Camera-Based Viewer Recognition for TV Audience Rating Accuracy

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for measuring TV audience ratings and advertisement effects rely heavily on viewer cooperation, which can lead to inaccuracies due to reliance on button inputs and infrared sensors, and lack real-time data analysis capabilities, resulting in differences between actual and measured ratings.

Innovation Solution

A system utilizing a viewing behavior measurement device that captures viewer images to recognize and analyze viewer behavior, generating and transmitting viewing behavior information only when changes occur, and an advertisement effect measurement server that analyzes this data to provide accurate, real-time advertisement effect measurements.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a people meter with button input is used to collect viewing behavior information, then viewer cooperation is required to input viewer information, but this leads to inaccuracies in audience rating measurement due to reliance on viewer cooperation

Engineering Contradiction:

Improveaudience rating measurement accuracyVSAvoidreliability of measured data

Core Design Contradiction:

Measurement precisionVSReliability

Solution Approach 1:

The patent replaces the mechanical button-input system with an automated image recognition system using cameras and facial recognition algorithms. The system automatically detects viewers' presence, identity, and viewing behavior without requiring manual button presses, thereby eliminating the reliability issues associated with viewer cooperation while maintaining measurement precision.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service measurement by automatically capturing viewer information through cameras and performing facial recognition without human intervention. The measurement system serves itself by autonomously collecting, processing, and transmitting viewing behavior data, removing the need for viewers to actively participate in data collection.

Inventive Principle:

Principle #25Self-service

2Speed

If data is collected and transmitted continuously at high frequency to achieve real-time analysis, then real-time advertisement effect measurement is improved, but data transmission volume and processing load increase significantly

Engineering Contradiction:

Improvereal-time data analysis speedVSAvoiddata transmission volume

Core Design Contradiction:

SpeedVSQuantity of substance

Solution Approach 1:

The patent extracts and transmits only the essential viewing behavior information and key changes in viewer state rather than continuously transmitting all raw data. By selecting and transmitting only relevant data points (such as viewer presence, identity, and significant behavior changes), the system achieves real-time analysis capability while minimizing data transmission volume.

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The system performs partial data collection and transmission by focusing on critical viewing events and changes rather than continuously monitoring and transmitting all parameters. This selective approach provides sufficient real-time information for advertisement effect measurement without the excessive data load of complete continuous monitoring.

Inventive Principle:

Principle #16Partial or excessive action

3Extent of automation

If infrared sensors are used to detect viewer presence, then automated detection is improved, but the system fails to accurately detect viewers who are sleeping or not staring at the television

Engineering Contradiction:

Improveautomated viewer detectionVSAvoidviewer presence detection accuracy

Core Design Contradiction:

Extent of automationVSMeasurement precision

Solution Approach 1:

The patent replaces infrared sensor-based detection with camera-based image recognition technology. The camera system captures visual information about viewers' presence, facial expressions, and attention direction, providing more accurate detection of actual viewing behavior including whether viewers are sleeping or not watching the television, thereby improving measurement precision while maintaining automation.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS8887186B2Analysis method and system for audience rating and advertisement effects based on viewing behavior recognition

Publication Date: 2014.11.11 ELECTRONICS & TELECOMM RES INST

AI summary

Provided are an analysis method and system for audience rating and advertisement effects based on viewing behavior recognition including a method of recognizing a viewer on the basis of image recognition through a camera and efficiently collecting data and a method of transmitting collected data on the basis of viewer recognition to an audience rating and advertisement effect measurement server. The present invention includes a viewing behavior measurement device configured to receive a viewer image obtained by imaging one or more viewers who are viewing broadcast content, recognize the one or more viewers from the received viewer image to generate viewing behavior information, and deliver the generated viewing behavior information, and an advertisement effect measurement server configured to analyze an advertisement effect for each viewer on the basis of the received viewing behavior information and generate and deliver an advertisement effect measurement result.