Gesture recognition method under complex background

The invention provides a method and a model for gesture recognition in a complex background. The method comprises the following steps: selecting a gesture picture public data set of the complex background as an experimental data set of the invention; performing random rotation on the image, and enhancing the image in a brightness conversion data enhancement mode; drawing a real frame for the gestures of the data set by using a labmeli tool; an improved gesture recognition model based on the yolov8 is constructed; inputting pictures and labels in the data set into a network model for training; after the training is completed, inputting a picture needing to be detected into the trained yolov8 network model; the model locates a gesture area in the picture and draws a prediction frame; and comparing the content in the prediction frame with the gesture of the real frame in training, and outputting a gesture recognition result. 本发明提供了一种复杂背景进行手势识别的方法，模型，包括：选用复杂背景的手势图片公开数据集作为本发明的实验数据集；对图像进行随机旋转，明亮度转换数据增强方式强化；使用labeli