科学家开发出新的手势识别模型,精确度达到96.92%,有助于假肢控制。 Scientists develop new hand gesture recognition model with 96.92% accuracy, aiding prosthetic control.

上海Jiao Tong大学的科学家创建了一种新的模型,用以识别手势,利用先进的神经网络技术将肌肉信号处理成二维图像。 Scientists at Shanghai Jiao Tong University have created a new model for recognizing hand gestures that uses advanced neural network technology to process muscle signals into two-dimensional images. 这种方法被称为CwCST-CNN,大大提高了分类精确度,达到96.92%,超过了现有方法。 This method, called cwCST-CNN, significantly improves classification accuracy to 96.92%, surpassing existing methods. 它在假肢控制和康复训练方面有很有希望的应用。 It has promising applications in prosthetic control and rehabilitation training.