Smart city road video continuous coverage method based on key area perception

Abstract

The invention provides a smart city road video continuous coverage method based on key area perception. The method comprises the steps of analyzing images acquired by image acquisition equipment basedon a neural network to obtain a vehicle thermodynamic diagram, performing heat statistical operation on a new thermodynamic diagram obtained after superposition operation is performed on the vehiclethermodynamic diagram, and adjusting a refresh rate of the image acquisition equipment of each road section according to heat statistical results of different road sections; carrying out projection transformation on the images acquired by the image acquisition equipment with the self-adjustable refresh rate, and then carrying out image splicing operation; and projecting the spliced images into a pre-built city CIM, realizing continuous coverage of city road videos, and performing visualization processing on information in the CIM in combination with a Web GIS technology. According to the method, the loss of the camera can be reduced while key information is effectively collected; heat statistics is performed based on the neural network, the refresh rate of the image acquisition equipment can be adjusted in real time, and the traffic condition of the road can be better reflected.

Description

technical field [0001] The invention belongs to the fields of computer vision and artificial intelligence, and in particular relates to a method for continuous coverage of road video in smart cities based on key area perception. Background technique [0002] In 2018, the Central Economic Work Conference was held in Beijing. The conference redefined infrastructure construction, defining 5G, artificial intelligence, industrial Internet, and Internet of Things as "new infrastructure construction", and then "strengthening the construction of a new generation of information infrastructure" was considered Included in the 2019 government work report, but the "new infrastructure" that has been implemented in real scenarios is rare. [0003] Currently, the cameras used for road monitoring are only responsible for the image acquisition of the monitoring area. Due to the existence of repeated acquisition and overlapping vehicles, it is difficult to integrate the information when proces...

AI Technical Summary

Problems solved by technology

[0002] In 2018, the Central Economic Work Conference was held in Beijing. The conference redefined infrastructure construction, defining 5G, artificial intelligence, industrial Internet, and Internet of Things as "new infrastructure construction", and then "strengthening the construction of a new generation of information infrastructure" was considered Included in the 2019 government work report, but the "new infrastructure" that has been implemented in real scenarios is rare

[0003] At present, the cameras used for road monitoring are only responsible for the image collection of the monitoring area. Due to the existence of repeated collection and overlapping vehicles, it is difficult to integrate the information when processing the image, resulting in a low utilization rate of the obtained image information.

At present, the cameras used for road information collection are generally set to a fixed refresh rate. The defect of this method is that the fixed high refresh rate will lead to excessive power consumption of the device or incomplete information collection at critical moments.

Images