An intelligent sound system and method

Abstract

The application relates to the technical field of sound intelligent control, in particular to an intelligent sound system and method; a first crossing mode, a second crossing mode and a third crossing mode are obtained according to the state of a crossing sliding door; first crossing data and first moving data of the first crossing mode are identified to obtain a first abnormal area and a first abnormal coefficient, and then first orientation data are determined by combining the first abnormal area with a first distance between the crossing sliding door; second crossing data of the second crossing mode are identified to obtain a second abnormal area and a second abnormal coefficient; second orientation data are determined according to a second distance between the second abnormal area and the crossing sliding door and the second abnormal coefficient; third crossing data and second moving data of the third crossing mode are identified to obtain a third abnormal area and a third abnormal coefficient, and then third orientation data are determined by combining the third abnormal area with a third distance between the crossing sliding door. The intelligent control of the railway crossing sound equipment is realized through the orientation data of different modes.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of sound intelligent control, in particular to an intelligent sound system and method. BACKGROUND

[0002] A railway crossing is a place where a railway intersects with a highway, street or other traffic route, also known as a railway level crossing. In a railway transportation system, many places are not suitable for building bridges or tunnels due to geographical or economic reasons; therefore, railway crossings become an inevitable choice. Railway crossings are usually provided with traffic signs, warning signals and control facilities to ensure the safety of railway transportation and other traffic modes.

[0003] With the development of railway transportation, the safety of railway crossings has attracted more and more attention. In the prior art, railway crossing alarm systems usually rely on sound alarms to remind pedestrians and vehicles; however, since the sound devices used for early warning are generally fixed in one direction, they cannot be well adjusted in terms of direction according to the control state of the railway crossing, as well as the running status of the train, the traffic conditions and the behavior of the personnel, etc., and cannot focus on the alarm of the area with obvious abnormalities, resulting in unsatisfactory early warning effect.

[0004] With the development of data technology, dynamic and flexible regional early warning reminders can be made according to the driving parameters of the train, the real-time traffic conditions of the crossing and the behavior of the personnel, which can improve the traffic safety and road passing efficiency, and avoid the safety hazards of traditional early warning methods.

[0005] Therefore, an intelligent sound system and method are provided. SUMMARY

[0006] The present application aims to provide an intelligent sound system and method; a first crossing mode, a second crossing mode and a third crossing mode are obtained according to the state of the crossing sliding door; the first crossing data and the first moving data of the first crossing mode are identified to obtain a first abnormal area and a first abnormal coefficient, and then the first direction data is determined by combining the first abnormal area with the first distance of the crossing sliding door; the second crossing data of the second crossing mode is identified to obtain a second abnormal area and a second abnormal coefficient; the second direction data is determined according to the second distance of the crossing sliding door and the second abnormal coefficient of the second abnormal area; the third crossing data and the second moving data of the third crossing mode are identified to obtain a third abnormal area and a third abnormal coefficient, and then the third direction data is determined by combining the third abnormal area with the third distance of the crossing sliding door. The present application realizes intelligent control of the sound devices of the railway crossing through the direction data of different modes.

[0007] To achieve the above-mentioned purpose, the present application provides the following technical scheme:

[0008] An intelligent sound system comprises:

[0009] The crossing mode recognition module recognizes a crossing sliding door of a railway crossing, and divides a first crossing mode, a second crossing mode and a third crossing mode according to a state of the crossing sliding door.

[0010] The first mode analysis module acquires first crossing data and first movement data in the first crossing mode, constructs a first mode recognition model to recognize the first crossing data and the first movement data, obtains a first abnormal area and a first abnormal coefficient, acquires a first distance between the first abnormal area and the crossing sliding door, and determines first orientation data of the sound equipment according to the first distance and the first abnormal coefficient.

[0011] The second mode analysis module acquires second crossing data in the second crossing mode, constructs a second mode recognition model to recognize the second crossing data, obtains a second abnormal area and a second abnormal coefficient, acquires a second distance between the second abnormal area and the crossing sliding door, and determines second orientation data of the sound equipment according to the second distance and the second abnormal coefficient.

[0012] The third mode analysis module acquires third crossing data and second movement data in the third crossing mode, constructs a third mode recognition model to recognize the third crossing data and the second movement data, obtains a third abnormal area and a third abnormal coefficient, acquires a third distance between the third abnormal area and the crossing sliding door, and determines third orientation data of the sound equipment according to the third distance and the third abnormal coefficient.

[0013] The sound intelligent control module controls the sound equipment according to the first orientation data, the second orientation data and the third orientation data.

[0014] In the crossing mode recognition module, the division process of the first crossing mode, the second crossing mode and the third crossing mode is as follows: the first crossing mode is a process in which the crossing sliding door is from opening to closing; the second crossing mode is that the crossing sliding door is in a closed state; and the third crossing mode is a process in which the crossing sliding door is from closing to opening.

[0015] The first mode recognition model comprises a first data acquisition layer, a first crossing data recognition layer and a first abnormality recognition layer.

[0016] The first data acquisition layer acquires first crossing data and first movement data; the first crossing data is monitored video data of the railway crossing during the first crossing mode; and the first movement data comprises a closing speed of the crossing sliding door and a first passable width in the first crossing mode.

[0017] The first intersection data recognition layer recognizes and obtains first passing individual data and first passing environment data in the first intersection data; the first passing individual data comprises first passing individual type, first passing individual speed, first passing individual distance and first passing crowded data; the first passing environment data comprises environment visibility, road surface slipperiness, road surface flatness and environment noise data in the first intersection mode;

[0018] The first abnormality recognition layer recognizes a first passing abnormality coefficient of each passing individual according to the first passing individual data, the first passing environment data and the first moving data; a first abnormal area is obtained by clustering and dividing according to the position of the passing individual and the first passing abnormality coefficient; and a first abnormality coefficient is obtained according to the first passing abnormality coefficient of the passing individual in the first abnormal area.

