Binocular image bus passenger flow calculation method associated with bus door state

Abstract

The invention belongs to the technical field of image processing, particularly to a binocular image bus passenger flow calculation method associated with a bus door state. On the basis of adopting busdoor linkage to perform double background elimination under door opening and closing backgrounds and a bus door state linkage motion trajectory tracking method, depth estimation is performed on a binocular image collected by a binocular camera, and a parallax error is determined for each pixel in the binocular image; and judgment is made according to a depth map made by the parallax errors, humanhead feature matching and motion trajectory tracking are performed on a suspected head, and passengers getting on and getting off in a human head area satisfying a condition are counted. According tothe method provided by the invention, on the basis of an original binocular image bus passenger flow calculation method, through a double background elimination processing technology under door opening and closing backgrounds in combination with bus door linkage, a passenger flow overstating error brought by bus door opening and closing is reduced; and by adoption of the motion trajectory tracking method based on the bus door state, a getting-on / getting-off passenger flow overstating error caused by back-and-forth shaking in the vicinity of an image recognition area is reduced.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to a bus passenger flow calculation method based on binocular images associated with vehicle door states. Background technique [0002] Buses are an important means of public transport for people to travel. Through the statistics of its passenger flow, we can know the passenger flow of each time period, each bus, each station, etc., and use these data to carry out intelligent scheduling, so that public resources can be more fully utilized, and passenger flow statistics are accurate. The degree directly affects the operation of the intelligent transportation system and the efficiency of bus operation. By displaying the status and changes of the current number of passengers, the bus station can take measures to prevent emergencies when there are a large number of passengers. [0003] The main methods of existing passenger flow statistics technology include weigh...

AI Technical Summary

Problems solved by technology

[0005] (1) When opening the door, taller objects (such as car doors) that already exist in the background environment are misidentified as human heads when passengers are identified, which is easily misjudged as the same as the movement of human heads, resulting in miscalculation of passenger flow

[0006] (2) Passengers in the car shaking in the recognition area will cause misjudgment of getting on / off the car, which is likely to cause miscalculation

[0007] (3) In crowded conditions, there may be situations where the door of a passenger closes when the passenger has just boarded the bus and has not moved, and there is a missed boarding record.

Images