Object detection method and apparatus

By dividing the image sequence into sub-image sequences and performing background modeling, the problem of background modeling algorithms being affected by interference factors is solved, thereby improving the accuracy of object detection and the ability to detect static targets.

CN116342890BActive Publication Date: 2026-05-26LENOVO (BEIJING) LTD
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
LENOVO (BEIJING) LTD
Filing Date
2023-03-29
Publication Date
2026-05-26

AI Technical Summary

Technical Problem

Background modeling algorithms are easily affected by factors such as object movement or people standing still in image sequences, leading to errors in background generation and affecting the accuracy of object detection.

Method used

By dividing the image sequence into sub-image sequences, background modeling is performed based on the pixel feature information within the sub-image sequences, reducing the impact of interference factors on background recognition, and the background modeling algorithm is updated using an object detection model.

Benefits of technology

It improves the accuracy of background recognition and the precision of object detection, reduces background modeling errors, and enhances the ability to detect static target objects.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN116342890B_ABST
    Figure CN116342890B_ABST
Patent Text Reader

Abstract

This application provides an object detection method and apparatus. The method includes: obtaining an image sequence, the image sequence including multiple frames of images; determining pixel feature information of the images; dividing the image sequence into at least one sub-image sequence based on the pixel feature information of each frame of images, the sub-image sequence including at least one frame of images; performing background modeling based on at least one frame of images in the sub-image sequence to obtain a background image corresponding to the sub-image sequence; and detecting whether a target object exists in the images of the image sequence based on the background images corresponding to each sub-image sequence.
Need to check novelty before this filing date? Find Prior Art