[0019] The process of determining the first orientation data according to the first distance and the first abnormality coefficient is as follows:

[0020] A three-dimensional coordinate system is constructed with the position of the sound equipment as the origin; wherein the z-axis of the three-dimensional coordinate system is the vertical line of the ground plane passing through the origin;

[0021] The first alternative orientation data of the sound equipment is obtained according to the line connecting the first abnormal area and the sound equipment; the first alternative orientation data is vector data;

[0022] The first orientation weight of the first alternative orientation data is calculated according to the first distance between the first abnormal area and the intersection sliding door and the first abnormality coefficient of the first abnormal area; and the first orientation data is obtained according to all the first alternative orientation data and the corresponding first orientation weight.

[0023] The second mode recognition model comprises a second data acquisition layer, a second intersection data recognition layer and a second abnormality recognition layer;

[0024] The second data acquisition layer acquires second intersection data; the second intersection data is monitoring video data of the railway intersection during the second intersection mode;

[0025] The second intersection data recognition layer performs sound preprocessing on the second intersection data to obtain second preprocessed intersection data; sound abnormal passing individuals and behavior abnormal passing individuals in the second preprocessed intersection data are recognized and obtained; a sound abnormality coefficient is recognized according to the sound type, sound decibel and sound frequency of the sound abnormal passing individual; and a behavior abnormality coefficient is recognized according to the type, passing speed and behavior type of the behavior abnormal passing individual;

[0026] The second anomaly identification layer clusters the position data of the sound anomaly passing individual, the sound anomaly coefficient, the position data of the behavior anomaly passing individual and the behavior anomaly coefficient to obtain a second anomaly area; and obtains a second anomaly coefficient according to the sound anomaly coefficient of the sound anomaly passing individual and the behavior anomaly coefficient of the behavior anomaly passing individual in the second anomaly area.

[0027] The process of sound pre-processing of the second crossing data in the second crossing data identification layer is as follows: according to the distance between the train and the railway crossing and the passing sound data of the train, the environmental noise caused by the passing of the train in the second crossing data is processed to obtain second pre-processed crossing data.

[0028] The third mode identification model comprises a third data acquisition layer, a third crossing data identification layer and a third anomaly identification layer.

[0029] The third data acquisition layer acquires third crossing data and second movement data; the third crossing data is monitored video data of the railway crossing during the third crossing mode; and the second movement data comprises the opening speed of the crossing sliding door and the second passable width in the third crossing mode.

[0030] The third crossing data identification layer identifies the third crossing data to acquire third passing individual data and third passing environment data in the third crossing data; the third passing individual data comprises third passing individual type, third passing speed, third passing individual distance and third passing crowd data; and the third passing environment data comprises environmental visibility, road surface slipperiness, road surface flatness and environmental noise data in the third crossing mode.

[0031] The third anomaly identification layer identifies the third passing individual data, the third passing environment data and the second movement data to obtain a third passing anomaly coefficient of each passing individual; clusters and divides the third passing anomaly coefficient according to the position of the passing individual to obtain a third anomaly area; and obtains a third anomaly coefficient according to the third passing anomaly coefficient of the passing individual in the third anomaly area.

[0032] An intelligent sound method, comprising:

[0033] The crossing sliding door of the railway crossing is identified, and the first crossing mode, the second crossing mode and the third crossing mode are divided according to the state of the crossing sliding door;

[0034] The first crossing data and the first movement data in the first crossing mode are acquired; a first mode identification model is constructed to identify the first crossing data and the first movement data, to obtain a first anomaly area and a first anomaly coefficient; a first distance between the first anomaly area and the crossing sliding door is acquired, and first orientation data of the sound device is determined according to the first distance and the first anomaly coefficient;

[0035] acquire second crossing data in a second crossing mode; construct a second mode recognition model to recognize the second crossing data, obtain a second abnormal area and a second abnormal coefficient; acquire a second distance between the second abnormal area and a crossing sliding door, and determine second orientation data of the sound equipment according to the second distance and the second abnormal coefficient;

[0036] acquire third crossing data and second movement data in a third crossing mode; construct a third mode recognition model to recognize the third crossing data and the second movement data, obtain a third abnormal area and a third abnormal coefficient; acquire a third distance between the third abnormal area and the crossing sliding door, and determine third orientation data of the sound equipment according to the third distance and the third abnormal coefficient;

[0037] control the sound equipment according to the first orientation data, the second orientation data and the third orientation data.

[0038] The division processes of the first crossing mode, the second crossing mode and the third crossing mode are as follows: the first crossing mode is a process that the crossing sliding door is from opening to closing; the second crossing mode is that the crossing sliding door is in a closed state; and the third crossing mode is a process that the crossing sliding door is from closing to opening.

[0039] Compared with the prior art, the present application has the following beneficial effects:

[0040] 1. The present application recognizes the state of the crossing sliding door of the railway crossing, takes the process that the crossing sliding door is from opening to closing as the first crossing mode, takes the state that the crossing sliding door is in a closed state as the second crossing mode, and takes the process that the crossing sliding door is from closing to opening as the third crossing mode; the stages of the process that the train passes through the railway crossing are accurately divided through the first crossing mode, the second crossing mode and the third crossing mode, thereby providing support for subsequent abnormal area recognition and early warning.

[0041] 2. The present application constructs a first mode recognition model to recognize the first crossing data and the first movement data; first, the first crossing data is recognized to obtain first passing individual data and first passing environment data; the first passing individual data, the first passing environment data and the first movement data are recognized to obtain a first passing abnormal coefficient of each passing individual; the first passing abnormal coefficient is clustered and divided according to the position of the passing individual to obtain a first abnormal area; the first abnormal coefficient is obtained according to the first passing abnormal coefficient of the passing individual in the first abnormal area; and the abnormal area and the abnormal degree during the first crossing mode are accurately measured through the first abnormal area and the first abnormal coefficient.

