A study on edge devices for image classification of the Tasmanian devil (Sarcophilus harrisii) for vaccine delivery

A target-specific bait dispenser is required for oral bait vaccination of the endangered Tasmanian devil (Sarcophilus harrisii) against the deadly devil facial tumour disease. Development of a broadly extendible dispenser would be a beneficial ecological tool. A camera-based edge device with an onboard deep learning model can be used to identify the target species using image classification. It is important to choose a suitable edge device for the dispenser’s effective application in power constrained environments. We evaluated four edge devices for the smart bait dispenser-ArduinoPro Nicla Vision, ArduinoPro Portenta H7 with Vision Shield (LoRa), Raspberry Pi 3 Model B+ and Raspberry Pi Zero 2 W. Two simple convolutional neural networks, and four fine-tuned pretrained models (MobileNetV2, MobileNetV3Small, ResNet50V2, and ResNet152V2) were trained on trail camera images of devil and non-devil species. These models were evaluated across four metrics and post-training quantised for deployment. The edge devices were assessed on inference latency of each model in seconds (s) and power consumption in watt (W). We found that a simple CNN based image classification model yielded the best overall result of ∼96% across all metrics. Raspberry Pi 3 Model B+ and Zero 2 W could run all the six models whereas Portenta failed to run the two ResNet models and Nicla Vision failed to run all four pretrained models. All edge devices were found to be quick enough for dispenser application (average inference latency of 0.718 s). Portenta consumed the lowest power during inference (0.19 W), idle (0.185 W), and light sleep (0.045 W) states whereas Nicla Vision consumed the lowest power during deep sleep (0.002 W). Overall, we found that simple traditional CNNs was best suited for species classification in a camera-based smart bait dispenser for vaccination of Tasmanian devils. We conclude that ArduinoPro Portenta H7 with Vision Shield is the most competent device for this edge AI application. We suggest integration of this system with a microcontroller unit like ATtiny85 to minimise inactive power consumption of the dispenser. This method can be readily adapted for other species in vaccination and supplemental feeding projects.