Environment Recognition and Wearable RGB Cameras: Implications for Autonomous Lower-Limb Biomechatronic Devices

The following RGB image dataset was experimentally collected using a wearable camera system and postprocessed for image classification of different walking environments conducive to controlling autonomous lower-limb biomechatronic devices (e.g., robotic lower-limb prostheses and exoskeletons). Overall, 34,254 sampled images were collected and individually labelled. The authors request that prospective users reference the following publication: Laschowski B, McNally W, Wong A, and McPhee J. (2019). Preliminary Design of an Environment Recognition System for Controlling Robotic Lower-Limb Prostheses and Exoskeletons. IEEE International Conference on Rehabilitation Robotics. Toronto, Canada.