An Image Recognition Algorithm Based on Self-Encoding and Convolutional Neural Network Fusion

摘要

With the advent of big data and the era of universal Internet, the powerful representation capability of convolutional neural networks in large-scale visual recognition is particularly important, but image recognition performance is limited by the influence of environmental noise. To this end, this paper proposes an image recognition algorithm based on self-encoding and convolutional neural network fusion. The noise reduction algorithm is used to reduce the noise of the input image, and then the trained convolutional neural network is used to identify the image. The noise reduction self-encoder not only makes the public training data set suitable for the actual environment, but also improves the usage rate of the training model, so that the model is decoupled from the specific environment. The experimental results show that the image recognition algorithm proposed in this paper can have a good recognition rate with/without ambient noise.