Active Terahertz Imaging For Weapons Screening Analyzed
Today's walk-through metal detectors cannot visibly see items underneath closing; rather, they use low-frequency electromagnetic radiation to attempt to detect metal that may be weapons.
An alternative approach, terahertz imaging, aims to see items, including weapons, underneath clothing. In particular, a Silicon Valley startup, Cover, is marketing this.
In this report, based on an analysis of several patents and scientific papers, we analyze the physics behind the approach, the advantages and disadvantages of using active terahertz imaging, operational considerations, and several fundamental limitations.
For more details on other approaches for weapons screening and how various screening devices work, see our previous tutorials:
- 4 Main Weapon Detection Approaches Explained
- Smart Metal / "Weapon" Detectors Tutorial
- How Conventional Metal Detectors Work
- How X-Ray Weapon Scanners Work
- How Millimeter Wave Airport Body Scanners Work
Active ********* ******* ***
****** ********* ******* *** ********* ** LiDAR/radar ** **** **** *************** ***** into *** ***** ** **** *** analyze *** ********* ********* ** ****** a ***** *** (*.*., *** *** is * ***** ***** ** *** field ** **** **** *** ********).
****** ********* ******* ** ******* ** cameras **** *****-** ***** ***** ****, as **** ****** ****** ** ******* in *** ***** ** **** ***** illuminating **** **** **** **** ** electromagnetic *********. ***** ** *** *********** distinction, *******, ** ******* ***** ******* do *** **** ***** **********, ******* active ********* ******* ******** *** ********* to ********, ********* *******, ***.
*** ********* ** ********* ********* ** that ** *** ********* **** ******** materials, ******** ** ** ******* ********* of ********* *******. ********* ******* *** create ***** **** **** ***-**** ********, allowing ** ** *** ******** ** weapons. *** *******, *** * ** range ************** ** * ****** ********** a ****** ********** * *-*****:
Advantages *** ***********
***** ****** ********* *******/***** *** ******* advantages **** *** **** ************ ******* terahertz ******** (** ***** ********** *** lack ** *********** **********) *** *** detect ***-******** ****** ******* (****** ********-***** devices), *** ******** *** ******* *********** technological *********** (**** ** **** *** shared **** ******* ********* *******) *** implementation **************, *********:
- ****** ********* *** ***** ****; ****, ****** *** ******* *** required *** **** ********** (** ********** one *** * **** ****), ** visitors ***** **** ** **** *** rotate *** * ****. *** ****** increases *** ********* ** *** ********* area *** *** **** ** *** device, *** ****** ********** ******** ** the *********.
- ******* ** **** ******* * **** more ******* ****(*** ** *** **** ** ************* between ******* *** *******), *** ***** is ******* *******/********** ** ********* **** for *** *********. **** ******* * need ** **** ***** ******* ** scan ****.
- ******* ******** ********* *** ***********(*** *******, ******* *******), ** **** typically ****** **** ** *** ********* emissions ****** ** ******* *** ******, thus ********* ******* **** *** ********.
- *** ********, **** ********, ** **** can ********** ************ ********* ******* **, *** ** the ********* ********* ** ***** *********, they *** ****** ********* ********* (***** essentially ***** *** ******** ****** ** terahertz).
- ***** ***** *****: ***** ****** ********* ******* *** theoretically **** ***** ***** ******* ** cameras, *** ******* ***** **** ********* in ********** ******/******* ** * **. Such * ***** **** **** ***** performance ** ****** *******.
- **** ************ ******* **** ******* **** (*** context, ************ ********* ******* ********* ******* are ****** ** $***,*** *** ******) and, ** *********, *** **** ********* than ***** ********* *** **** "******* detectors."
- ********** *******: ***** * *** ******* ***** such *****, *** ***** ** ******* on *** ***, **** ******** ********* have ********* *** *******, *****, ** and ***** **** ***** **, ********* production **********, *****, *** *********** *****.
** *** ******** ****, *** ********* imaging, ** *** **** *** *********** improving, ***** *** *** ********* ** automate ******* ********* *** ********* *** need *** ** ******** ** **** at *****.
Emerging ********
** *******, ****** ********* ******* ** not ****** **** *** ******* *********, as **** ***** **** *** ******* offering **** * ****** —*********. *** *******, *******, ******* ** parcel ********, *** ****** ******* ******** limited *****. ** **** ********* *** company *** *** ******** * ********* post ** *** ******, ********* ** how **** ** *** ***** ***** that.
