Planning, Construction and Operation Management of Civil Airports in China Based on Big Data Technology

Abstract

 Airports are intricate ecosystems with a lot going on. It might be difficult to oversee passenger movement, maintain security, and provide a flawless travel experience. Nonetheless, facial recognition and big data technologies are transforming airport operations, increasing their security, personalization, and efficiency. The training data used in facial recognition algorithms can introduce bias. For the technology to be used fairly and ethically, training datasets must be carefully chosen, and bias must be tested often. To overcome this complication, Management of Airports based on Big Data using optimized Dual Stream Spectrum Deconvolution Neural Network   is proposed. Initially, the images collected from the Labelled Faces in the Wild dataset are given as input. Afterward, the data are fed to pre-processing. In pre-processing, Adaptive Variational Bayesian Filtering (AVBF) is used for redundant of data and noise removal. The pre-processing output is fed to Fano-Factor Constrained Tunable Quality Wavelet Transform (FCTQWT) for feature extraction. The extracted features such as Check-in Counters, Security Screening and Boarding Gates are extracted. Then it is given for classification using Dual Stream Spectrum Deconvolution Neural Network (DSSDNN) optimized with Water Strider Algorithm (WSA) for classifying the person’s faces as known and unknown. Finally it under goes Radio Frequency Identification (RFID)Card Scanning for identifying objects just through the tags and then it passing Recommendation, Location Tracking for the passengers in the airport. The proposed MOA-DSSDNN-WSA-BD approach is implemented in MATLAB. The performance of the proposed MOA-DSSDNN-WSA-BD approach attains 23.3%, 25.4% and 21.9% high precision, 24.4%, 28.1% and 27.6% high recall and 24.1%, 22.4% and 27.5%  high F1-Score compared with existing methods such as Structural and operational management of Turkish airports (SOM-CART-TA),Flight delay prediction for commercial air transport (FDP-DBN-CAT), and  Flight Delay Prediction Based on Aviation Big Data and Machine Learning (FDP-LSTM-ABD) models respectively.