[0042] 3、The second mode recognition model is constructed to recognize the second crossing data; sound abnormal passing individuals, sound abnormal coefficients, behavior abnormal passing individuals and behavior abnormal coefficients are obtained according to the second crossing data recognition; the second abnormal area is obtained by clustering according to the position data of the sound abnormal passing individuals, the sound abnormal coefficients, the position data of the behavior abnormal passing individuals and the behavior abnormal coefficients; the second abnormal coefficients are obtained according to the sound abnormal coefficients of the sound abnormal passing individuals and the behavior abnormal coefficients of the behavior abnormal passing individuals in the second abnormal area; the abnormal area and the abnormal degree during the second crossing mode are accurately measured through the second abnormal area and the second abnormal coefficients.

[0043] 4、The third mode recognition model is constructed to recognize the third crossing data and the second moving data; the third passing individual data and the third passing environment data are obtained according to the third crossing data; the third passing abnormal coefficients of each passing individual are obtained through the third passing individual data, the third passing environment data and the second moving data; the third abnormal area is obtained by clustering and dividing according to the position of the passing individual and the third passing abnormal coefficients; the third abnormal coefficients are obtained according to the third passing abnormal coefficients of the passing individual in the third abnormal area; the abnormal area and the abnormal degree during the third crossing mode are accurately measured through the third abnormal area and the third abnormal coefficients.

[0044] 5、The position of the sound equipment is taken as the origin to construct a three-dimensional coordinate system; the alternative orientation data of the sound equipment is obtained according to the connection line of the abnormal area and the sound equipment; the orientation weight of the alternative orientation data is obtained by measuring according to the distance between the abnormal area and the crossing door and the abnormal coefficients of the abnormal area; the orientation data is obtained according to all the alternative orientation data and the corresponding orientation weight; the warning focus of the sound equipment can accurately cover the abnormal area according to the orientation data, and the warning effect is effectively improved. BRIEF DESCRIPTION OF DRAWINGS

[0045] Figure 1 It is a structure schematic view of an intelligent sound system of the present application;

[0046] Figure 2 It is a structure schematic view of the first mode recognition model of the present application;

[0047] Figure 3 It is a flow schematic view of an intelligent sound method of the present application;

[0048] Figure 4 It is a highway passing scene schematic view of a railway crossing of the present application;

[0049] Figure 5 It is a railway passing scene schematic view of a railway crossing of the present application. DETAILED DESCRIPTION

[0050] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings of the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of the present application.

[0051] Embodiment one

[0052] The present application provides a kind of intelligent sound system, its structure is as shown in Figure 1 It includes intersection mode identification module, first mode analysis module, second mode analysis module, third mode analysis module and sound intelligent control module.

[0053] Intersection mode identification module, the intersection sliding door of railway crossing is identified, and first intersection mode, second intersection mode and third intersection mode are obtained according to the state division of the intersection sliding door.

[0054] The division process of the first intersection mode, second intersection mode and third intersection mode in the intersection mode identification module is as follows: the first intersection mode is the process that intersection sliding door is opened to closed;The second intersection mode is that intersection sliding door is in closed state;The third intersection mode is the process that intersection sliding door is closed to opened.

[0055] The present application identifies the state of intersection sliding door of railway crossing, and the process that intersection sliding door is opened to closed is taken as the first intersection mode, the closed state of intersection sliding door is taken as the second intersection mode, and the process that intersection sliding door is closed to opened is taken as the third intersection mode;The stage of the process that train passes through railway crossing is accurately divided through the first intersection mode, second intersection mode and third intersection mode, which provides support for subsequent abnormal area identification.

[0056] First mode analysis module, obtains the first intersection data and first movement data in the first intersection mode;First mode identification model is constructed to identify the first intersection data and first movement data, and first abnormal area and first abnormal coefficient are obtained;The first distance between first abnormal area and intersection sliding door is obtained, and the first orientation data of sound equipment is determined according to the first distance and first abnormal coefficient.

[0057] The first mode identification model is constructed based on deep neural network, and the structure is as shown in Figure 2 It includes first data acquisition layer, first intersection data identification layer and first abnormal identification layer.

[0058] The first data acquisition layer acquires first intersection data and first movement data; the first intersection data is monitored video data of a railway crossing during a first intersection mode; and the first movement data includes a closing speed of a movable door of the crossing and a first passable width during the first intersection mode.

[0059] The first intersection data identification layer identifies first pass individual data and first pass environment data in the acquired first intersection data; the first pass individual data includes a first pass individual type, a first pass individual speed, a first pass individual distance and first pass congestion data; and the first pass environment data includes environmental visibility, road surface slipperiness, road surface flatness and environmental noise data within the first intersection mode.

[0060] The first pass individual type includes pedestrians, electric vehicles, cars, trucks and tank trucks, etc.; the first pass congestion data is acquired by taking a pass individual as the center, determining a circular region according to a set radius threshold, and determining the first pass congestion data according to the number and type of pass individuals in the circular region; and the environmental noise data is acquired by a noise measuring instrument.

[0061] The first anomaly identification layer identifies a first pass anomaly coefficient of each pass individual according to the first pass individual data, the first pass environment data and the first movement data; a first anomaly region is divided by clustering according to the position of the pass individual and the first pass anomaly coefficient; and a first anomaly coefficient is obtained according to the first pass anomaly coefficient of the pass individual in the first anomaly region.

[0062] The first mode recognition model is constructed to identify the first intersection data and the first movement data; the first pass individual data and the first pass environment data are identified according to the first intersection data; the first pass anomaly coefficient of each pass individual is identified according to the first pass individual data, the first pass environment data and the first movement data; the first anomaly region is divided by clustering according to the position of the pass individual and the first pass anomaly coefficient; the first anomaly coefficient is obtained according to the first pass anomaly coefficient of the pass individual in the first anomaly region; and the first anomaly region and the first anomaly coefficient are used to accurately measure the anomaly region and the anomaly degree during the first intersection mode.

