The current flight delay not only affects the normal operation of the current flight, but also spreads to the downstream flights through the flights schedule, resulting in a wide range of flight delays. The analysis and prediction of flight delay propagation in advance can help civil aviation departments control the flight delay rate and reduce the economic loss caused by flight delays. Due to the small number of data samples that can constitute flight chains, it is difficult to construct flight chain data. In recent years, the analysis of the flight delay propagation problem is generally based on traditional machine learning methods with a small sample size. After obtaining a large amount of raw data from the China Air Traffic Management Bureau, we have constructed 36,287 pieces of three-level flight chain data. Based on these data, we tried to use a deep learning method to analyze and forecast flight delays. In the field of deep learning, there are CNN models and RNN models that deal with classification problems well. Based on these two classes of models, we modify and innovate the study of the problem of flight delay propagation and prediction. Firstly, the CNN-based CondenseNet algorithm is used to predict the delay level of the three-level flight chain data. Based on this, the CondenseNet network is improved by inserting CBAM modules and named CBAM-CondenseNet. The experimental results show that the improved algorithm can effectively improve the network performance, and the prediction accuracy can reach 89.8%. Compared with the traditional machine learning method, the average prediction accuracy increased by 8.7 percentage points. On the basis of the CNN model, we also considered the superiority of the LSTM (Long Short-Term Memory network) considering the processing time sequence information, and then constructed the CNN-MLSTM network and injected the SimAM module to enhance the attention of flight chain data. In the experiment of flight delay propagation prediction, the accuracy rate is 91.36%, which is a significant improvement compared to using the CNN or LSTM alone.

Infrared image small target detection and identification is a massive challenge with vast applications in essential places such as military industries and airports. Homoplastically, the unclear outline of the target in infrared images, and the complex image background, and the tiny percentage of targets in infrared images make it challenging to accurately acquire the targets’ features and other characteristics, all of which make the research of target detection and identification in infrared images significantly challenging. In this paper, we first contribute an open dataset of Single-frame Infrared Small Target dataset dubbed CAUC-SIRST. After that, the similarity object enhancement module is obtained based on the input feature maps by calculating the Wasserstein distance between the local target and the local background. Then a heterogeneous parallel backbone network structure is constructed, and the feature maps obtained from three different backbone network channels are fused and stitched together. As above, the general convolution channel adaptively extracts feature information; the SimAM channel increases the attention of potential targets, and the similarity object enhancement channel increases the weight of targets. Finally, fusing the feature maps of the three different channels makes it possible to increase the weight of potential targets while retaining the original basic information, this method is named HPN-SOE. Embedding the algorithm into the camera can form a sensor for infrared target detection and recognition. With extensive experimental results demonstrate our outstanding performance, which outperforms other existing methods by achieving a detection accuracy of 85.7% and a detection speed of 31.2 fps.

Identifying small objects is essential in airport clearances, such as UAVs and bird flocks affecting the safety of the entire airport. Due to small objects and complex backgrounds, it is a formidable task to learn iconic representations from images. Convolution neural networks (CNNs) is the dominant framework in the past few years. Although CNNs can effectively mitigate local redundancy with convolutional kernels, it is also extremely challenging for convolutional kernels to obtain more feature information about the small object. Alternatively, the prediction box loss function has a critical influence on the eventual model performance. To resolve these problems, a revolutionary small object recognition network is proposed in this paper. In the structure, we propose a hybrid convolutional structure, which is able to obtain object features by different convolutional approaches to achieve more information about the small objects. Following that, we use the Double-One Path Aggregation Network (DO-PAN) feature fusion structure to further integrate the small object information. Finally, we propose all sides consistency intersection over union (AIoU) loss, which enables the model to predict the object location more exactly. Combining the three contributions, we designate this structure as HDA-Net. It can simply achieve 96.1% box mAP50 on the UAV-VIT dataset object recognition task, the inference speed on Tesla T4 is 28.46 fps.

With the rapid development of civil aviation transportation industry, the passenger throughput in civil aviation is increasing, while the problem of flight delays is becoming more and more serious. For flight delay prediction under big data, deep learning methods can be applied to make high-precision predictions. Since data preprocessing is one of the most important parts, the method based on hybrid encoding is proposed in this paper. Firstly, the flight and meteorological data are fused with the associated primary key, Since weather data has a greater impact on flight delay. Then, the fused data is encoded according to different data types. Min-Max encoding is used for continuous features, and CatBoost encoding is adopted for discrete features respectively. Finally, the data set which has been preprocessed can be put into the deep convolutional neural network ResNet to verify the effect. The experimental results show that the prediction accuracy rate of flight delay level can reach 94.02% on the structured data set after hybrid encoding.