Power efficiency for license plate recognition (LPR) under unconstrained scenarios is a crucial factor in many edge-based real-world applications, e.g., autonomous vehicles whose power budget is limited. While a bulk of prior works have explored LPR approaches for unconstrained situations on CPU and GPU servers, these methods result in huge power dissipation, and are ineffective in challenging scenes. In this work, we present a mixed-precision (MP) Transformer hardware architecture to meet the requirements of power efficiency and satisfactory LPR accuracy in unconstrained scenarios, dedicated to implementing power-efficient edge accelerators for difficult LPR tasks. Firstly, MP-LPR Transformer model is proposed, where novel mixed-precision quantization and non-linear approximation techniques are tailored for low bit-width inference. Secondly, vector tiling systolic array Transformer architecture (VTSATA) is proposed with unique vector tiling systolic array (VTSA) and vector ALU (VALU) designs, where VTSA is used to accelerate matrix multiplications, and VALU is used to process non-linear operations. Finally, a FPGA accelerator prototype based on our approach is developed. Experimental results demonstrate that our hardware platform can reduce power consumption more than $20\times$ compared to GPU platforms, and for challenge subset on CCPD dataset, our method can further improve nearly $1\%$ accuracy with respect to the state-of-the-art performance.

The huge computational complexity and memory requirement of deep neural networks (DNNs) impede their deployability on resource-constrained edge-computing devices. To tackle this challenge, numerous model compression solutions have emerged, where Power-of-Two (PoT) quantization scheme demonstrates its potential in reducing computational complexity with complex multiplications replaced by bit-shifting arithmetic operations. However, most of existing methods resort to using full-precision hyper-parameters to avoid significant accuracy degradation, and so they achieve sub-optimal performance on the reduction of computation and memory cost. Therefore, we propose a novel efficient inference approach for resource-limited embedded devices. In our approach, integer-only PoT quantization scheme is designed to quantize both weights and activations into only PoT representations with fewer bit-width to achieve bit-wise inference. Additionally, a distribution-loss regularizer is developed to minimize the retraining disturbation introduced by quantizer. Furthermore, output weights are efficiently encoded in binary format by a two-stage compression pipeline. Finally, comprehensive experiments are conducted with different models and datasets. Experimental results indicate that our approach outperforms state-of-the-art approaches in terms of accuracy and compression rate, with more than 2 × the compression ratio previous methods achieve and 1.2 × accuracy improvement, under 3-bit precision with ResNet56.

As a substitute for metal, carbon foam is vital in the electromagnetic shielding industry. Nevertheless, the diameter and density of carbon foam cells are still mostly assessed manually. This research offers a Deep-Res-MixAttention segmentation method to effectively minimize manual labor and increase measurement efficiency. Moreover, the method consists of two modules: the MixAttention module is intended to improve feature extraction skills, and we use the multiscale deep residual module to collect edge information. In addition to enhancing the segmentation capability of incomplete carbon foam, the loss function is adjusted to address the dataset imbalance issue. Additionally, we propose the bidirectional selection rotation calipers algorithm to intelligently determine the density and diameter. The results reveal that the optimized network’s IoU and acc carbon reach 91.05% and 88.31%. Finally, the calculation errors of the average diameter and density are under control at 1.79% and 7.09%, respectively. The approach has a high application value for assessing the electromagnetic shielding effectiveness of carbon foam.