China Subway Facial Recognition System Examined

By: Charles Rollet, Published on Jun 24, 2019

A China city of 6+ million people has installed facial recognition-enabled gates in subways, allowing commuters to enter stations by simply showing their face, thus bypassing ticket/card/phone use. How accurate and efficient is the system?

China Subway Facial Recognition System

In this note, IPVM examines the use of facial recognition for public transit in the PRC. We find that implementation problems persist chiefly due to the systems' use of identification instead of verification, and take a closer look at:

  • System Background & Scale
  • Who Supplies this
  • Purpose
  • How It Works
  • Identification, Not Verification
  • Effectiveness/Issues
  • Government Response
  • Human Rights Concerns

* ***** **** ** 6+ ******* ****** *** installed ****** ***********-******* ***** in *******, ******** ********* to ***** ******** ** simply ******* ***** ****, thus ********* ******/****/***** ***. How ******** *** ********* is *** ******?

China Subway Facial Recognition System

** **** ****, **** examines *** *** ** facial *********** *** ****** transit ** *** ***. We **** **** ************** problems ******* ******* *** to *** *******' *** of********************* **************, *** **** * closer **** **:

  • ****** ********** & *****
  • *** ******** ****
  • *******
  • *** ** *****
  • **************, *** ************
  • *************/******
  • ********** ********
  • ***** ****** ********

[***************]

**********

*** ******* **** **** fully ****** *** ** only *** **** ** far: ***** (***********.* *******) ***** ******** its *****-**** ****** ****** ***** ** **** year,**** ****** ***** *****. Here ** **** * Jinan ****** **** *** booth ******** ***** ****, as ***** ** * local *************:

*****

**** *** ********** **** also **** ************ *********************,*** ***** **** ***-*** trials **** **** ***** actually ***********.

Purpose, ***** ******** ******

******* ***** ***** ****** Xinhua*********** *****'* ****** ** more ********* **** *********** gates:

*** ****** *********** ****** enables ** ****** ** pass *** ******** **** every ******, ***** ***** traditional ******** *** **** only ** ****** *******.

***** ***** ***** *************** ****** ******* ***** **** ** the ****** ** "**** one ** * *******", an ******** *****.

******** ******** ******* **** not ***** ************ *** the ***** ****, ***** facial *********** ** ****** transport *** * ***** public ******** **** ** terms ** ******** ********: fooling * **** ***-***** ticket ****** ** ************* harder **** **** ****** an ********* ****** **** ticket, ** ***** *** transit ***** ** ******* commuter.

Supplier - ******

*** ******* ****** *** terminals ********, * ** ******/******** startup ***** ** ******** with ***+ *********, ********* to **********.

*************** ****** *********** ********* have ****** ******** *********, a****** ******* ** *** field.

**** ****, ********** ****** $*** ********* * ****** * round ** ********* *** by *******'* ** ***. Orbbec *** **** ****** at $* *******, ****** it *** ** *****'* AI ********. *** ******** have*********** ** * ********** to *********, ******* *******-****** facial *********** *******, ******** Orbbec ******* **** ******** on ** *******.

How ** *****

********* *** *** *** facial *********** ****** (***** is ********) ** *********** the ******'* "***** *****" app *** *********** ***** face ** ****** * selfie. **** ***** **** has **** ********, ********* then **** ** ** specific ***** ** *** subway ******* **** **** rec ********* ***** *** size ** ** **** fixed ** *** ***, then **** **** *** camera ** ** ******* in.

** ***** ** ********* ***** ******,*** ****** ** ***** integrated **** *** ******** subway *****:

******* **** *** ***'* face ** ********, ** theoretically, ***** ********* *** take **** ******* **** if **** ** *** have ***** ***** ** wallet. *** *** *** subway ***** *** ****-*** enabled: **** ***** *********** ********* * ** *** gates ******** **** *** system:

Identification, *** ************

*** ***** ****** ****** is ************* ********* **** other ***-**-**** ******* ** China. *** *******, ** is ******** ***** ** **** *****/*********** by ***** ***'* ****- *******, ************* ***** ***** ********* numbers *****,***** ** ******** **** to ** ******* ***** the **** *** ********** registered ***/*** ****. *** system **** ******** *** customer's **** ** ** existing *****, ** *:* matching ** **** ************.

****** *** *****/********** *******, Jinan's ****** **** *** require ********* ***'* ********* number ** *** ***** form ** ** - only ***'* **** ** needed. ***** ******* **** use**************, *** ************. **** means *** ***** ****** AI ** ******** ** identify ********* ******* **** thousands ** ****** ** the ******, ** '*:*' matching.

