Object detection method and device, computer device and storage medium

By acquiring object behavior features from the target application, utilizing a probabilistic prediction model and applying corrections, the low accuracy of object abnormal behavior detection in existing technologies is solved, achieving higher detection accuracy.

CN122097974APending Publication Date: 2026-05-29SHENZHEN TENCENT COMP SYST CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
SHENZHEN TENCENT COMP SYST CO LTD
Filing Date
2024-11-28
Publication Date
2026-05-29

AI Technical Summary

Technical Problem

In existing technologies, methods that rely on business experience to manually define detection rules to detect abnormal behavior of objects have low accuracy and are prone to misjudgment.

Method used

By acquiring the behavioral characteristics of objects in the target application, a probabilistic prediction model is used to predict whether the objects have abnormal behavior, and the prediction results are corrected based on the behavioral characteristics to determine whether the objects have abnormal behavior.

Benefits of technology

It improves the accuracy of object detection, reduces false positives and false negatives, and enhances the precision of abnormal behavior identification.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122097974A_ABST
    Figure CN122097974A_ABST
Patent Text Reader

Abstract

The application discloses an object detection method and device, computer equipment and a storage medium, and belongs to the technical field of computers. The method comprises the following steps: acquiring a behavior feature of an object in a target application in the target application, wherein the behavior feature is used for reflecting the behavior of the object in the target application; inputting the behavior feature into a probability prediction model to obtain a first probability of the object, wherein the probability prediction model is used for predicting whether the object in the target application has an abnormal behavior in the target application, and the first probability refers to the probability that the object has an abnormal behavior in the target application; correcting the first probability based on the behavior feature to obtain a second probability of the object; and if the second probability of the object is greater than a preset threshold, it is determined that the object has an abnormal behavior in the target application. The application corrects the deviation of the probability prediction model based on the behavior feature of the object, and improves the accuracy of detecting the object with an abnormal behavior.
Need to check novelty before this filing date? Find Prior Art