The invention discloses a vehicle manned identification method. The method comprises the following steps: acquiring image data of a vehicle passing through a monitoring area; identifying a vehicle image and a figure image from the image data by adopting a deep residual convolutional neural network model; obtaining vehicle image matrix data and figure image matrix data according to the vehicle image and the figure image; conducting analyzing according to the vehicle image, the figure image, the vehicle image matrix data and the figure image matrix data, and judging whether the vehicle has a manned behavior or not; and if the manned behavior is monitored, storing the image, and sending out an alarm signal. According to the method, the vehicle and the person are identified by using the deep residual convolutional neural network model in the acquired vehicle monitoring image, and the vehicle image and the person image are analyzed so that whether the vehicle has the manned behavior can beaccurately judged, the illegal manned behavior of the vehicle can be effectively monitored and prevented, the identification accuracy is high and the application range is wide.
本发明公开了一种车辆载人识别方法,包括:获取车辆经过监控区域的图像数据;从所述图像数据中采用深层式残差卷积神经网络模型识别出车辆图像和人物图像;根据车辆图像和人物图像得到车辆图像矩阵数据和人物图像矩阵数据;根据车辆图像、人物图像、车辆图像矩阵数据和人物图像矩阵数据进行分析,判断车辆是否有载人的行为;若监测到有载人行为,则保存该图像,并发出报警信号。该方法对获取的车辆监控图像中用深层式残差卷积神经网络模型识别出车辆和人,并对车辆图像和人物图像进行分析,准确判断出车辆是否存在载人行为,能有效监测和预防车辆违规载人的行为,识别的准确性高,使用范围广。
Vehicle manned identification method and system, terminal and medium
一种车辆载人识别方法、系统、终端及介质
2020-12-08
Patent
Electronic Resource
Chinese
Vehicle illegal manned detection method and system, terminal and medium
European Patent Office | 2021
|Optical system for manned vehicle terminal guidance
Engineering Index Backfile | 1963
|