School of Optoelectronic Engineering, Changchun University of Science and Technology, Changchun 130022, China.
Key Laboratory of Optoelectronic Measurement, Optical Information Transmission Technology of Ministry of Education, School of Optoelectronic Engineering, Changchun University of Science and Technology, Changchun 130022, China.
Sensors (Basel). 2021 Dec 31;22(1):304. doi: 10.3390/s22010304.
The presence of fake pictures affects the reliability of visible face images under specific circumstances. This paper presents a novel adversarial neural network designed named as the FTSGAN for infrared and visible image fusion and we utilize FTSGAN model to fuse the face image features of infrared and visible image to improve the effect of face recognition. In FTSGAN model design, the Frobenius norm (), total variation norm (), and structural similarity index measure () are employed. The F and TV are used to limit the gray level and the gradient of the image, while the is used to limit the image structure. The FTSGAN fuses infrared and visible face images that contains bio-information for heterogeneous face recognition tasks. Experiments based on the FTSGAN using hundreds of face images demonstrate its excellent performance. The principal component analysis (PCA) and linear discrimination analysis (LDA) are involved in face recognition. The face recognition performance after fusion improved by 1.9% compared to that before fusion, and the final face recognition rate was 94.4%. This proposed method has better quality, faster rate, and is more robust than the methods that only use visible images for face recognition.
虚假图片的存在会影响特定情况下可见光人脸图像的可靠性。本文提出了一种新颖的对抗神经网络，名为 FTSGAN，用于红外和可见光图像融合，并利用 FTSGAN 模型融合红外和可见光图像的人脸图像特征，以提高人脸识别效果。在 FTSGAN 模型设计中，使用了 Frobenius 范数（）、全变差范数（）和结构相似性指数度量（）。F 和 TV 用于限制图像的灰度级和梯度，而则用于限制图像结构。FTSGAN 融合了包含生物信息的红外和可见光人脸图像，用于异构人脸识别任务。基于 FTSGAN 的实验使用了数百张人脸图像，证明了其出色的性能。主成分分析（PCA）和线性判别分析（LDA）用于人脸识别。与融合前相比，融合后的人脸识别性能提高了 1.9%，最终人脸识别率达到 94.4%。与仅使用可见光图像进行人脸识别的方法相比，该方法具有更好的质量、更快的速度和更强的鲁棒性。
