Research On The Method Of Recognition Of The Anomie Behavior Of Air Baggage Self-service Check-in

The Self-service Baggage Check-in system can automatically check the weight,size and number of pieces of baggage,and transfer them to the transmission and sorting system after they are qualified.The standardization of passenger self-service operation process is the risk point of Baggage Self-service Check-in.The anomie behavior will cause the failure of transportation system and even affect the safety of air transportation.Among them,it is a typical risk point to check two or more pieces of baggage at the same time.At present,the detection methods based on two-dimensional and three-dimensional point cloud images of baggage are difficult to detect accurately because of the similarity and stacking of baggage.Therefore,it is of great significance to study the method of detecting the actual number of pieces of baggage based on the depth image sequence.The main work of this paper includes:Firstly,the definition of behavior,Self-service Baggage Check-in process and common Self-service Baggage Check-in behavior are researched and analyzed,and normative behavior and anomie behavior are defined,and a modeling method of Check-in behavior based on the motion history image is proposed.The method of data set collection for Check-in behavior is designed,and the data set for Check-in behavior is established.Secondly,the SIFT-BOW-SVM and C3 D networks based on machine learning methods are applied to the actual application environment of this paper to identify Check-in behavior.Thirdly,a recognition algorithm based on improved motion history image and convolutional neural network is proposed.Aiming at the background interference problem,a method of extracting key target regions based on the GrabCut algorithm is studied.At the same time,it is compared with the inter-frame difference method and the hybrid Gaussian background modeling combined with the inter-frame difference method.For the problems of information redundancy and real-time,the algorithm of key frame extraction and motion history image improvement is proposed;a convolutional neural network model is constructed for feature extraction and classification.Finally,this paper introduces the software and hardware structure of the Self-service Baggage Check-in system,and designs a Check-in behavior recognition system based on it,and tests the system.