[0063] The process of determining the first orientation data according to the first distance and the first anomaly coefficient is as follows:

[0064] A three-dimensional coordinate system is constructed with the position of the sound equipment as the origin; the z-axis of the three-dimensional coordinate system is a vertical line of the ground plane passing through the origin.

[0065] The first alternative orientation data of the sound equipment is acquired according to the line connecting the first anomaly region and the sound equipment; and the first alternative orientation data is vector data.

[0066] According to the first distance between the first abnormal area and the crossing sliding door and the first abnormal coefficient of the first abnormal area, the first orientation weight of the first alternative orientation data is calculated; and the first orientation data is obtained according to all the first alternative orientation data and the corresponding first orientation weights.

[0067] The present invention constructs a three-dimensional coordinate system with the position of the audio equipment as the origin; obtains the first alternative orientation data of the audio equipment based on the connection line between the first abnormal area and the audio equipment; calculates the first orientation weight of the first alternative orientation data based on the first distance between the first abnormal area and the crossing sliding door and the first abnormal coefficient of the first abnormal area; obtains the first orientation data based on all the first alternative orientation data and the corresponding first orientation weights; and according to the first orientation data, the early warning focus of the audio equipment can accurately cover the abnormal area, effectively improving the early warning effect.

[0068] The second pattern analysis module obtains the second crossing data in the second crossing pattern; constructs a second pattern recognition model to identify the second crossing data, obtains the second abnormal area and the second abnormal coefficient; obtains the second distance between the second abnormal area and the crossing sliding door, and determines the second orientation data of the audio equipment based on the second distance and the second abnormal coefficient.

[0069] The second pattern recognition model includes a second data acquisition layer, a second crossing data recognition layer and a second anomaly recognition layer;

[0070] The second data acquisition layer acquires second railway crossing data; the second railway crossing data is surveillance video data of the railway crossing during the second railway crossing mode;

[0071] The second road crossing data recognition layer performs sound preprocessing on the second road crossing data to obtain second preprocessed road crossing data; identifies individuals with abnormal voices and individuals with abnormal behavior in the second preprocessed road crossing data; obtains a sound abnormality coefficient based on the sound type, sound decibel, and sound frequency of the individuals with abnormal voices; and obtains a behavior abnormality coefficient based on the type, speed, and behavior type of the individuals with abnormal behavior;

[0072] The sound types include the sound of vehicle horns, the sound of people quarreling, etc.; the abnormal behaviors include pedestrians approaching, climbing over the sliding gate of the intersection, vehicles overtaking, vehicles approaching the sliding gate of the intersection, etc.

[0073] The second abnormality identification layer clusters the position data, abnormal sound coefficient, position data and abnormal behavior coefficient of individuals with abnormal voices to obtain a second abnormal area; and obtains a second abnormality coefficient based on the abnormal sound coefficient and abnormal behavior coefficient of individuals with abnormal behavior in the second abnormal area.

[0074] The second mode recognition model is constructed to recognize the second crossing data; sound abnormal passing individuals, sound abnormal coefficients, behavior abnormal passing individuals and behavior abnormal coefficients are obtained according to the second crossing data; the second abnormal area is obtained by clustering according to the position data of the sound abnormal passing individuals, the sound abnormal coefficients, the position data of the behavior abnormal passing individuals and the behavior abnormal coefficients; the second abnormal coefficients are obtained according to the sound abnormal coefficients of the sound abnormal passing individuals and the behavior abnormal coefficients of the behavior abnormal passing individuals in the second abnormal area; the abnormal area and the abnormal degree during the second crossing mode are accurately measured through the second abnormal area and the second abnormal coefficients.

[0075] In the second crossing data recognition layer, the process of sound pre-processing of the second crossing data is as follows: according to the distance between the train and the railway crossing and the passing sound data of the train, the environmental noise caused by the passing of the train in the second crossing data is processed to obtain the second pre-processed crossing data.

[0076] The third mode analysis module obtains third crossing data and second moving data in the third crossing mode; a third mode recognition model is constructed to recognize the third crossing data and the second moving data, to obtain a third abnormal area and a third abnormal coefficient; a third distance between the third abnormal area and the crossing sliding door is obtained, and the third orientation data of the sound equipment is determined according to the third distance and the third abnormal coefficient.

[0077] The third mode recognition model comprises a third data acquisition layer, a third crossing data recognition layer and a third abnormality recognition layer.

[0078] The third data acquisition layer obtains third crossing data and second moving data; the third crossing data is the monitoring video data of the railway crossing during the third crossing mode; the second moving data comprises the opening speed of the crossing sliding door and the second passable width in the third crossing mode.

[0079] The third crossing data recognition layer recognizes the third crossing data to obtain third passing individual data and third passing environment data in the third crossing data; the third passing individual data comprises third passing individual type, third passing speed, third passing individual distance and third passing congestion data; the third passing environment data comprises environmental visibility, road surface slipperiness, road surface flatness and environmental noise data in the third crossing mode.

[0080] The third abnormality recognition layer obtains the third passing abnormal coefficient of each passing individual according to the third passing individual data, the third passing environment data and the second moving data; the third abnormal area is obtained by clustering and dividing according to the position of the passing individual and the third passing abnormal coefficient; the third abnormal coefficient is obtained according to the third passing abnormal coefficient of the passing individual in the third abnormal area.

[0081] The third mode recognition model is constructed to recognize the third crossing data and the second mobile data; first, the third crossing individual data and the third crossing environment data are recognized according to the third crossing data; the third crossing abnormality coefficient of each crossing individual is recognized through the third crossing individual data, the third crossing environment data and the second mobile data; the third abnormal area is obtained by clustering and dividing according to the crossing individual position and the third crossing abnormality coefficient; the third abnormality coefficient is obtained according to the third crossing abnormality coefficient of the crossing individual in the third abnormal area; and the abnormal area and the abnormal degree during the third crossing mode are accurately measured through the third abnormal area and the third abnormality coefficient.