************, **** ******** ******** ****** ****** attention, ** *** ************ ******* * *****-**(** ***** ** ******* $** ******* of *** *****) ** ** ******* to ************* ****'* ********** ** ***** on "********** ****** *********." ** ******* to ** ****, *******, ** *** limitations ** ****** ********* ******* **** be ******** *** ** ******* ** the ** **** ****** *** **.
Terahertz ***********
********* ***** *** *************** ***** **** frequencies ** *** ***** ** *** GHz ** **** *** (** * THz), ***** ** **** ********** ****** than **-** (*.* ***/*.****), *** **** wavelengths **
********* *********** (**** *** *** ** 900 ***) *** * ******** ****** for ********* ******* ********* ** (*) they *** ***** ***-**** ********** ** large ********* (****+), *** (*) **** have *** *********** ************, *** ******* clothing ********* *** *********** *********** ** radiation, ** *** ********* ** ******** scientific ****** (*,*):
******* ***, **** ** **** ******** is ******* **** ******** ****** ********** a ******compromise ******* ******* ********** *** ***********/********* *********** [Petkie, Douglas T., et al. 2008] [emphasis added]
*** ***** ******* ** * ********* technology *** ******** ******* ** ******-***** concealed ******* .... *** ********* **** around *** *** *** ****** ******* it **** ** *trough ** ** *********** *********** ******, while providing sufficiently **** ******* ********** for a favorable tradeoff between antenna size and standoff range. [Cooper, Ken B., et al. 2011] [emphasis added]
4 *********** *******
********* *********** ********* * **** *****, but ****** **** *****, ***** *** four ******** ******** ***** **** ******* the ****** ***********, **** ** ********* obstruction **** *********** *********** (**********, **********) by ***** ****** *** ****** *********, as**** ********** ******:
***** *******, ** ****, ***** ******* trade-offs ** ****/********** ** *** ********* device, *********** ******* ********, *** *********** of *** *******:
******** ~*** *** *** ~*** *** windows — *** *** **** ****** mentioned ********* ****** *** ******** *********. According ** **** (*** ***** ******* above), (*) *** ~*** *** *********** penetrate ***** ****** **** *** *** but **** ***** **********, *** (*) devices ***** ~*** *** *** **** expensive *** **** ********* **** *** ones ***** *** ***.
**** *** **** ******* ******** ** the ~*** *** *** ~*** *** ranges *** ******** *********, *** *** company, *********, **** ~*** *** *********** for ***** ******* (***** ********** ******** the ****** *********** *** *** **** limited **********).
Terahertz **. ***** ****
***** ********** ** *** ** **** of * ******* *** ********* ******* (as ***** *****), ** *** ******* penetration ***** **** *** ***** **** due ** *** **** ***** *******, which ******* ********* (******* ** *** specific *************** ** ***** *********). **** limitation ** ****** ** **** ******* devices ** ******** ********, **** *** exception ** ************ *-*** ********.
** ********, **** ***** *** ******:
(*) ** * ****** ** *** fully ******* ** *** ******** ***, for *******, ******* ** * ****, a *********** ****** ** ********* ********* will ** ******** ** *** ****, making ********* **** *********.
(*) ********** ***** ****** ********* ******* will **** ** ****** *** ******** to ****** ** ****** ***** **** to *** ******* ** **** *** scanners (** ********** *** *** * back ****). *** ****** ******* ********* friction, *** *** ****** ********* *** footprint *** *** ********* *** *** cost ** *** ******.
Absorption ** ********* ******** *********
******* ********** ** ******, *********, *** denim ** * ********* ******* ** terahertz *******, ** ** ****** ********* wavefronts ** **** ******* **** ****** clothing *****. *******, ***** *********, **** as ******* ** ****, **** ***** transmission ************ (*.*., ***** ***** **** go *******, ******* **** ***** ***** will ****** *** * ********* ******).
* **** ******** ********** ********** ** * *** *** *****, leather ******* ***** ****** **%+ ** incident ********* *********:
*********,**** ********** *************** ********* ******** ********* ****** *** resulting ****, **** ******* ********* *** multiple ****** ********** *** *****:
Terahertz **. *****
********** ** ********* ********* ******* ******* on ***** *******, *** ***/**** ******** can ** ****** ** ********* *****, obscuring *******. **** ******* ******* ** the ********* ********* ** ***** ********* and ***** ******** ********* *** **********.
