Traffic counting method and equipment

Abstract

A traffic counting method and traffic counting equipment are disclosed. The traffic counting method comprises the steps of: recognizing first and second regions of interest (ROI) in a traffic view including moving objects; acquiring first image data representing the first ROI and second image data representing the second ROI; analyzing the first and second image data along with time; and counting the moving objects and determining moving directions of the moving objects based on the analysis of the first and second image data.

Description

technical field [0001] The present disclosure relates generally to computer vision techniques, and more particularly to traffic counting systems and methods. Background technique [0002] Traffic counting refers to counting the number of moving objects moving in a specific direction within a given period of time. For example, an object can be a person or a vehicle. Traffic counting has various applications, such as determining the number of people entering and / or exiting a building, the number of passengers entering and / or exiting a bus or train, the number of vehicles entering and / or exiting a parking lot, Number of pedestrians / vehicles moving on, etc. [0003] In recent years, flow counting has gained enormous attention in several industrial and service sectors. For example, consumer businesses (eg, retail stores, supermarkets, etc.) are increasingly aware of the importance of monitoring the number of visitors to their establishments. Consumer traffic data can be analy...

AI Technical Summary

Problems solved by technology

However, humans are prone to error, fatigue, and cannot monitor traffic without interruption

For example, workers may miscount in crowded places

Also, labor is expensive, so traffic counting is usually performed in an ad hoc way

The data scope may only cover short time spans, such as hours or days, and does not provide insight into long-term trends in traffic

As a result, manually obtained data often lack accuracy and detail

Images