Mobile Video Call Quality Measurement Using OCR Timestamp Overlay

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current video call quality measurement methods, such as the Perceptual Evaluation of Video Quality (PEVQ) standard, face difficulties in measuring actual video call quality on mobile terminals due to limitations in comparing video frames and requiring modifications to the terminal, and are not suitable for calls lasting over 20 seconds.

Innovation Solution

A method and system that insert a sequentially increasing index and timestamp into video frames captured by a mobile terminal's camera, allowing for accurate quality index calculation based on transmission parameters like delay, jitter, and freezing, without needing a reference video source, using OCR text recognition and a terminal agent with a modified video capture function.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If the PEVQ standard is used to measure video quality by comparing with a reference video source, then measurement precision is improved, but device complexity increases and it cannot measure actual video calls through mobile terminal cameras

Engineering Contradiction:
Improvevideo quality measurement precisionVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts only the essential timing information (timestamp and sequence number) from video frames and transmits it as overlay text, rather than transmitting entire video frames for comparison. This reduces the measurement system to its core function of tracking frame delivery timing, significantly simplifying device complexity while maintaining measurement precision through OCR recognition of the extracted timing data

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

Instead of comparing actual video frames with reference frames, the patent creates a simplified copy of timing information by embedding timestamp and sequence number data as visible text overlays in the video stream. The receiving end OCR-scans these text overlays to reconstruct timing information, avoiding complex frame-by-frame comparison while achieving accurate quality measurement

Inventive Principle:
Principle #26Copying

2Measurement precision

If the PEVQ standard with 6-20 second reference video is used, then measurement precision for short videos is improved, but adaptability to video calls of any duration deteriorates

Engineering Contradiction:
Improvemeasurement precision for short videoVSAvoidadaptability to video call duration
Core Design Contradiction:
Measurement precisionVSAdaptability or versatility

Solution Approach 1:

The patent implements a dynamic measurement system where timestamp and sequence number overlays are continuously embedded in real-time video streams of any duration. The receiving terminal continuously OCR-scans these dynamic overlays throughout the entire call, allowing accurate quality measurement whether the call lasts 10 seconds or 10 hours, thus achieving full adaptability to varying video call durations

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent performs preliminary embedding of timestamp and sequence number information into video frames at the transmitting end before transmission. This pre-prepared timing information is always available in the video stream regardless of call duration, enabling the receiving end to immediately begin quality measurement without waiting for reference videos or setting up complex comparison protocols

Inventive Principle:
Principle #10Preliminary action

3Measurement precision

If terminal modification is required to implement video call quality measurement, then measurement precision is improved, but ease of operation deteriorates

Engineering Contradiction:
Improvevideo call quality measurement precisionVSAvoidease of use
Core Design Contradiction:
Measurement precisionVSEase of operation

Solution Approach 1:

The patent makes the video call application itself self-measuring by embedding timing information directly into the video stream it generates. The application automatically performs OCR scanning of its own transmitted video frames to extract timing data for quality measurement, eliminating the need for external measurement tools or terminal modifications while maintaining high measurement precision

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent makes the video call application multi-functional by integrating both video call functionality and quality measurement functionality into a single application. The same application that captures and transmits video also embeds timing overlays and performs OCR scanning, allowing users to make video calls with automatic quality measurement without any additional software or terminal modifications

Inventive Principle:
Principle #6Universality (Multi-functionality)

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

Enables simple and accurate measurement of video call quality experienced by users on mobile terminals, supporting calls of any duration without modifying the terminal or network, and providing a quality index based on transmission metrics.

Implementation Method 1

extracting the index and the timestamp from the stored video frame through an optical character reader (OCR) text recognition

Methodology Applied
Scientific EffectOptical character recognition:

Data Source

PatentUS10271044B2Method and system for checking video call quality of mobile terminal
Publication Date: 2019.04.23 NAVER CLOUD CORP
  • US10271044B2 patent drawing
  • US10271044B2 patent drawing
  • US10271044B2 patent drawing

AI summary

Disclosed is a method of measuring a video call quality of a mobile terminal, the method including establishing a video call between a transmitting terminal and a receiving terminal; modifying a video capture function provided from a terminal operating system (OS) and inserting a sequentially increasing index and timestamp into each video frame that is automatically acquired through a camera every time a video call application (app) of the transmitting terminal calls the video capture function; transmitting, by the transmitting terminal, the video frame to the receiving terminal; storing, by the receiving terminal, the video frame at a ratio higher than a frame per second (FPS) of the transmitting terminal; and extracting the index and the timestamp from the stored video frame through an optical character reader (OCR) text recognition and calculating a quality index.