[0082] The sound intelligent control module controls the sound equipment according to the first orientation data, the second orientation data and the third orientation data.

[0083] The present application further provides an intelligent sound method, and a flow thereof is as shown in the figure, which comprises the following steps. Figure 3

[0084] The crossing sliding door of the railway crossing is recognized, and the first crossing mode, the second crossing mode and the third crossing mode are obtained according to the state of the crossing sliding door;

[0085] The first crossing data and the first mobile data in the first crossing mode are obtained; the first mode recognition model is constructed to recognize the first crossing data and the first mobile data, and the first abnormal area and the first abnormality coefficient are obtained; the first distance between the first abnormal area and the crossing sliding door is obtained, and the first orientation data of the sound equipment is determined according to the first distance and the first abnormality coefficient;

[0086] The second crossing data in the second crossing mode is obtained; the second mode recognition model is constructed to recognize the second crossing data, and the second abnormal area and the second abnormality coefficient are obtained; the second distance between the second abnormal area and the crossing sliding door is obtained, and the second orientation data of the sound equipment is determined according to the second distance and the second abnormality coefficient;

[0087] The third crossing data and the second mobile data in the third crossing mode are obtained; the third mode recognition model is constructed to recognize the third crossing data and the second mobile data, and the third abnormal area and the third abnormality coefficient are obtained; the third distance between the third abnormal area and the crossing sliding door is obtained, and the third orientation data of the sound equipment is determined according to the third distance and the third abnormality coefficient;

[0088] The sound equipment is controlled according to the first orientation data, the second orientation data and the third orientation data.

[0089] ​The first crossing mode, the second crossing mode and the third crossing mode are obtained according to the state of the crossing sliding door; the first abnormal area and the first abnormal coefficient are obtained by identifying the first crossing data and the first moving data of the first crossing mode, and the first orientation data is determined by combining the first abnormal area and the first distance of the crossing sliding door; the second abnormal area and the second abnormal coefficient are obtained by identifying the second crossing data of the second crossing mode; the second orientation data is determined according to the second distance of the crossing sliding door, the second abnormal area and the second abnormal coefficient; the third abnormal area and the third abnormal coefficient are obtained by identifying the third crossing data and the second moving data of the third crossing mode, and the third orientation data is determined by combining the third abnormal area and the third distance of the crossing sliding door. The intelligent control of the railway crossing sound equipment is realized through the orientation data of different modes.

[0090] Embodiment two

[0091] An intelligent sound system is used to control the actual railway crossing; the railway crossing is the intersection of the highway and the railway, the highway passing scene is as shown in Figure 4 , and the rail passing scene is as shown in Figure 5 .

[0092] The intelligent sound system provided by the application has the structure as shown in Figure 1 ; and comprises a crossing mode identification module, a first mode analysis module, a second mode analysis module, a third mode analysis module and a sound intelligent control module.

[0093] The crossing mode identification module identifies the crossing sliding door of the railway crossing, and the first crossing mode, the second crossing mode and the third crossing mode are obtained according to the state of the crossing sliding door.

[0094] The division process of the first crossing mode, the second crossing mode and the third crossing mode in the crossing mode identification module is that the first crossing mode is the process of the crossing sliding door from opening to closing; the second crossing mode is that the crossing sliding door is in a closed state; and the third crossing mode is the process of the crossing sliding door from closing to opening.

[0095] The state of the crossing sliding door of the railway crossing is identified, the process of the crossing sliding door from opening to closing is taken as the first crossing mode, the crossing sliding door in the closed state is taken as the second crossing mode, and the process of the crossing sliding door from closing to opening is taken as the third crossing mode; the stages of the process of the train passing through the railway crossing are accurately divided through the first crossing mode, the second crossing mode and the third crossing mode, so as to provide support for subsequent abnormal area identification and early warning.

[0096] The first mode analysis module obtains first intersection data and first movement data in the first intersection mode, constructs a first mode recognition model to recognize the first intersection data and the first movement data, obtains a first abnormal area and a first abnormal coefficient, obtains a first distance between the first abnormal area and a moving door of the intersection, and determines first orientation data of the sound equipment according to the first distance and the first abnormal coefficient.

[0097] The first mode recognition model is constructed based on a deep neural network, and has a structure as shown in Figure 2

[0098] The first data acquisition layer obtains the first intersection data and the first movement data; the first intersection data is monitored video data of a railway intersection during the first intersection mode; and the first movement data includes a closing speed of the moving door of the intersection in the first intersection mode and a first passable width.

[0099] The first intersection data recognition layer recognizes first pass individual data and first pass environment data in the first intersection data; the first pass individual data includes a first pass individual type, a first pass individual speed, a first pass individual distance, and first pass crowded data; and the first pass environment data includes environmental visibility, road surface slipperiness, road surface flatness, and environmental noise data in the first intersection mode.

[0100] The first pass individual type includes a pedestrian, a battery car, a car, a truck, and a tank car; the first pass individual distance is a distance between a pass individual and the moving door of the intersection; the first pass crowded data is obtained by determining a circular area with the pass individual as the center and according to a set radius threshold, and determining the first pass crowded data according to a number and a type of pass individuals in the circular area; and the environmental noise data is obtained by a noise measuring instrument.

[0101] The first pass individual data at a start time of the first intersection mode is recognized to obtain data as shown in Table 1.

[0102] Table 1: First pass individual data table

[0103] First Pedestrian ID First Pedestrian Type First Pedestrian Speed First Pedestrian Distance S01 Electric Car 2.24 m / s 16.25m S02 Car 4.58 m / s 18.34m S03 Car 5.24 m / s 15.26m S04 Bus 3.38 m / s 10.85m

[0104] The first abnormal recognition layer recognizes a first pass abnormal coefficient of each pass individual according to the first pass individual data, the first pass environment data, and the first movement data, clusters and divides the first abnormal area according to a position of the pass individual and the first pass abnormal coefficient, and obtains the first abnormal coefficient according to the first pass abnormal coefficient of the pass individual in the first abnormal area.

