IoT Camera Management via Audience Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge in managing camera devices for public landscapes in smart cities is that a large number of scenic spots exceed the capacity of available camera collection devices, leading to unsatisfied visiting needs of citizens, necessitating an efficient method to optimize camera device management.
Innovation Solution
An IoT system with a cloud platform is implemented, comprising user platforms, a service platform, and a management platform, using machine learning models to predict audience counts and determine which camera devices to cancel or adjust based on audience demand, ensuring optimal camera device allocation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If the count of camera devices is increased to cover more scenic spots, then the visiting needs of citizens can be satisfied, but the device complexity and management difficulty increase
Solution Approach 1:
The system enables self-service through automated audience count prediction using machine learning models. The service platform automatically predicts future audience counts and generates cancellation recommendations without manual intervention, allowing the system to manage itself dynamically based on predicted demand
Solution Approach 2:
The system implements feedback loops where historical audience data is continuously fed into machine learning models to improve prediction accuracy. The management platform receives feedback from the service platform about predicted audience counts and adjusts camera device allocation accordingly, creating a closed-loop control system
2Device complexity
If the count of camera devices is reduced to simplify management, then the device complexity decreases, but the visiting needs of citizens cannot be satisfied
Solution Approach 1:
The system transforms static camera device allocation into a dynamic process by using machine learning models to predict future audience counts. Camera devices are dynamically added or canceled based on predicted demand, allowing the system to adapt to changing conditions without permanent over-provisioning
Solution Approach 2:
The system changes the parameter of camera device allocation from fixed to variable by introducing audience count predictions. The management platform adjusts the number and location of camera devices based on predicted audience parameters, optimizing coverage while managing complexity
3Productivity
If camera devices are allocated statically without prediction, then the device complexity is low, but the resource utilization efficiency is poor
Solution Approach 1:
The system performs preliminary action by predicting future audience counts before making camera device allocation decisions. The service platform forecasts audience numbers in advance, allowing the management platform to proactively adjust device allocation to match upcoming demand
Solution Approach 2:
The system replaces manual or mechanical decision-making processes with machine learning-based prediction models. Instead of static allocation rules, the system uses computational models to automatically determine optimal camera device placement based on predicted audience behavior
Data Source
AI summary
The present disclosure provides a method and an Internet of Things system for managing a camera device of a public landscape in a smart city. The method is implemented based on the Internet of Things system, the system including a plurality of user platforms, a service platform, a management platform, and a plurality of object platforms. The method includes: counting, based on the service platform, a count of audience of the landscape images corresponding to the different user platforms in a preset future duration, and sending the count of the audience to the management platform, wherein the count of audience of the landscape images corresponding to the different user platforms in the future duration is determined through processing the count of the audience of the landscape images in a preset historical duration based on a third prediction model, and the third prediction model is the machine learning model; and determining, based on the management platform, a camera device to be canceled, the camera device to be canceled being a camera device corresponding to a landscape image whose the count of the audience does not satisfy a preset condition.