In recent years, unsupervised person reidentification technology has made great strides. The technology retrieves images of interested persons under different cameras from massive repositories of unlabeled images. However, in the current research, there are some existing problems, such as the influence of pedestrians appearing across cameras and pseudo-label noise. To solve these problems, we conduct research in two ways: removing the camera bias and dynamically updating the memory model. In removing the camera bias, based on a learnable channel attention module, the features that are only related to cameras can be extracted from the feature map, thereby removing the camera bias in the global features and obtaining the features that can represent the pedestrians. In regards to dynamically updating the memory model, since the instance features do not necessarily belong to the identity represented by the pseudo-label, we adopt a method to update the memory dynamically according to the distance between the instance features and the category features so that the category features tend to be true. We combine the removal of the camera bias and the dynamic updating of the memory model to better solve problems in this field. Extensive experimentation demonstrates the superiority of our method over the state-of-the-art approaches on fully unsupervised Re-ID tasks.

Video-based person re-identification is an important research topic in computer vision that entails associating a pedestrian’s identity with non-overlapping cameras. It suffers from severe temporal appearance misalignment and visual ambiguity problems. We propose a novel self-supervised human semantic parsing approach (SS-HSP) for video-based person re-identification in this work. It employs self-supervised learning to adaptively segment the human body at pixel-level by estimating motion information of each body part between consecutive frames and explores complementary temporal relations for pursuing reinforced appearance and motion representations. Specifically, a semantic segmentation network within SS-HSP is designed, which exploits self-supervised learning by constructing a pretext task of predicting future frames. The network learns precise human semantic parsing together with the motion field of each body part between consecutive frames, which permits the reconstruction of future frames with the aid of several customized loss functions. Local aligned features of body parts are obtained according to the estimated human parsing. Moreover, an aggregation network is proposed to explore the correlation information across video frames for refining the appearance and motion representations. Extensive experiments on two video datasets have demonstrated the effectiveness of the proposed approach.

A surrounding vehicles behavior prediction method was presented for intelligent vehicles. The surrounding vehicles’ behavior is hard to predict since the significant uncertainty of vehicle driving and environmental changes. This method adopts bidirectional long short-term memory (BiLSTM) model combined with an encoder to ensure the memory of long-time series training. By constructing an attention mechanism based on BiLSTM, we consider the importance of different information which could guarantee the encoder’s memory under long sequence. The designed attention-bidirectional LSTM (Att-BiLSTM) model is adopted to ensure the surrounding vehicles’ prediction accuracy and effectiveness.

To address the problems of low recognition accuracy of traditional early fire warning systems in actual scenarios, a newly developed naive Bayes (NB) algorithm, namely, improved naive Bayes (INB), was proposed. An optimization method based on attribute weighting and an orthogonal matrix was used to improve the NB algorithm. Attribute weighting considers the influence of different values of each attribute on classification performance under every decision category; the orthogonal matrix weakens the linear relationship between the attributes reducing their correlations, which is more closely related to the conditional independence assumption. Data from the technology report of the National Institute of Standards and Technology (NIST) regarding fire research were used for the simulation, and eight datasets of different sizes were constructed for INB training and testing after filtering and normalization. A ten-fold cross-validation suggests that INB has been effectively trained and demonstrates the stable ability in fire alarms when the dataset contains 190 sets of samples; namely, the INB can be fully trained by using small datasets. A support vector machine (SVM), a back propagation (BP) neural network, and NB were selected for comparison. The results showed that the recognition accuracy, average precision, average recall, and average $\rm{F}_{1}$ measure of INB were 96.1%, 97.3%, 97.2%, and 97.3%, respectively, which is the highest among the four different algorithms. Additionally, INB has a better performance compared to NB, SVM, and BP neural networks when the training time is short . In conclusion, INB can be used as a core algorithm for fire alarm systems with excellent and stable fire alarm capabilities.

The rapid development of social media leads to the spread of a large amount of false news, which not only affects people’s daily life but also harms the credibility of social media platforms. Therefore, detecting Chinese fake news is a challenging and meaningful task. However, existing fake news datasets from Chinese social media platforms have a relatively small amount of data and data collection in this field is relatively old, thus being unable to meet the requirements of further research. In consideration of this background, we release a new Chinese Weibo Fake News dataset, which contains 26320 fake news data collected from Weibo. In addition, we propose a fake news detection model based on data augmentation that can effectively solve the problem of a lack of fake news, and we improve the generalization ability and robustness of the model. We conduct numerous experiments on our Chinese Weibo Fake News dataset and successfully deploy the model on the web page. The experimental performance proves the effectiveness of the proposed end-to-end model for detecting fake news on social media platforms.

Many pretrained deep learning models have been released to help engineers and researchers develop deep learning-based systems or conduct research with minimall effort. Previous work has shown that at secret message can be embedded in neural network parameters without compromising the accuracy of the model. Malicious developers can, therefore, hide malware or other baneful information in pretrained models, causing harm to society. Hence, reliable detection of these vicious pretrained models is urgently needed. We analyze existing approaches for hiding messages and find that they will ineluctably cause biases in the parameter statistics. Therefore, we propose steganalysis methods for steganography on neural network parameters that extract statistics from benign and malicious models and build classifiers based on the extracted statistics. To the best of our knowledge, this is the first study on neural network steganalysis. The experimental results reveal that our proposed algorithm can effectively detect a model with an embedded message. Notably, our detection methods are still valid in cases where the payload of the stego model is low.

Clarifying the mechanism of fungi growth is of great significance for maintaining the quality during grain storage. Among the factors that affect the growth of fungi spores, the most important factors are temperature, moisture content and storage time. Therefore, through this study, a multivariate linear regression model among several important factors, such as the spore number and ambient temperature, rice moisture content and storage days, were developed based on the experimental data. In order to build a more accurate model, we introduce a random forest algorithm into the fungal spore prediction during grain storage. The established regression models can be used to predict the spore number under different ambient temperature, rice moisture content and storage days during the storage process. For the random forest model, it could control the predicted value to be of the same order of magnitude as the actual value for 99% of the original data, which have a high accuracy to predict the spore number during the storage process. Furthermore, we plot the prediction surface graph to help practitioners to control the storage environment within the conditions in the low risk region.