Identification ** * *** ******

************** ** **** **** difficult **** ************ ***** the ****** **** ******* the ********'* **** ******* many ****** ****** **** just ***.

**** *** ******** *** companies **** *:* ********, for *******,** *** **** **** year******* ***************** *** ****** *********** terminals ***** ******* ** to **,*** *****. ***** that *** **** *** a ******/**********, * ****** transit ****** *** **** hundreds ** ********* ** even ******** ** *****, a **** ***** ****** matching ******* **** ****.

*************/********

*** ****** ** **** Jinan's ****** ****** *********** solution *** *********** ******** which **** **** ******** its ******* *** ****, with ***** *** ******** remaining ************.

*** ***** ****** ******************* "*** **** *** didn't ******* **** * tried ** *** ** face ** *** **** the *******. **** **** it's **** **** * marketing *******." * ***** reporter*** * ****** *** ******, ******* verification **** ***** **** seconds, ************* ****** ****** ********* ** ***** ******* 30-40 ****** *** ****** (or *** ****** ***** 1-2 *******).

**** ** * ********** from *** ********'* ***** - *** ***** *** on *** **** "******** user *** *** *****":

******* ***** ** ********* mismatches. ********** ******** *************** *** **** ***** & ********** ** *** Jinan ***** *** *** found **** *** *********** issues ******* ** ***** or ***** ********* ******** will ** ***** ** the ***** **********:

***high ***** ** ****** ************ between twins, triplets, and [other identical siblings] may bring risks and payment issues like registration failure, abnormal ********, ******** **** *********, or multiple fare deduction, etc, which may bring damage to your personal properties. By clicking "Agree" to the Terms of Service, you *********** ********* **** *****. [emphasis added]

*** ***** **** ******** Jinan ***** **** *** liability ** ***** ** hacking ** ****** *******.

Social ***** *********

**** ******* ****** ***** users **** **** **** had ****** **** *** system, **********:

  • ***** ******** ***** **, I ***** **** **** faces *** *** **********. The *********** ****** ** very ********. ** *** don’t *** ********** *** 10 *******, ***'** **** up ** ***** **** card. (******)
  • *** **** *********** ***'* as **** ** ** school's ******* (******)
  • ***** **** *** **** to ** *** **** rec, *** **** ** take **** ******* *** or ** *****'* **** (******)

Government ******** ** ******

** ********, *** ***** subway *********** * ***** **************** ** ******* ********** about *** ****** *********** system. *** ******* ****** users *** ******** ***** faces **** **** ******/**********, along **** ************* ********** in *** ******** ****** image ******* ******.

***** **** *** ****** is ******* ** ******* learning *** *** ******* over ****, *** *********** that ***** **** * place **** ***** ***** with ** ***** ** record ***** **** *** avoid ****** *********.

Privacy/Human ****** ********

** ** ******* **, the *** ** ********* information ** *** ****** transit ***** **** ** even ****** *** ******** forces ** ***** ******'* movements. ** ********** *********, a ****-***** ****** ***** it ******* *** ****** to *** ********* ******* or **** ****** ***** cards. *** ***** *** used ****** ******* ************ for ************* ********:

  • ** ****, ******* ** the ****** ***** **** sect **** ********* ***** on ***** ************ ******** of **** ******* ******** on * ***, ********* to *********** ** ***** **** Freedom ***** (*** **** 127)
  • ****** *** ****** **** protests ** **** ****, protestors ***** *** ************ ******-**** ******* **** cash******* ** ***** ***** subway *****, ***** ***** be **** *** ******** purposes
  • ** ********, ***** *** PRC ** ********** * campaign ** *****************************, *********** ***** ****** recognition *************** ***** ********

*****, ***** ** ** indication **** ****** *********** for ****** ******* ** spreading ** *****, ** that ** **** ** made **********. ************, *** Jinan ****** ******* ******** voluntary.

**********

***** *** ***** ******'* use ** **** *** made **** *********, ***** remains *** *** **** using ****** *********** ** a **** ****** ******* solution ** *** *******, with *** ********* ************ involving *:* ******** ****** a *********** ******.

Comments (4)

** **** ***** ** Shenzhen *** *****'* ** overcrowded ***** *** **** queues, *** *** *** payment ******** *** *** security ***** ****** *** even *** **** *** stations.

** *** ********* *** of ****** **** ****** is ****** **** ** is ********* **** ******* to *** ** ****, and ** *** ******* time ** *** ********** in *** ****** ** time *****.

