目标监控方法、装置、计算机设备及存储介质

本申请公开了一种目标监控方法、装置、计算机设备及存储介质，包括：获取第一媒体信息；利用深度神经网络识别第一媒体信息中的所有目标物的身份形成身份集合，根据身份集合确定目标物的数量N；获取第二媒体信息；对第二媒体信息中的目标物进行去重处理；从去重处理之后的第二媒体信息中确定待追踪目标物的身份以及待追踪目标物的数量M；根据所有目标物的数量N和所述待追踪目标物的数量M判断是否有目标物跟踪丢失；若有目标物跟踪丢失，则根据待追踪目标物和所述身份集合确定跟踪丢失的目标物身份。上述方式不仅体现了目标监控的便捷性和智能性，还可以大大提高数量统计的效率。 The invention discloses a target monitoring method and device, computer equipment and a storage medium. The target monitoring method comprises the steps: obtaining first media information; identifyingidentities of all target objects in the first media information by utilizing a deep neural network to form an identity set, and determining the number N of the target objects according to the identity set; obtaining second media information; performing duplicate removal processing on the target object in the second media information; determining the identity of a to-be-tracked target object and the number M of the to-be-tracked target objects from the second media information after the de-duplication processing; according to the number N of all th