Image recognition is a hot topic in the field of computer vision and pattern recognition, it is widely used in identification, automatic control, human-computer interaction systems. With the development of civil aviation, image recognition has become an important tool to ensure civil aviation security. In this article, firstly, tensor is used to represent the image, which can preserve more structure information of image than traditional vector representation. Then, combining a new tensor distance (NTD) and multilinear discriminant subspace analysis (MLDSA), a novel dimensionality reduction approach named NTD-MLDSA is proposed, and the transformation matrices can be obtained by employing an iterative strategy. Different from the Euclidean distance (ED), which bases on orthogonal assumption, NTD takes into account the spatial relationships of elements and can reflect the real distance between tensors. Experimental results show that the propose approach is more appropriate for dimensionality reduction of image objects than other classical dimension reduction methods, based on benchmark recognition databases Yale, ORL and USPS, the low dimensional data obtained by NTD-MLDSA improves the classification accuracy.
The airport ground taxiing phase is identified as a significant contributor to fuel consumption and carbon emissions within the aviation transport system. However, refined management and research remain challenged by data gaps and model imprecision. A research framework is proposed, combining high-precision Quick Access Recorder (QAR) data, statistical analysis, and machine learning. Main-engine fuel consumption differences across aircraft types, airports, and taxi-in and taxi-out operations were analyzed to identify the principal determinants of taxiing fuel consumption. A direct taxi-time–fuel baseline and an Adaptive Mechanism Learning (AML) model were constructed to predict aircraft taxiing fuel consumption. The AML model achieved a test-set coefficient of determination (R2) of 0.9672, while aircraft-specific R2 values exceeded 0.97 for A320, A350, and B777 aircraft. The results highlight the importance of taxiing duration and engine operating state in estimating main-engine fuel consumption and associated CO2 emissions. The findings provide a basis for refined fuel-consumption assessment and operational decision support for greener airport operations.
Tail striking is a typical safety event in the area of civil aviation, which is directly related to the aircraft pitch angle at landing. Based on 2933 A319 flights’ non-exceedance quick access recorder (QAR) data from Dali airport, the relationship between key flight parameters during the final approach and landing pitch angle is explored. Functional data analysis and the Group Lasso method are used to select the most important flight parameters, and cluster analysis and weighted logistic regression are used to identify and predict a “high-risk” flight pattern. Here, “high risk” refers to a flight pattern associated with a higher probability of large landing pitch attitude, which is used as a proxy indicator of potential tail-strike risk rather than as evidence of an actual tail-strike event. Finally, flight operation recommendations are provided. The research results indicate that the airspeed, pitch angle and engine speed are most closely related to the landing pitch angle. An unusually high-risk flight pattern is identified, characterized by “high airspeed, high attitude, low thrust” caused by improper energy management of light-load flights. About 32.4% of flights in this pattern land with “large landing attitude”, which means the landing pitch angle is larger than the 95% sample percentile. A prediction model for the high-risk pattern is established using QAR parameters at the heights of 500 ft, 450 ft, and 400 ft, with an accuracy rate of 99.7% on the test data. The prediction in advance at 400 ft can provide pilots with sufficient time to take necessary operations.