[0105] ​The first mode recognition model is constructed to recognize the first intersection data and the first moving data; first, the first passing individual data and the first passing environment data are recognized according to the first intersection data; the first passing abnormality coefficient of each passing individual is recognized through the first passing individual data, the first passing environment data and the first moving data; the first abnormal area is obtained by clustering and dividing according to the position of the passing individual and the first passing abnormality coefficient; the first abnormality coefficient is obtained according to the first passing abnormality coefficient of the passing individual in the first abnormal area; the abnormal area and the abnormal degree during the first intersection mode are accurately measured through the first abnormal area and the first abnormality coefficient.

[0106] The process of determining the first orientation data according to the first distance and the first abnormality coefficient is as follows:

[0107] The position of the sound equipment is taken as the origin to construct a three-dimensional coordinate system; wherein the z-axis of the three-dimensional coordinate system is the vertical line of the ground plane passing through the origin;

[0108] The first alternative orientation data of the sound equipment is obtained according to the connection line between the first abnormal area and the sound equipment; the first alternative orientation data is vector data;

[0109] The first orientation weight of the first alternative orientation data is calculated according to the first distance between the first abnormal area and the intersection sliding door and the first abnormality coefficient of the first abnormal area; the first orientation data is obtained according to all the first alternative orientation data and the corresponding first orientation weight.

[0110] The position of the sound equipment is taken as the origin to construct a three-dimensional coordinate system; the first alternative orientation data of the sound equipment is obtained according to the connection line between the first abnormal area and the sound equipment; the first orientation weight of the first alternative orientation data is calculated according to the first distance between the first abnormal area and the intersection sliding door and the first abnormality coefficient of the first abnormal area; the first orientation data is obtained according to all the first alternative orientation data and the corresponding first orientation weight; the warning focus of the sound equipment can accurately cover the abnormal area according to the first orientation data, and the warning effect is effectively improved.

[0111] The second mode analysis module obtains the second intersection data in the second intersection mode; a second mode recognition model is constructed to recognize the second intersection data, and the second abnormal area and the second abnormality coefficient are obtained; the second distance between the second abnormal area and the intersection sliding door is obtained, and the second orientation data of the sound equipment is determined according to the second distance and the second abnormality coefficient.

[0112] The second mode recognition model includes a second data acquisition layer, a second intersection data recognition layer and a second abnormality recognition layer;

[0113] The second data acquisition layer acquires second crossing data; the second crossing data is monitored video data of a railway crossing during a second crossing mode;

[0114] The second crossing data identification layer performs sound preprocessing on the second crossing data to obtain second preprocessed crossing data; identifies sound abnormal passing individuals and behavior abnormal passing individuals in the second preprocessed crossing data; identifies a sound abnormal coefficient according to a sound type, a sound decibel and a sound frequency of the sound abnormal passing individuals; and identifies a behavior abnormal coefficient according to a type, a passing speed and a behavior type of the behavior abnormal passing individuals.

[0115] The sound type includes a vehicle horn sound, a crowd quarrel sound and the like; and the behavior abnormality includes a pedestrian approaching, a crossing door jumping, a vehicle overtaking, a lane changing, a vehicle approaching a crossing door and the like.

[0116] The behavior abnormal data during the second crossing mode is shown in Table 2.

[0117] Table 2: Behavior Abnormal Passing Individual Data Table

[0118] Abnormal Pedestrian ID Type Speed Behavior Type T01 Pedestrian 1.86 m / s Approaching Crossing Gate T02 Car 2.51 m / s Changing Lane T03 Car 2.64 m / s Approaching Crossing Gate

[0119] The second abnormality identification layer clusters the position data of the sound abnormal passing individuals, the sound abnormal coefficient, the position data of the behavior abnormal passing individuals and the behavior abnormal coefficient to obtain a second abnormal area; and obtains a second abnormal coefficient according to the sound abnormal coefficient of the sound abnormal passing individuals and the behavior abnormal coefficient of the behavior abnormal passing individuals in the second abnormal area.

[0120] The application constructs a second mode identification model to identify the second crossing data; identifies sound abnormal passing individuals, a sound abnormal coefficient, behavior abnormal passing individuals and a behavior abnormal coefficient according to the second crossing data; clusters the position data of the sound abnormal passing individuals, the sound abnormal coefficient, the position data of the behavior abnormal passing individuals and the behavior abnormal coefficient to obtain a second abnormal area; obtains a second abnormal coefficient according to the sound abnormal coefficient of the sound abnormal passing individuals and the behavior abnormal coefficient of the behavior abnormal passing individuals in the second abnormal area; and accurately measures the abnormal area and the abnormal degree during the second crossing mode through the second abnormal area and the second abnormal coefficient.

[0121] In the second crossing data identification layer, the process of performing sound preprocessing on the second crossing data is as follows: according to the distance between a train and a railway crossing and the passing sound data of the train, the environmental noise caused by the train passing in the second crossing data is processed to obtain the second preprocessed crossing data.

[0122] A third mode analysis module obtains third intersection data and second movement data in the third intersection mode, constructs a third mode recognition model to recognize the third intersection data and the second movement data, obtains a third abnormal area and a third abnormal coefficient, obtains a third distance between the third abnormal area and a movable door of the intersection, and determines third orientation data of the sound equipment according to the third distance and the third abnormal coefficient.

[0123] The third mode recognition model comprises a third data acquisition layer, a third intersection data recognition layer and a third abnormality recognition layer.

[0124] The third data acquisition layer obtains third intersection data and second movement data, the third intersection data is monitored video data of a railway intersection during a third intersection mode, and the second movement data comprises an opening speed of a movable door of the intersection and a second passable width in the third intersection mode.

