Face recognition algorithms based on deep learning methods have become increasingly popular. Most of these are based on highly precise but complex convolutional neural networks (CNNs), which require significant computing resources and storage, and are difficult to deploy on mobile devices or embedded terminals. In this paper, we propose several methods to improve the algorithms for face recognition based on a lightweight CNN, which is further optimized in terms of the network architecture and training pattern on the basis of MobileFaceNet. Regarding the network architecture, we introduce the Squeeze-and-Excitation (SE) block and propose three improved structures via a channel attention mechanism—the depthwise SE module, the depthwise separable SE module, and the linear SE module—which are able to learn the correlation of information between channels and assign them different weights. In addition, a novel training method for the face recognition task combined with an additive angular margin loss function is proposed that performs the compression and knowledge transfer of the deep network for face recognition. Finally, we obtained high-precision and lightweight face recognition models with fewer parameters and calculations that are more suitable for applications. Through extensive experiments and analysis, we demonstrate the effectiveness of the proposed methods.

Most recent learning algorithms for single image dehazing are designed to train with paired hazy and corresponding ground truth images, typically synthesized images. Real paired datasets can help to improve performance, but are tough to acquire. This paper proposes an unsupervised dehazing algorithm based on GAN to alleviate this issue. An end-to-end network based on GAN architecture is established and fed with unpaired clean and hazy images, signifying that the estimation of atmospheric light and transmission is not required. The proposed network consists of three parts: a generator, a global test discriminator, and a local context discriminator. Moreover, a dark channel prior based attention mechanism is applied to handle inconsistency haze. We conduct experiments on RESIDE datasets. Extensive experiments demonstrated the effectiveness of the proposed approach which outperformed previous state-of-the-art unsupervised methods by a large margin.

A high dynamic range and low-noise CMOS (complementary metal-oxide-semiconductor transistor) front-end interface integrated circuit (IC) with multi-channel detection is presented in this paper. Two different current detection channels, composed of a trans-impedance amplifier (TIA) and an integrator-differentiator TIA, are used to boost the current detection range. A capacitance-coupled instrument amplifier (CCIA) is also included to realize high precision voltage detection. A fourth-order sigma-delta modulator using a second-order loop filter and a second-order noise shaping integral quantizer is adopted to realize effective number of bits above 16 bits. The presented interface IC is implemented in 0.18-μm CMOS process with supply voltage of 3.3 V, and a proto-type electrochemical sensor platform with miniaturized sensor array is developed to verify the functionality of the interface IC. Measurement results indicate that the designed interface IC achieves 157.8 dB current detection dynamic range, and the measured input-referred current noise and voltage noise floor are 1.04 pA and 58.4 nV within 10 kHz integration bandwidths, respectively.