***** **** ******* ****** to **** * ******* work, ***** ** **** there ** ** ***** in *** **** *****. I ***** ******* *** objective ** *** *** speed *** *** ******** people.

***** *** * ****** Gates ** *** *******, not *** ** *** article ******.

***'** *****, ****** *** pointing **** ***. *'** corrected *** *******.

**** ** *** (*****) that ****** **** ****** up * ****** ? May ** *** ******* brothers ***** **** ** give **** * ***

Login to read this IPVM report.
Why do I need to log in?
IPVM conducts unique testing and research funded by member's payments enabling us to offer the most independent, accurate and in-depth information.

Related Reports

How China's Pay By Facial Recognition Works on Apr 02, 2019
Many social media posts have variously celebrated or warned about the growing use of facial recognition for payments in China. An example of one...
Hikvision AI Problems Criticized By Chinese Publication on Apr 09, 2019
Hikvision's facial recognition works poorly, causes delays and worsens learning, according to a new investigation by one of China's leading...
19 Facial Recognition Providers Profiled on Apr 23, 2019
IPVM interviewed 19 facial recognition providers at ISC West to understand their claimed accuracy, success and positioning. 9 from China, where...
IBM Shuts Down Intelligent Video Analytics on May 02, 2019
Abruptly, IBM has discontinued its Intelligent Video Analytics solution, it confirmed to IPVM. IBM's video analytics have faced recent media...
11 Facial Recognition Providers Review (Secutech) on May 09, 2019
Adding to our 19 Facial Recognition Providers Profiled report from ISC West, IPVM focused on facial recognition technology for our Day 2 coverage...
China Jaywalking Facial Recognition Guide on May 27, 2019
News reports touting the PRC's AI prowess often showcase facial recognition cameras being used to automatically catch and fine jaywalkers. In...
Facial Recognition Systems Fail Simple Liveness Detection Test on May 17, 2019
Facial recognition is being widely promoted as a solution to physical access control but we were able to simply spoof 3 systems because they had no...
UK Facial Recognition Essex Errors Report on Jul 05, 2019
Facial recognition trials in the UK have generated significant controversy and debate over the past few years. This week, it flared again when Sky...
China Uyghur Analytic Projects Require Intel And NVIDIA, Intel Condemns, NVIDIA Silent on Dec 02, 2019
At least 8 PRC China police projects require NVIDIA and Intel chips to power their Uyghur-detecting analytics, according to procurement documents...
Clearview AI Alarm - NY Times Report Says "Might End Privacy" on Jan 20, 2020
Over the weekend, the NY Times released a report titled "The Secretive Company That Might End Privacy as We Know It" about a company named...

Most Recent Industry Reports

Latest London Police Facial Recognition Suffers Serious Issues on Feb 24, 2020
On February 20, IPVM visited another live face rec deployment by London police, but this time the system was thwarted by technical problems and...
Masks Cause Major Facial Recognition Problems on Feb 24, 2020
Coronavirus is spurring an increase in the use of medical masks, which new IPVM test results show cause major problems for facial recognition...
Every VMS Will Become a VSaaS on Feb 21, 2020
VMS is ending. Soon every VMS will be a VSaaS. Competitive dynamics will be redrawn. What does this mean? VMS Historically...
Video Surveillance 101 Course - Last Chance on Feb 20, 2020
This is the last chance to join IPVM's first Video Surveillance 101 course, designed to help those new to the industry to quickly understand the...
Vulnerability Directory For Access Credentials on Feb 20, 2020
Knowing which access credentials are insecure can be difficult to see, especially because most look and feel the same. Even insecure 125 kHz...
AI/Smart Camera Tutorial on Feb 20, 2020
Cameras with video analytics, sometimes called 'Smart' camera or 'AI' cameras, etc. are one of the most promising growth areas of video...
China Manufacturer Suffers Coronavirus Scare on Feb 20, 2020
Uniview suffered a significant health scare last week after one of its employees reported a fever and initially tested positive for coronavirus....
Cheap Camera Problems at Night on Feb 19, 2020
Cheap cameras generally have problems at night, despite the common perception that integrated IR makes cameras mostly the same, according to new...
Milestone Launches Multiple Cloud Solutions on Feb 18, 2020
Milestone is going to the cloud, becoming one of the last prominent VMSes to do so. Milestone is clearly late but how competitive do these new...
Video Surveillance Architecture 101 on Feb 18, 2020
Video surveillance can be designed and deployed in a number of ways. This 101 examines the most common options and architectures used in...