一种面向表情识别的ROI区域二级投票机制

Subjects: Computer Science >> Integration Theory of Computer Science submitted time 2018-05-24 Cooperative journals: 《计算机应用研究》

文元美 欧阳文 凌永权

Abstract： Aiming at the problem of how to more efficiently use the distributed features that convolutional neural network have learned from training images, this paper proposed a ROI(regions of interest) region secondary voting mechanism for facial expression recognition. Firstly, it divided into the image a series of ROI images, and input it into the convolutional neural network for training. Then, it input into the ROI images of the test image the convolutional neural network, getting all ROI images’ results. Lastly, it used the secondary voting mechanism to determine the final category of test image. In addition, aiming at the problem of convolutional neural network cannot learn spatial position information such as rotation, this paper introduced the STN (spatial transformer network) to make convolutional neural network useful in complex condition. Experiments show the ROI region secondary voting mechanism can more effectively use the distributed features which learned by convolutional neural network, compared with the method of voting directly using ROI images, the accuracy is increased by 1.1%. The introduction of STN can effectively improve the robustness of convolutional neural network, compared with non-introduced STN networks, the accuracy is increased by 1.5%.

 Hits 1629  Downloads 845  Comment