Content recognition method, system and storage medium thereof
Abstract:
A system may obtain a first recognition record file. The first recognition record file may include recognition records. The recognition records may include respective coordinates for a monitored video received by a monitoring device. The respective coordinates may be indicative of a specified event previously recognized based on the monitored video. The system may determine a monitoring boundary based on the respective coordinates of the recognition records for the monitored video received by the monitoring device. The system may receive at least one frame of subsequent monitored video. The system may extract in response to receipt of the at least one frame of the subsequent monitored video, image data from the at least one frame of the subsequent monitored video. The extracted image data from the at least one frame may be located in the monitoring boundary as applied to the at least one frame of the subsequent monitored video.
Public/Granted literature
Information query
Patent Agency Ranking
0/0