A novel asymmetric image encryption and authentication scheme based on equal modulus decomposition (EMD) in the Fresnel transform domain is proposed. First, the Fresnel spectrum of the plaintext is sparsely sampled. Then a sparse presentation of the Fresnel spectrum is divided into two complex-valued masks with equal modulus based on EMD, both of which are necessary for decryption and authentication. The decrypted image renders little information on the original image visually. Furthermore, a nonlinear correlation is used to authenticate the decrypted image with the plaintext. In the scheme proposed here, the illuminating wavelength and the diffraction distance strengthen the security of the proposed scheme significantly, and the two complex-valued masks are both essential for decryption and authentication. The percentage of extracted pixels does not have to be determined deliberately. The feasibility and validity of the scheme have been proved by numerical simulations.

We propose an asymmetric optical image encryption scheme with silhouette removal by using interference and equal modulus decomposition (EMD). Plaintext is first separated into two complex value masks with the same modulus using EMD in the Fresnel transform domain. The two masks are encoded into four phase-only masks (POMs), two of which are treated as ciphertexts and other two as plaintext-dependent private keys by using the inverse Fresnel transform with different diffraction distances and interference-based encryption. Any information about the plaintext, including its silhouette, cannot be retrieved using one, two, or even three of the four POMs. Our scheme also avoids the constraint of the same modulus in EMD and eliminates the vulnerability against the iterative amplitude-phase attack and advanced iterative amplitude-phase attack. Numerical simulations were used to verify the validity and security of our proposed method.

Aiming at the problems of fuzzy detection characteristics, high false positive rate and low accuracy of traditional network intrusion detection technology, an improved intelligent intrusion detection method of industrial Internet of Things based on deep learning is proposed. Firstly, the data set is preprocessed and transformed into 122 dimensional intrusion data set after one-hot coding; Secondly, aiming at the problem that convolution network cannot deal with data with long-distance attributes, Bidirectional long short-term memory (BiLSTM) is used to mine the relationship between data features; At the same time, the Batch Normalization mechanism is introduced to speed up the training of deep neural network. After the activation function performs nonlinear transformation on the input data of the previous layer, it is normalized to ensure the trainability of the network. The experimental results on NSL-KDD data set show that the accuracy of the proposed CNN-BiLSTM model is 96.3%, the detection rate is 97.1%, and the performance is the best.

Aiming at the problems of over reliance on labor and low generalization of traditional emotion analysis methods based on dictionary and machine learning, an emotion analysis model of microblog comment text based on deep learning is proposed. Firstly, text is obtained through microblog crawler program. After data preprocessing, including data cleaning, Chinese word segmentation, removal of stop words, and so on, the Skip-gram model is used for word vector training on a large-scale unmarked corpus, and then the trained word vector is used as the text input of CNN-BiLSTM model, which combines Bidirectional Long-Short Term Memory (BiLSTM) neural network and Convolution Neural Network (CNN). Considering the historical context information and the subsequent context information, BiLSTM can better use the temporal relationship of text to learn sentence semantics. CNN can extract hidden features from the text and combine them. Finally, after Adamax optimization training, the emotion type of microblog comment text is output. The proposed model combines the learning advantages of BiLSTM and CNN. The overall accuracy of text emotion analysis has been greatly improved, with an accuracy of 0.94 and an improvement of 8.51% compared with the single CNN model.

No abstract