Real-time convolutional networks for sonar image classification in low-power embedded systems
This enables autonomous underwater vehicles to perform efficient perception with limited computational resources, though it is incremental as it builds on existing methods like Fire modules.
The paper tackled the problem of real-time sonar image classification on low-power embedded systems by proposing networks that use aggressive max-pooling, achieving 98.8-99.7% accuracy with 41-61 ms inference times on a Raspberry Pi 2, corresponding to speedups of 28.6-19.7.
Deep Neural Networks have impressive classification performance, but this comes at the expense of significant computational resources at inference time. Autonomous Underwater Vehicles use low-power embedded systems for sonar image perception, and cannot execute large neural networks in real-time. We propose the use of max-pooling aggressively, and we demonstrate it with a Fire-based module and a new Tiny module that includes max-pooling in each module. By stacking them we build networks that achieve the same accuracy as bigger ones, while reducing the number of parameters and considerably increasing computational performance. Our networks can classify a 96x96 sonar image with 98.8 - 99.7 accuracy on only 41 to 61 milliseconds on a Raspberry Pi 2, which corresponds to speedups of 28.6 - 19.7.