Abstract
Most of the emphasis of the study for neural networks on impaired face processing in autism has been on triplet-based autism/non-autism infant prediction. Children with ASD are most likely to be examined by their face’s individual components (such as, eyes, nose, and mouth), instead of combining the characteristics into a holistic face memory. Our proposed system identifies multiclass picture recognition by composing a GAN-based neural network and a deep CNN, preceded by L2 regularization and face embedding, thereby accurately predicting autistic children. In Autism infant detection, the Triplet Loss reduces the difference between an anchor and a positive with almost the same identity and significantly increases the variance between the source and a negative with a different label. The way we pick hard related triplets from inside the mini-batches is the key limitation in terms of batch size. We use a batch size of about 1800 examples in most studies. The GTLBFNet loss is based on the FaceNet model, which has training set containing three images of faces. Two of the images are of the same person (one is the Anchor and the other is the Positive), and the third picture is of a different person (Negative). Any representation of an autistic child's face is processed by the FaceNet model, which encodes the features in a 128-dimensional space and outputs a vector of size 128. The GTLBFaceNet model employs a GAN generator to precisely guess the expected outcomes. Our analysis accurately predicts autism dataset labels in 98.3% of cases.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
S. Raj, S. Masood, Analysis and detection of autism spectrum disorder using machine learning techniques. Procedia Comput. Sci. 167, 994–1004 (2020)
Q. Junfei, et al., A survey of machine learning for big data processing. EURASIP J. Adv. Sign. Process. 1(2016), 1–16 (2016)
R.J. Landa, Efficacy of early interventions for infants and young children with, and at risk for, autism spectrum disorders. Int. Rev. Psychiatry 30(1), 25–39 (2018)
D. Bzdok, A. Meyer-Lindenberg, Machine learning for precision psychiatry (2017). arXiv preprint arXiv:1705.10553
J. Lee, et al., Quantifying performance of machine learning methods for neuroimaging data. NeuroImage 199, 351–365 (2019)
S. Efstratios, et al., Predicting the diagnosis of autism spectrum disorder using gene pathway analysis. Mol. Psychiatry 19(4), 504–510 (2014)
J.N. Constantino, T. Charman, Diagnosis of autism spectrum disorder: reconciling the syndrome, its diverse origins, and variation in expression. Lancet Neurol. 15(3), 279–291 (2016)
S.H. Lee, M.J. Maenner, C.M. Heilig, A comparison of machine learning algorithms for the surveillance of autism spectrum disorder. PLoS ONE 14(9), e0222907 (2019)
A. Rachid, et al., Euclidean and geodesic distance between a facial feature points in two-dimensional face recognition system. Hum. Comput. Interact. 1, 5 (2017)
H. Anibal Sólon, et al., Identification of autism spectrum disorder using deep learning and the ABIDE dataset. NeuroImage Clin. 17, 16–23 (2018)
T. Rajat Mani, et al., Classifying autism spectrum disorder using the temporal statistics of resting-state functional MRI data with 3D convolutional neural networks. Front. Psychiatry 11, 440 (2020)
A. Nicola, et al., Deep learning and multiplex networks for accurate modeling of brain age. Front. Aging Neurosci. 11, 115 (2019)
M. Schrimpf, Brain-inspired recurrent neural algorithms for advanced object recognition. Diss. Master’s thesis, Technical University Munich, LMU Munich, University of Augsburg, 2017
A. Hermans, L. Beyer, B. Leibe, In defense of the triplet loss for person re-identification (2017). arXiv preprint arXiv:1703.07737
X. Liu, et al., Hard negative generation for identity-disentangled facial expression recognition. Pattern Recognit. 88, 1–12 (2019)
A. Zhang, et al., EEG data augmentation for emotion recognition with a multiple generator conditional Wasserstein GAN. Complex Intell. Syst. 021, 1–13
S.A. Babu, R.J.S. Raj, A.X. VM, N. Muthukumaran, DCT based enhanced tchebichef moment using huffman encoding algorithm (ETMH), in 2021 3rd International Conference on Intelligent Communication Technologies and Virtual Mobile Networks (ICICV) (2021), pp. 522–527. https://doi.org/10.1109/ICICV50876.2021.9388504
S. Babu Anantha, P. Eswaran, C. Senthil Kumar, Lossless compression algorithm using improved RLC for grayscale image. Arab. J. Sci. Eng. 41(8), 3061–3070 (2016)
S.A. Babu, E. Perumal, Efficient approach of run length coding technique using lossless grayscale image compression (E-RLC), in 2018 3rd International Conference on Inventive Computation Technologies (ICICT) (2018), pp. 680–686. https://doi.org/10.1109/ICICT43934.2018.9034377
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Joshua Samuel Raj, R., Anantha Babu, S., Jegatheesan, A., Arul Xavier, V.M. (2022). A GAN-Based Triplet FaceNet Detection Algorithm Using Deep Face Recognition for Autism Child. In: Peter, J.D., Fernandes, S.L., Alavi, A.H. (eds) Disruptive Technologies for Big Data and Cloud Applications. Lecture Notes in Electrical Engineering, vol 905. Springer, Singapore. https://doi.org/10.1007/978-981-19-2177-3_18
Download citation
DOI: https://doi.org/10.1007/978-981-19-2177-3_18
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-19-2176-6
Online ISBN: 978-981-19-2177-3
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)