Rapid Deployment of Domain-specific Hyperspectral Image Processors with Application to Autonomous Driving

TL;DR

This paper presents a method for deploying efficient hyperspectral image segmentation networks on low-cost embedded systems for autonomous driving, focusing on model redesign, quantization, and hardware adaptation.

Contribution

It introduces a process to adapt and optimize a lightweight hyperspectral image segmentation network for resource-constrained embedded platforms in autonomous vehicles.

Findings

Successful deployment of a low-power hyperspectral segmentation network on a commercial AI coprocessor.

Quantization techniques maintained segmentation accuracy while reducing computation.

Demonstrated feasibility of real-time hyperspectral processing in autonomous driving systems.

Abstract

The article discusses the use of low cost System-On-Module (SOM) platforms for the implementation of efficient hyperspectral imaging (HSI) processors for application in autonomous driving. The work addresses the challenges of shaping and deploying multiple layer fully convolutional networks (FCN) for low-latency, on-board image semantic segmentation using resource- and power-constrained processing devices. The paper describes in detail the steps followed to redesign and customize a successfully trained HSI segmentation lightweight FCN that was previously tested on a high-end heterogeneous multiprocessing system-on-chip (MPSoC) to accommodate it to the constraints imposed by a low-cost SOM. This SOM features a lower-end but much cheaper MPSoC suitable for the deployment of automatic driving systems (ADS). In particular the article reports the data- and hardware-specific quantization techniques utilized to fit the FCN into a commercial fixed-point programmable AI coprocessor IP, and proposes a full customized post-training quantization scheme to reduce computation and storage costs without compromising segmentation accuracy.