Abstract
Face image super-resolution imaging is an important technology which can be utilized in crime scene investigations and public security. Modern CNN-based super-resolution produces excellent results in terms of peak signal-to-noise ratio and the structural similarity index (SSIM). However, perceptual quality is generally poor, and the details of the facial features are lost. To overcome this problem, we propose a novel deep neural network to predict the super-resolution wavelet coefficients in order to obtain clearer facial images. Firstly, this paper uses prior knowledge of face images to manually emphases relevant facial features with more attention. Then, a linear low-rank convolution in the network is used. Finally, image edge features from canny detector are applied to enhance super-resolution images during training. The experimental results show that the proposed method can achieve competitive PSNR and SSIM and produces images with much higher perceptual quality.
Highlights
Face Super-Resolution (SR) is an important subset of image super-resolution technology for public security
It is worth noting that super-resolution method based on Convolutional Neural Network (CNN) can achieve good performance in terms of peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM) while the output images are often over smooth resulting in poorer perceptual quality
The purpose of the first phase is to focus the attention on the facial features and enable translation invariance to the wavelet-based CNN. (2) We introduce the method of [27] to supplement the image edge with information extracted by the canny edge detection operator
Summary
Face Super-Resolution (SR) is an important subset of image super-resolution technology for public security. Most SR method based on deep learning initially used interpolation of the low-resolution image for a high-resolution image first before it is computed in the neural network, which incurred higher computational cost. It is worth noting that super-resolution method based on CNN can achieve good performance in terms of peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM) while the output images are often over smooth resulting in poorer perceptual quality. Wavelet-based method performs wavelet transform on the high-resolution image to obtain wavelet sub-band coefficients. With further development of deep learning, many methods based on estimating the high frequency coefficients have been proposed in [20,21,22,23,24,25]. In [22], a deep neural network model which combines wavelet transform and CNN is proposed to predict missing details in the wavelet coefficients of low-resolution images.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
