Conference abstractPublished 22 Nov 2022
A Humanand#39;s Heat Signature and Background Subtraction Hybrid Approach For Crowd Counting and Analysis
Educational Leadership International an educational consulting firm, United States
Abstract
Select any words in the title or abstract, or press this and choose a sentence.
A new method was developed based on human thermal and motion characteristics for crowd counting and classification. An important technique for automated crowd estimation and type determination is described. Without any camera calibration or assumption of prior information about the input images, the proposed methodology estimates crowd density and no human interference is required so that it can be used effectively in a fully automated crowd control system. For crowd counting purposes, two new features have been proposed: the first reflects human thermal characteristics and is represented by the ratio between their temperature and the temperature of their ambient environment. The second defines the features of human motion and is determined by the ratio between the velocity of human motion and the rigidity of the ambient environment. For human beings, each ratio should surpass a certain predetermined threshold. Improved crowd counting efficiency is achieved using these investigated characteristics and less time was taken. In addition, the two features are combined and used together for one of three key forms of crowd classification, which are: fully mobile, fully static, or a combination of both types. There are also several beneficial features of the proposed system as it is, privacy protection of crowd counting system, effective for homogeneous and in-homogeneous crowds, does not rely on any direction in motion detection, does not require a particular crowd size. The experimental findings show the approach’s efficacy.
Subjects
Computer Science & ITAsk about this article
Answers come only from this article's title, abstract and keywords, each sentence pointing to where it comes from. Your question is sent to OpenAI to write the answer. Scholars Central does not store it.
Follow
New articles from what you follow go into your email digest, once you agree to alert emails.
- JournalAdvances in Robotics & Automation
- SubjectComputer Science & IT
- BylineArticles whose byline prints “Mayank Khurana”
