手势识别在人机交互领域具有重要作用。利用肤色检测算法对手势图像进行分割处理，基于轻量级卷积神经网 络MobileNetV2，自建数据集训练，通过迁移学习，建立适用于移动设备的手势识别模型。

Gesture recognition plays an important role in the field of human-computer interaction. Skin color detection algorithm is used for gesture image segmentation. Based on convolutional neural network MobileNetV2, the data training set is established. Through transfer learning, the gesture recognition model suitable for mobile devices is established.