YOLOv4-Tiny-Based Coal Gangue Image Recognition and FPGA Implementation

Abstract

Nowadays, most of the deep learning coal gangue identification methods need to be performed on high-performance CPU or GPU hardware devices, which are inconvenient to use in complex underground coal mine environments due to their high power consumption, huge size, and significant heat generation. Aiming to resolve these problems, this paper proposes a coal gangue identification method based on YOLOv4-tiny and deploys it on the low-power hardware platform FPGA. First, the YOLOv4-tiny model is well trained on the computer platform, and the computation of the model is reduced through the 16-bit fixed-point quantization and the integration of a BN layer and convolution layer. Second, convolution and pooling IP kernels are designed on the FPGA platform to accelerate the computation of convolution and pooling, in which three optimization methods, including input and output channel parallelism, pipeline, and ping-pong operation, are used. Finally, the FPGA hardware system design of the whole algorithm is completed. The experimental results of the self-made coal gangue data set indicate that the precision of the algorithm proposed in this paper for coal gangue recognition on the FPGA platform are slightly lower than those of CPU and GPU, and the mAP value is 96.56%; the recognition speed of each image is 0.376 s, which is between those of CPU and GPU; the hardware power consumption of the FPGA platform is only 2.86 W; and the energy efficiency ratio is 10.42 and 3.47 times that of CPU and GPU, respectively.

A Novel Methodology for Classifying Electrical Disturbances Using Deep Neural Networks

“…Batch normalization helps to reduce the problem of internal covariate drift, which occurs when there is high variation in the input data. This can lead to slower convergence and overfitting [34]. Batch normalization is a powerful technique that improve the performance of deep neural networks [35].…”

Editorial for the Beyond Moore’s Law: Hardware Specialization and Advanced System on Chip

“…Their aim is to support multiple network functions while achieving a high performance at 100 Gbps. In the deep learning field, Xu et al (reference [ 2 ]) propose a low-power design for the YOLOv4-tiny model using an FPGA. Their design utilizes 16-bit fixed-point operators, which trade precision for the achievement of over 10 times and 3 times the power dissipation compared to CPU and GPU, respectively.…”