Abstract：Face alignment in thermal infrared images is crucial for accurately extracting facial temperature data, as its positioning precision directly impacts the accuracy of temperature analysis in various facial regions. However, most prevalent face alignment algorithms designed for visible face images encounter limitations when directly applied to infrared thermal images, resulting in insufficient accuracy. To address this issue, a face alignment algorithm specifically designed for thermal infrared images, leveraging a multi-scale convolution attention mechanism, is introduced. This algorithm effectively integrates the multi-scale convolutional attention mechanism with an inverted residual convolutional network, while incorporating the wing loss as the loss function to further enhance the network models feature extraction capabilities. On both an open thermal infrared face dataset and a self-collected facial palsy dataset, the algorithm achieves normalized mean errors of 3.23% and 3.94%, respectively. This represents a significant improvement in localization accuracy for facial features compared to traditional methods, extending its applicability to populations with facial palsy. This advancement holds immense potential for various applications.
PU E P, TANG S L. Application of infrared thermography in power system fault diagnosis［J］. Electric Power Technology, 2009 (7): 50-56.
ZHANG Y C, HOU X Y. The Temperature Segregation Area of Asphalt Pavement Detection Method Based on Infrared Thermal Images［J］. Acta Metrologica Sinica, 2017, 38(1): 23-27.
［7］
KHABARLAK K, KORIASHKINA L. Fast facial landmark detection and applications: A survey［J］.
［8］
LIU X, FENG J, ZHANG R, et al. Quantitative assessment of Bells palsy-related facial thermal asymmetry using infrared thermography: A preliminary study［J］. Journal of Thermal Biology, 2021, 100: 103070.
［10］
KOPACZKA M KOLK R, SCHOCK J, et al. A thermal infrared face database with facial landmarks and emotion labels［J］. IEEE Transactions on Instrumentation and Measurement, 2018, 68(5): 1389-1401.
［11］
NAGUMO K, KOBAYASHI T, OIWA K, et al. Face alignment in thermal infrared images using cascaded shape regression［J］. International Journal of Environmental Research and Public Health, 2021, 18(4): 1776.
［13］
HOANG V T, HUANG D S, JO K H. 3-D facial landmarks detection for intelligent video systems［J］. IEEE Transactions on Industrial Informatics, 2020, 17(1): 578-586.
［16］
ZHANG K, ZHANG Z, LI Z, et al. Joint face detection and alignment using multitask cascaded convolutional networks［J］. IEEE Signal Processing Letters, 2016, 23(10): 1499-1503.
［18］
GUO X J, LI S Y, YU J K, et al. PFLD: A practical facial landmark detector［J］. arXiv:2019:1902.10859v2.
ZHENG Z P, ZENG H S, DING C J, et al. Summary on the Infrared Thermal Imaging Temperature Measurement Technology and Its Application［J］. Infrared Technology, 2003, (1): 96-98.
arXiv: 2022: 2101.10808v3.
［19］
DING B. LENet: Lightweight and Efficient LiDAR Semantic Segmentation Using Multi-Scale Convolution Attention［J］. arXiv: 2023: 2301.04275v3.
QU Y, HUAN K W, AN P L, et al. Study on Active Dual-wavelength Unfrared Laser Thermometry Calibration Experiments［J］. Acta Metrologica Sinica, 2021, 42(2): 137-143.
LIU X, WANG Y, LUAN J. Facial paralysis detection in infrared thermal images using asymmetry analysis of temperature and texture features［J］. Diagnostics, 2021, 11(12): 2309.
［6］
NAKANISHI R, IMAI-MATSUMURA K. Facial skin temperature decreases in infants with joyful expression［J］. Infant Behavior and Development, 2008, 31(1): 137-144.
［9］
KAZEMI V, SULLIVAN J. One millisecond face alignment with an ensemble of regression trees［C］// IEEE Computer Society. 2014 IEEE Conference on Computer Vision and Pattern Recognition. Columbus OH, USA, 2014: 1867-1874.
［12］
SUN Y, WANG X, TANG X. Deep convolutional network cascade for facial point detection［C］// IEEE Computer Society. 2013 IEEE Conference on Computer Vision and Pattern Recognition. Portland OR, USA, 2013: 3476-3483.
［14］
POSTER D, HU S, NASRABADI N, et al. An examination of deep-learning based landmark detection methods on thermal face imagery［C］// IEEE Computer Society. 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops. Long Beach, CA, USA, 2019: 980-987.
［15］
KOWALSKI M, NARUNIEC J, TRZCINSKI T. Deep alignment network: A convolutional neural network for robust face alignment［C］// IEEE Computer Society. 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops. Honolulu HI, USA, 2017: 2034-2043.
［17］
CHU W T, LIU Y H. Thermal facial landmark detection by deep multi-task learning［C］// Institute of Electrical and Electronics Engineers Inc. 2019 IEEE 21st International Workshop on Multimedia Signal Processing. Kuala Lumpur, Malaysia, 2019: 1-6.
［20］
FENG Z H, KITTLER J, AWAIS M, et al. Wing loss for robust facial landmark localisation with convolutional neural networks［C］//IEEE Computer Society. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City UT, USA, 2018: 2235-2245.
2024, Vol. 45 
