Due to additive operation’s dominated computation and simplified network in binary convolutional neural network (BCNN), it is promising for Internet of Things scenarios which demand ultralow power consumption. By means of fully exploiting the in-memory computing advantages and low current consumption design using multilevel cell (MLC) spin-toque transfer magnetic random access memory (STT-MRAM), this paper proposes an MLC-STT-computing in-memory-based computing in-memory architecture to achieve convolutional operation for BCNN to further reduce the power consumption. Simulation results show that compared with the resistive random access memory (RRAM)- and spin orbit torque-STT-MRAM-based counterparts, the architecture proposed in this paper reduces power consumption by ~ $35{\times}$ and 59% in Modified National Institute of Standards and Technology data set, respectively.

Robust and fast traffic sign recognition is very important but difficult for safe driving assistance systems. This study addresses fast and robust traffic sign recognition to enhance driving safety. The proposed method includes three stages. First, a typical Hough transformation is adopted to implement coarse-grained location of the candidate regions of traffic signs. Second, a RIBP (Rotation Invariant Binary Pattern) based feature in the affine and Gaussian space is proposed to reduce the time of traffic sign detection and achieve robust traffic sign detection in terms of scale, rotation, and illumination. Third, the techniques of ANN (Artificial Neutral Network) based feature dimension reduction and classification are designed to reduce the traffic sign recognition time. Compared with the current work, the experimental results in the public datasets show that this work achieves robustness in traffic sign recognition with comparable recognition accuracy and faster processing speed, including training speed and recognition speed.

Non-Maximum Suppression (NMS) is an indispensable post-processing step in face detection. The vast majority of face detection methods need NMS to merge the candidate detected face boxes that belong to the same face. However, the standard NMS is a greedy and local optimization technique which suffers from several shortcomings, such as high complexity (O(N2)), high latency and large power consumption. This paper alleviates these problems and presents an efficient hardware architecture for NMS, meanwhile, carries out the optimization for the calculation unit to achieve the reduction of area accordingly. Based on the multi-thread computing, this paper utilizes sliding window to obtain parallelism and uses Position-based bit table (PBBT) technique for the enhancement of data accessing and data reusing, which greatly decreases the cost of memory access and power consumption. The proposed hardware architecture is implemented in TSMC 28-nm technology. Experiments show that the power consumption is 6.142 mW and the latency is 12.79 μs to cluster 1000 candidate boxes, whose energy efficiency is higher than those state-of-the-art methods by 3798× and 358×, respectively.

How to describe an image accurately with the most useful information but at the same time the least useless information is a basic problem in the recognition field. In this paper, a novel and high precision feature called BG2D2LRP is proposed, accompanied with a corresponding face recognition system. The feature contains both static texture differences and dynamic contour trends. It is based on Gabor and LBP theory, operated by various kinds of transformations such as block, second derivative, direct orientation, layer and finally fusion in a particular way. Seven well-known face databases such as FRGC, AR, FERET and so on are used to evaluate the veracity and robustness of the proposed feature. A maximum improvement of 29.41% is achieved comparing with other methods. Besides, the ROC curve provides a satisfactory figure. Those experimental results strongly demonstrate the feasibility and superiority of the new feature and method.

The integral histogram image plays an important role in accelerating the feature computation in vision algorithms. However, the computational process of the integral histogram, called integral computation, has high computational complexity and numerous memory access operations, which limit its wide application. This brief proposes an in-memory computational architecture based on Spin Transfer Torque Magnetic Random Access Memory (STT-MRAM) to solve these problems. The architecture can work in two different modes depending on the requirements: the integral computation mode and the memory mode. The architecture can figure out the integral histogram when in the integral computation mode, and just store the data directly when in the memory mode. Utilizing the nonvolatile, high density and low power characteristics of STT-MRAM, we integrate the computational units into the memory array to achieve parallel computation. Reduced number of data transmission between storage units and computation units contributes to cut down the latency and energy consumption. The evaluation results show that, comparing with the state-of-the-art work, our architecture provides 1.1x ~ 9.0x performance improvements and reduces 87.4% ~ 97.3% energy consumption for 64 x 64 ~ 512 x 512 size images, just with a 8% area overhead.

In this paper, we propose a multi-modal 2D + 3D face recognition method for a smart city application based on a Wireless Sensor Network (WSN) and various kinds of sensors. Depth maps are exploited for the 3D face representation. As for feature extraction, we propose a new feature called Complete Local Derivative Pattern (CLDP). It adopts the idea of layering and has four layers. In the whole system, we apply CLDP separately on Gabor features extracted from a 2D image and depth map. Then, we obtain two features: CLDP-Gabor and CLDP-Depth. The two features weighted by the corresponding coefficients are combined together in the decision level to compute the total classification distance. At last, the probe face is assigned the identity with the smallest classification distance. Extensive experiments are conducted on three different databases. The results demonstrate the robustness and superiority of the new approach. The experimental results also prove that the proposed multi-modal 2D + 3D method is superior to other multi-modal ones and CLDP performs better than other Local Binary Pattern (LBP) based features.