***** ** ****'* *****, *** ******** ***** **** * mannequin, *** ** **** ** ***-*********** to *** ******** *********, **** ******* it **** * *** *-*****:
* ********* ******* ** ***** * plastic ********* ** **** ** ** highly ************ ** *** ***, ***** humans *** ******* ******.Thus *** ********* *** ******* **** * **** ****** ***** (***** ** ********* ********* ** *** ***), which in turn was covered by plastic wrap to prevent the outer clothing from becoming moist. [emphasis added]
********* ********* ********** ****** ********, ****, fog, *** **** ******** *** **** negatively ****** ******* ********* ***** ********* imaging.
Range ********
******** ********* ***** **** **** ** pass ******* *** ******** *** *** reflection **** *** ****** ******* *** detector, ***** **** ****** ********, *** would *** * ***** ***** *** a ***** ******** ** *** ******, as**** ********** ******. *** ***** ** *** ******** signal ** ******* ** *** ******** axis, *** *** ***** ** ******* on *** ********** ****.
*** ****** ***** **** ******** ***** because ***** **** ** *** ********* emission ****** ******* ******-***** ********, **** of *** ***** ******* **** ** the ********. *** *****-**** **** *********** to *** **** ** * ********* weapon. ***** ******** *** ****** ******, one *** ****** * ** ***** map *** *** ** ** **** unnatural '******' ************* ** ********* *******:
***** ** *************** *** *********, ** practice, ***** *** ********* ** *****-***** 2D ******** *** **********, ***** ********* colors *** **** ** ******** ********* ranges, ** *** ***** ** ******** papers:
Privacy ********
***** ***** ** ******* ********** ** privacy ******** **** ********* *******, *** can ***** **** ***** ****** ** one ******* ********* ******* *** **** the ******** ** * ***** **** with * ***-**** ********.
**** ** *** ********** ****** ** patents **** **** ******** ****** ***** this, **** ******** ********** ******* ********** **** *******:
**** ** *** ***** ***********. ** your **** ******, *** *** ***, this ** *** ***** [**** *** pipes], *** ** ***** ****** *** bloody ******* *** *****. *** *** can **** ***, *** *** ******* through *** ******* ** **** ** you *** *** * ***.Sometimes ** *** *** ****. ****'* *** ** *** ******* [emphasis added]
**** **** ** *** ** *** tests ********* ** ********** ****** **** IPVM *******, *** ***** ******* **** exclusively ****, *** *** ****** **** was ****** ******* ** *** *****.
*** *******, ***** ********** **** **** scanners *********** *** **** *******, *** after ***** ******** **** ****** ** the ******, *** ************* **** ****** to **** ** **** * ********** - ************ * ******** ***** **** with * ****** ** * ***** being:
*** **** ******* ** ********** **** body ********, ****** ********** **** ******* **** ******** Work.
**** ******** **** **** ***** ** resolved ********** ******, *** ** ** something **** ********* ** *** ***** should ******** *****.
AI/CNN *** ****** ***********
*** ****** ************ ** ** *** CNN **** **** **** ****** ***********, so ** *** ** **** ************* and *** ******* * ***** ********. The ********* ** ******** ***** ** the ****** ** ***-**** ******* ****** out ** * ***** ****, ***** is **** ** **/*** ** ****** them.
Detecting ******* **** **** ******** ******
**** ***** **** ***, **** ** marketing ********* *** ********** ******, ******* detection ** *****, *** ***** ** is * ***** *** ****, **** believes **** ********* ******* **** **** distinct ****** ** * **** ******* task *** ***** *******.
**** * ******* ****** ** ********** terahertz *******, ** ** **** ** definitively ***** *** **** **** ******* on ****** ** **** *****.
Detecting ******* ** *********
*** ********** ****** *** ******* **** reviewed ****** ********* ********* ******* ** backpacks *** ****. **** ******** **** the *********** ******** ** ***** ********** for ******* ********* ** *** ******** for ********* ******* ** ********* ****.
** **** ***** ** ******* *** be ******** ** ******, *******, *** other ******** ******* ** ****, ********* these ******* **** ** ************ *********.
************ *********** **** **** *** ******* implications *** *******, *** ** *** main ******* *** ******* *********, ** school ********* ***** **** ** *** other ********* ************ (*** *******, *-*** bag ********), ***** **** (*) ******** the ******* **** ** *** **********, and (*) **** **** *** ********* process.
