ST-MNIST

ST-MNIST是一个新颖的神经形态触觉数据集，由新加坡国立大学创建，包含23名参与者手写的0-9数字数据。数据集通过100个仿生事件驱动触觉传感器阵列收集，总计6953个样本。创建过程中，参与者使用笔在一个触觉传感器阵列上书写数字，数据以事件形式异步记录。ST-MNIST数据集旨在为神经形态和机器人研究社区提供资源，推动基于时空事件数据的算法开发，特别是在评估脉冲神经网络方面。

ST-MNIST is a novel neuromorphic tactile dataset developed by the National University of Singapore. It contains handwritten digit data (digits 0–9) from 23 participants. The dataset was collected using a 100-element biomimetic event-driven tactile sensor array, with a total of 6953 samples. During data collection, participants wrote digits on the tactile sensor array with a pen, and the data was recorded asynchronously in event-based form. The ST-MNIST dataset aims to provide a resource for the neuromorphic and robotics research communities, advancing the development of algorithms based on spatio-temporal event data, particularly for the evaluation of spiking neural networks.