYT) (scholarship number 747650) and the Tecnológico de Monterrey, sponsors of his doctoral studies and for which his participation in this article is possible.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Author Contributions

Conceptualization, C.H.-A.; methodology, C.H.-A.; software, C.H.-A. and J.A.A.-S.; validation, J.M.C.-P., A.I.T.-C., and am-T.; formal analysis, C.H.-A., J.M.C.-P., A.I.T.-C, and am-T.; investigation, C.H.-A., J.A.A.-S., and A.I.T.-C.; resources, J.M.C.-P., and am-T.; data acquisition, C.H.-A., and J.A.A.-S.; data curation C.H.-A.; writing-original draft preparation, C.H.-A., J.A.A.-S., and A.I.T.-C.; writing-review and editing, A.I.T.-C, and am-T.; visualization, C.H.-A., J.A.A.-S., and A.I.T.-C.; supervision, J.M.C.-P., and am-T.; project administration, am-T.; funding acquisition, am-T. All authors have read and agreed to the published version of the manuscript.

Informed consent

All participants gave their informed consent prior to participation by submitting their videos and photographs for academic use.

Data availability statement

The generated dataset will be kept as pictures and videos of minors were used for the consolidation of the “minor” class with the informed consent of their legal tutors. The dataset can be used by Universidad de Monterrey (UDEM) for academic research in the topic. If someone wishes to use the dataset, they can send a request to the corresponding author describing the academic purposes of their research.

Additional information

Funding

The data in this work was supported with resources and the use of facilities at the Universidad de Monterrey (UDEM). The funding for the publication of this paper was covered by Universidad de Monterrey (UDEM). Am-T. also thanks CONAHCYT and its program “Sistema Nacional de Investigadoras e Investigadores (SNI)” for the support received as SNI level l (grant number 377932).