[0125] The third intersection data recognition layer recognizes the third intersection data, obtains third passing individual data and third passing environment data in the third intersection data, the third passing individual data comprises a third passing individual type, a third passing speed, a third passing individual distance and third passing congestion data, and the third passing environment data comprises environmental visibility, road surface slipperiness, road surface flatness and environmental noise data in the third intersection mode.

[0126] The third abnormality recognition layer recognizes a third passing abnormal coefficient of each passing individual according to the third passing individual data, the third passing environment data and the second movement data, obtains a third abnormal area according to the position of the passing individual and the third passing abnormal coefficient, and obtains a third abnormal coefficient according to the third passing abnormal coefficient of the passing individual in the third abnormal area.

[0127] The third mode recognition model is constructed to recognize the third intersection data and the second movement data, the third passing individual data and the third passing environment data are recognized according to the third intersection data, the third passing abnormal coefficient of each passing individual is recognized according to the third passing individual data, the third passing environment data and the second movement data, the third abnormal area is obtained according to the position of the passing individual and the third passing abnormal coefficient, the third abnormal coefficient is obtained according to the third passing abnormal coefficient of the passing individual in the third abnormal area, and the abnormal area and the abnormal degree during the third intersection mode are accurately measured through the third abnormal area and the third abnormal coefficient.

[0128] The sound intelligent control module controls the sound equipment according to the first orientation data, the second orientation data and the third orientation data.

[0129] In order to verify the early warning effect of the intelligent sound system, the intelligent sound system is verified.

[0130] The verification process adopts system 1 and system 2, the system 1 is an intelligent sound system of the application; the system 2 is a fixed orientation sound early warning system;

[0131] In the verification process, the position selection of the sound equipment includes position 1, position 2 and position 3;

[0132] The verification is divided into three stages, corresponding to the first crossing mode, the second crossing mode and the third crossing mode; in the first crossing mode, the average value of the first crossing abnormal coefficient of the crossing individual is taken as the first verification index; in the second crossing mode, the average value of the second abnormal coefficient of the crossing individual is taken as the second verification index; in the third crossing mode, the average value of the third crossing abnormal coefficient of the crossing individual is taken as the third verification index; the data obtained through multiple verifications is shown in Table 3.

[0133] Table 3 Intelligent sound system early warning effect data table

[0134]

[0135] According to the data in Table 3, under the same position condition, the early warning effect of system 1 is better than that of system 2.

[0136] According to the state of the crossing sliding door, the first crossing mode, the second crossing mode and the third crossing mode are obtained; the first crossing data and the first moving data of the first crossing mode are identified to obtain the first abnormal area and the first abnormal coefficient, and then the first orientation data is determined in combination with the first distance between the first abnormal area and the crossing sliding door; the second crossing data of the second crossing mode is identified to obtain the second abnormal area and the second abnormal coefficient; the second orientation data is determined according to the second distance between the second abnormal area and the crossing sliding door and the second abnormal coefficient; the third crossing data and the second moving data of the third crossing mode are identified to obtain the third abnormal area and the third abnormal coefficient, and then the third orientation data is determined in combination with the third distance between the third abnormal area and the crossing sliding door. Through the orientation data of different modes, the intelligent control of the railway crossing sound equipment is realized.

[0137] Although the embodiments of the application have been shown and described, it is to be understood that for the purpose of the present application, the changes, modifications, replacements and variations of these embodiments can be made by those skilled in the art without departing from the principles and spirit of the application, and the scope of the application is defined by the appended claims and their equivalents.

Claims

1. An intelligent sound system, characterized in that: include: A crossing mode recognition module identifies the sliding gates at the railway crossing and classifies the sliding gates into a first crossing mode, a second crossing mode, and a third crossing mode according to the state of the sliding gates; A first pattern analysis module obtains first crossing data and first movement data in the first crossing pattern; constructs a first pattern recognition model to identify the first crossing data and the first movement data, obtaining a first abnormal area and a first abnormal coefficient; obtains a first distance between the first abnormal area and the crossing sliding door, and determines first orientation data of the audio device based on the first distance and the first abnormal coefficient; The first movement data includes a closing speed of the sliding door of the crossing and a first passable width in the first crossing mode; The first crossing data is surveillance video data of a railway crossing during a first crossing mode, and first passing individual data and first passing environment data are obtained by identification; the first passing individual data includes a first passing individual type, a first passing individual speed, a first passing individual distance, and first passing congestion data; The first traffic congestion data is obtained by: taking the passing individual as the center of the circle, determining a circular area according to a set radius threshold, and determining the first traffic congestion data according to the number and type of the passing individuals within the circular area; A second pattern analysis module obtains second crossing data in the second crossing pattern; constructs a second pattern recognition model to identify the second crossing data, obtains a second abnormal area and a second abnormal coefficient; obtains a second distance between the second abnormal area and the crossing sliding door, and determines second orientation data of the audio device based on the second distance and the second abnormal coefficient; a third pattern analysis module for acquiring third crossing data and second movement data in the third crossing pattern; constructing a third pattern recognition model to identify the third crossing data and the second movement data, obtaining a third abnormal area and a third abnormal coefficient; obtaining a third distance between the third abnormal area and the crossing sliding door, and determining third orientation data of the audio device based on the third distance and the third abnormal coefficient; The audio intelligent control module controls the audio device according to the first orientation data, the second orientation data and the third orientation data.

2. The intelligent audio system according to claim 1, characterized in that: In the road crossing pattern recognition module, the process of dividing the first road crossing pattern, the second road crossing pattern and the third road crossing pattern is as follows: the first road crossing pattern is the process of the road crossing sliding door from opening to closing; The second crossing mode is when the crossing sliding door is in a closed state; the third crossing mode is when the crossing sliding door is in a process from closed to open.

