Retention rate statistical method based on human head detection
The invention discloses a retention rate statistical method based on human head detection, and belongs to the field of visual target detection and tracking, and the method comprises the steps: obtaining a video image containing a human head; performing target positioning detection on the human head...
Saved in:
Main Authors | , |
---|---|
Format | Patent |
Language | Chinese English |
Published |
01.09.2023
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | The invention discloses a retention rate statistical method based on human head detection, and belongs to the field of visual target detection and tracking, and the method comprises the steps: obtaining a video image containing a human head; performing target positioning detection on the human head in the video image to obtain a detection frame; performing feature extraction on the detection frame to obtain a target feature, and adding an identity (ID) to the target feature; adding the target features into a tracking list, converting detection information in the tracking list into a retention rate statistical table, and calculating through the statistical table to obtain target detection times; and based on the frame rate of the video image and the target detection frequency, calculating target residence time, setting a time threshold, and based on the target residence time and the time threshold, calculating the residence rate of the target in a certain time period. The technical problem that in the prior ar |
---|---|
Bibliography: | Application Number: CN202310499967 |