An Open-Source Tool for Classification Models in Resource-Constrained Hardware

Sensor applications often face three main restrictions: power consumption, cost, and lack of infrastructure. Most of the challenges imposed by these limitations can be addressed by embedding Machine Learning (ML) classifiers in the sensor hardware, creating smart sensors able to interpret the low-level data stream. However, for this approach to be cost-effective, we need highly efficient classifiers suitable to execute in resource-constrained hardware, such as low-power microcontrollers. In this paper, we present an open-source tool named EmbML - Embedded Machine Learning that implements a pipeline to develop classifiers for resource-constrained hardware. We describe EmbML implementation details and comprehensively analyze its classifiers considering accuracy, classification time, and memory usage. Moreover, we compare the performance of EmbML classifiers with classifiers produced by related tools to demonstrate that our tool provides a diverse set of classification algorithms that are both compact and accurate. Finally, we validate EmbML classifiers to recognize disease vector mosquitoes in a smart sensor and trap application.