Differentiating *********
******* ****** ****** ********* ** ********** papers ** *** ******* ** ************* the ********* ** ********* ******* (*.*., a ******** *** **. * ******* toy ***). ******* ****, ******** ** * ******** *********** *****-**, Cover, ****, "* ** ***** **’** ******** be **** ** ****** *** ********** between * ****** *** *** * [hand]gun."
***** **** ** * **** ********* aspect ** ***** ******* **** ********* weapons ** *********, ** *** ***** provide ******** ******* ** *********.
Video-Frame ****
***** ******* ********** ********* *****-**** ***** rates **** ****** ********* *******, *** highest ***** **** **** *** ******** implemented ** ********** ****** **** ** IPVM *** * **, ***** ** equivalent ** **** ***** * ****** and ** ~** ***** **** **** conventional **** ******* (** **). ** remains ** ** **** ** ****** frame ***** **** ** ********, *** if ***, **** **** ******* *********** limitations **** * ********* ** ****** objects.
*** *******, *** * ************* ** NASA ********** ** * ****** ****** being ******* ** *** ****** **** a * ** ***** ****:
Image **********
*********, ***** *********** ** ********* ***** are ***** **** ***** ** ***** cameras, *** *** ***** ****** ** pixels *** **** ********* ** *** manufacturer/setup. **** **********, *******,*********** *** ********* * ******* ********* on * ***** ***** ** ***** 1,836 ****** *** ****** (***** *** pixel *********** ** * *.*" * 0.4 ******):
******** * *********** ***** ******* *** handgun ***** ******** ** *** ** subjective, ** ******* **** * ************ estimate *** *** ******* ***** ****** needed ** ****** * ******* **** good ********** ** **** **** ** N_tot = ** × **=****, ************* to ***. **, ***** *** ***** density ** ***** * **^(-*)
**** **** ** **** ****, *** target **** *** ******* ** * human *****, ** **** ****-**** *****, one ***** **** **** ******. ** general,*********** ************* *** ****** ** ****** ******* on ******** *******:
*** ******* ****** ** ***** ****** required *** * *** ******* ***** to ** ********* ******* ** ****** factors **** ** *** ******** *** size ** *** ****** ***** ********, the ******* ***********, *** ***** **** spot ****, *** ******** ***** ** view, *** *** ******* ***** *******.
*** *******, ***** *** ********* ***** of * ******* **** ********* ***** resolutions:
Active ********* ** ******* *********
******** ********, ***** ********* ********* *** security ********* ** *** * ***** idea, ** ** *** **** **** for **** **** * ****** ***. However, ** *** **** ***********.
******* ********* ********, ***** *** ************ available, ** *** **** ********* ********** (unlike ****** *******) *** *** ********* emitted ***** ** ******* ** *** field ** ****. ****** ***** ** passive ******* **** ******** *** ** the ********** ** ************ ******* * human **** *** * ********* ****** and *** ** *** ********* ******** properties ** ********* *********.
************, ****** ******* ** *** ****** on *** *********** ********** ******* ** object *** * **** (* ********** of *** ******* ********) *** *** offer ****** ********** **** ******* *******. On *** ***** ****, ******* ******* are **** ******* *** *** ** expensive ** ****** ****, ** ******** papers ****. *** *******, ****'* ********** wrote ******* *****:
******** ** ****** ******* *******, ******* imagers **** *** ********** ** * simpler *** *****-**** ******** ************, *** they *** ****** ********** ***** **** good ********, **** ** ******* ** interference *******. *******, ****** ******* **** actively ********** * ******, ******* ******* must ******** * *********** ******** ******* detection *********** *** ******* *****
*** **** ******* ** ******* ********* imagers, **** **** ****** ********* ********** *********,***** ***** / "******" ********* ********,******* ********* ** *********** ********** *****, ********** ********* *** ******** ****.
Active ********* **. ******** ***** *********
***** ****** ********* ******* ***** ********* of **** ******** *** ***-******** *******, they *** **** ********* **** ********-***** systems (**** ******* ********* ******* *** more ********* **** **** ******** *******), require *** *******, *** **** ******* performance ** ******* ** ****.
****** **** ****, *** ********-***** ******** is * ************* **** ****** ****** of ********* ******* **** *********-*****, ** metal ********* **** **** ****** *** multiple *******, *** **'* **** ~ five ***** ***** ********-***** "******* *********" have **** **************.