A Hardware-Friendly Rotation Convolution Neural Network and its FPGA Implementation for Remote Sensing Scene Classification

Directly performing high-accuracy remote sensing scene classification (RSSC) on the edge devices facilitates realtime processing of ground information. Current mainstream RSSC neural networks face challenges in hardware deployment or lack high classification accuracy. This paper proposes a hardware-friendly and high-accuracy multi-branch rotation convolutional neural network (mBR-CNN) and its FPGA accelerator for RSSC tasks. The proposed neural network leverages the rotational properties of $3 \times 3$ convolutional kernels to construct mBR-CNN with rotation invariance. Additionally, the proposed FPGA accelerator innovatively incorporates a multi-branch rotation convolution operator, supported by a multi-branch parallel pipeline structure to achieve optimal throughput. Deployed on the Xilinx Virtex7 XC7VX690T FPGA, it utilizes less than 40 % of on-chip resources, achieving processing speeds exceeding 2000 frames per second (fps), with a state-of-the-art classification accuracy of 96.19% on the UC-Merced Land Use Dataset.