3. The intelligent audio system according to claim 1, characterized in that: The first pattern recognition model includes a first data acquisition layer, a first crossing data recognition layer and a first anomaly recognition layer; The first data acquisition layer acquires first crossing data and first movement data; The first road crossing data recognition layer identifies and obtains first individual pass data and first traffic environment data in the first road crossing data; the first individual pass data includes first individual pass type, first individual pass speed, first individual pass distance, and first traffic congestion data; the first traffic environment data includes environmental visibility, road surface slipperiness, road surface smoothness, and environmental noise data within the first road crossing mode; the first individual pass distance is the distance between the individual pass and the road crossing sliding door; The first abnormality identification layer obtains the first abnormality coefficient of each passing individual based on the first passing individual data, the first passing environment data and the first movement data; obtains the first abnormal area based on the clustering of the passing individual position and the first abnormality coefficient; and obtains the first abnormality coefficient based on the first abnormality coefficient of the passing individual in the first abnormal area.

4. The intelligent audio system according to claim 1, wherein: The process of determining the first orientation data according to the first distance and the first anomaly coefficient is as follows: A three-dimensional coordinate system is constructed with the location of the audio equipment as the origin; wherein the z-axis of the three-dimensional coordinate system is a perpendicular line to the ground plane passing through the origin; Acquiring first candidate orientation data of the audio device according to a connection line between the first abnormal area and the audio device; the first candidate orientation data is vector data; According to the first distance between the first abnormal area and the crossing sliding door and the first abnormal coefficient of the first abnormal area, the first orientation weight of the first alternative orientation data is calculated; and the first orientation data is obtained according to all the first alternative orientation data and the corresponding first orientation weights.

5. The intelligent audio system according to claim 4, characterized in that: The second pattern recognition model includes a second data acquisition layer, a second crossing data recognition layer and a second anomaly recognition layer; The second data acquisition layer acquires second railway crossing data; the second railway crossing data is surveillance video data of the railway crossing during the second railway crossing mode; The second road crossing data recognition layer performs sound preprocessing on the second road crossing data to obtain second preprocessed road crossing data; Identifying and obtaining individuals passing through with abnormal sounds and individuals passing through with abnormal behaviors in the second pre-processed crossing data; According to the voice type, sound decibel and sound frequency of individuals with abnormal voices, the sound abnormality coefficient is identified; according to the type, speed and behavior type of individuals with abnormal behavior, the behavior abnormality coefficient is identified; The second abnormality identification layer clusters the position data, abnormal sound coefficient, position data and abnormal behavior coefficient of individuals with abnormal voices to obtain a second abnormal area; and obtains a second abnormality coefficient based on the abnormal sound coefficient and abnormal behavior coefficient of individuals with abnormal behavior in the second abnormal area.

6. The intelligent audio system according to claim 5, characterized in that: In the second crossing data recognition layer, the process of sound preprocessing of the second crossing data is: according to the distance between the train and the railway crossing and the sound data of the train passing, the environmental noise caused by the passage of the train in the second crossing data is processed to obtain the second preprocessed crossing data.

7. The intelligent audio system according to claim 1, characterized in that: The third pattern recognition model includes a third data acquisition layer, a third crossing data recognition layer and a third anomaly recognition layer; The third data acquisition layer acquires third crossing data and second movement data; the third crossing data is surveillance video data of the railway crossing during the third crossing mode; the second movement data includes the opening speed of the crossing sliding door and the second passable width in the third crossing mode; The third crossing data recognition layer recognizes the third crossing data and obtains third individual passerby data and third traffic environment data from the third crossing data; the third individual passerby data includes third individual passerby type, third passerby speed, third individual passerby distance, and third traffic congestion data; the third traffic environment data includes environmental visibility, road surface slipperiness, road surface smoothness, and environmental noise data within the third crossing mode; The third abnormality identification layer obtains the third abnormality coefficient of each passing individual based on the third passing individual data, the third passing environment data and the second movement data; obtains the third abnormal area based on the clustering of the passing individual position and the third abnormality coefficient; and obtains the third abnormality coefficient based on the third abnormality coefficient of the passing individual in the third abnormal area.

8. An intelligent audio method, characterized in that, include: Identify the sliding gates at railway crossings and classify them into the first, second, and third crossing modes based on their status. Obtaining first crossing data and first movement data in a first crossing pattern; constructing a first pattern recognition model to identify the first crossing data and the first movement data, obtaining a first abnormal area and a first abnormal coefficient; obtaining a first distance between the first abnormal area and the crossing sliding door, and determining first orientation data of the audio device based on the first distance and the first abnormal coefficient; The first movement data includes a closing speed of the sliding door of the crossing and a first passable width in the first crossing mode; The first crossing data is surveillance video data of a railway crossing during a first crossing mode, and first passing individual data and first passing environment data are obtained by identification; the first passing individual data includes a first passing individual type, a first passing individual speed, a first passing individual distance, and first passing congestion data; The first traffic congestion data is obtained by: taking the passing individual as the center of the circle, determining a circular area according to a set radius threshold, and determining the first traffic congestion data according to the number and type of the passing individuals within the circular area; Obtaining second crossing data in a second crossing pattern; constructing a second pattern recognition model to identify the second crossing data, obtaining a second abnormal area and a second abnormal coefficient; obtaining a second distance between the second abnormal area and the crossing sliding door, and determining second orientation data of the audio device based on the second distance and the second abnormal coefficient; Obtain third crossing data and second movement data in a third crossing mode; construct a third pattern recognition model to identify the third crossing data and the second movement data, and obtain a third abnormal area and a third abnormal coefficient; obtain a third distance between the third abnormal area and the crossing sliding door, and determine third orientation data of the audio device based on the third distance and the third abnormal coefficient; The audio device is controlled according to the first orientation data, the second orientation data, and the third orientation data.

9. The intelligent audio method according to claim 8, characterized in that: The division process of the first crossing mode, the second crossing mode and the third crossing mode is: the first crossing mode is the process of the crossing sliding door from opening to closing; the second crossing mode is the crossing sliding door in the closed state; the third crossing mode is the process of the crossing sliding door from closing to opening.

Images