Training Deep Neural Networks for Wireless Sensor Networks Using Loosely and Weakly Labeled Images

TL;DR

This paper introduces a cost-effective domain generalization method for training deep neural networks on low-quality, domain-shifted images in wireless sensor networks, achieving significant error reduction and enabling real-time recognition.

Contribution

The paper proposes the CEDG algorithm, which effectively transfers models from source to target domains with minimal labor, specifically tailored for resource-constrained WSNs.

Findings

41.12% reduction in category-level averaged error

Network requires only 7 million multiplications per prediction

Enables real-time recognition on digital signal processors

Abstract

Although deep learning has achieved remarkable successes over the past years, few reports have been published about applying deep neural networks to Wireless Sensor Networks (WSNs) for image targets recognition where data, energy, computation resources are limited. In this work, a Cost-Effective Domain Generalization (CEDG) algorithm has been proposed to train an efficient network with minimum labor requirements. CEDG transfers networks from a publicly available source domain to an application-specific target domain through an automatically allocated synthetic domain. The target domain is isolated from parameters tuning and used for model selection and testing only. The target domain is significantly different from the source domain because it has new target categories and is consisted of low-quality images that are out of focus, low in resolution, low in illumination, low in photographing angle. The trained network has about 7M (ResNet-20 is about 41M) multiplications per prediction that is small enough to allow a digital signal processor chip to do real-time recognitions in our WSN. The category-level averaged error on the unseen and unbalanced target domain has been decreased by 41.12%.