摘要：【中文】本发明公开了一种基于可迁移注意力神经网络的脑电情感识别方法及装置，其中，方法包括：(1)获取一个脑电情感数据库，分为训练集和测试集；(2)建立基于可迁移注意力神经网络的脑电情感识别网络，所述脑电情感识别网络包括特征提取器和情感分类器，所述特征提取器包括依次连接的深度特征提取模块、局部注意力子网和全局注意力子网；(3)网络进行训练，总损失为情感分类器损失加上注意力熵损失后再减去注意力子网和全局注意力子网损失，通过随机梯度下降法更新网络参数；(4)提取待识别的脑电情感数据，将其作为一个测试集样本输入训练好的脑电情感识别网络，并按照步骤(3)对脑电情感识别网络调整，得到识别的情感类别。本发明识别准确率更高。 【EN】The invention discloses an electroencephalogram emotion recognition method and device based on a migratable attention neural network, wherein the method comprises the following steps: (1) acquiring an electroencephalogram emotion database, and dividing the database into a training set and a test set; (2) establishing a brain electric emotion recognition network based on a migratable attention neural network, wherein the brain electric emotion recognition network comprises a feature extractor and an emotion classifier, and the feature extractor comprises a depth feature extraction module, a local attention subnet and a global attention subnet which are sequentially connected; (3) training the network, wherein the total loss is the loss of the emotion classifier plus the loss of the attention entropy, then subtracting the loss of the attention subnet and the loss of the global attention subnet, and updating the network parameters by a random gradient descent method; (4) and (4) extracting electroencephalogram emotion data to be recognized, inputting the electroencephalogram emotion data serving as a test set sample into the trained electroencephalogram emotion recognition network, and adjusting the electroencephalogram emotion recognition network according to the step (3) to obtain the recognized emotion category. The invention has higher identification accuracy.

摘要：【中文】本发明公开了一种基于域选择迁移回归的跨数据库微表情识别方法及装置，包括：(1)获取两个微表情数据库，分别作为训练数据库和测试数据库，其中，每个微表情数据库中包含有若干微表情视频和对应的微表情类别标签；(2)将训练数据库和测试数据库中的微表情视频转换为微表情图像序列，并从中提取出灰度人脸图像，再经过分块后提取人脸局部区域特征；(3)建立域选择迁移回归模型，并采用人脸局部区域特征对其进行学习，得到一个连接人脸局部区域特征与微表情类别标签之间的稀疏投影矩阵；(4)对于待识别的微表情，按照步骤(2)得到人脸局部区域特征，并采用学习到的稀疏投影矩阵，得到对应的微表情类别标签。本发明准确率更高。 【EN】The invention discloses a cross-database micro-expression recognition method and device based on domain selection migration regression, which comprises the following steps: (1) acquiring two micro expression databases which are respectively used as a training database and a testing database, wherein each micro expression database comprises a plurality of micro expression videos and corresponding micro expression category labels; (2) converting the micro expression videos in the training database and the testing database into micro expression image sequences, extracting gray face images from the micro expression image sequences, and extracting local area features of the face after blocking; (3) establishing a domain selection migration regression model, and learning the model by adopting the local facial region characteristics to obtain a sparse projection matrix connecting the local facial region characteristics and the micro-expression class labels; (4) and (3) for the micro expression to be recognized, obtaining the local area characteristics of the human face according to the step (2), and obtaining a corresponding micro expression category label by adopting the learned sparse projection matrix. The invention has higher accuracy.