本发明公开了一种基于视频的人体行为网络模型及识别方法，包括：3D卷积层、四层Block模块、全局平均池化层以及Softmax激活层。首先，所述3D卷积层，将输入的连续视频帧进行3D卷积操作，然后进入Block模块；所述Block模块主要由改进的3D残差块和SE模块构成，首先利用3D残差块对视频帧进行特征提取，然后经过SE模块，让网络自动获取每个特征通道的重要程度，从而增加重要通道的利用率；最后，将通过全局平均池化层的特征输入到Softmax激活层，输出人体行为类别。本发明的优点是：简单、快捷，且识别精度高，提高了提取时空特征的能力，解决了网络模型的参数较多以及准确率较低的问题。

【 英文摘要 】The invention discloses a video-based human behavior network model and a video-based human behavior network identification method. The video-based human behavior network model comprises a 3D convolution layer, a four-layer Block module, a global average pooling layer and a Softmax activation layer. First, the 3D convolution layer performs a 3D convolution operation on input continuous video frames, and then enters a Block module; The Block module is mainly composed of an improved 3D residual Block and an SE module, firstly, the 3D residual Block is used to perform feature extraction on the video frame, and then the network automatically obtains the importance degree of each feature channel through the SE module, thereby increasing the utilization rate of the important channel; Finally, the features passed through the global average pooling layer are input to the Softmax activation layer, outputting the human behavior class. The method has the advantages of simplicity, rapidness and high identification precision, improves the capability of extracting space-time characteristics, and solves the problems of more parameters and lower accuracy of a network model.

【 用途 】

【 技术功效 】

【技术分类】

【行业分类】