Covid security


contact tracing purposes working with the NHSX teams if a person tested positive in the test day prior to the event. All data from the pilot Fans Are Back event will then be fed back to DCMS, Public Health England and the NHS. The BioSURE rapid Covid-19 test is the only antibody kit available which tests for IgA, IgM and IgG. Man- ufactured in the UK, the CE certified and MHRA registered triple antibody test is able to identify if a person cur- rently has Corona virus (Covid-19), or if they have had a previous infection. Providing qualitative results in under 10 minutes, this test is praised as a ‘game-changer’ for the screening of high volumes of people to increase


the efficiency of Covid test- ing and crowd safety.


Lockdown Amid


of an immi- nent


threat lock-


down in the UK, it could be some time before


these


new solutions have any real impact. The good news is that these technolo- gies exist to hasten the recovery when it starts.


The face mask problem


In February, some provinces and mu- nicipalities in China made it mandato- ry to wear masks when in public. News reports soon followed of residents and police chastising the non-compliant, a trend that’s now seen globally as mask wearing has been rapidly adopted as a social minimum. Akash Takyar was shocked at how things were being handled in China, and he wondered if his software company – LeewayHertz - could offer a more peaceful solutions. Takyar recognized how important it is to wear a mask to slow the spread of Covid-19. Rather than leave members of the public to monitor each other, he wanted to develop a computer program that could look at images and detect whether people are wearing masks. His San Francisco-based company is one of many now pioneering mask rec- ognition to get people to comply for the public good. So far, masks have been confounding


traditional facial


recognition software—but these new machine learning tools could con- ceivably be used in private or public spaces to measure compliance and ostensibly take that out of the hands of individuals. Tryolabs’ software iden- tifies whether visible faces are uncov- ered or wearing masks as they pass by a closed-circuit television camera. For the most part, people who flout these mandates, even if they can af- ford to follow them get away with that noncompliance. For businesses that


18


have workers returning to indoor fa- cilities, noncompliance could lead to others in the workplace getting infect- ed. Ultimately, it could be a great loss for a business if there was an outbreak because someone was asymptomatic and failed to wear a mask, says Takyar. But “face data is as precious as a fin- gerprint,” says Deborah Raji, a fellow at the AI Now Institute at New York University. And those who have had qualms about facial recognition won- der whether mask recognition soft- ware, however well-intentioned as it may be. Today’s facial recognition software studies the features around the eye, nose, mouth, and ears to identify an individual whose picture is already supplied, either by the individual or in a criminal database. Wearing a mask obstructs this recognition—an issue that many systems have already en- countered, and others have solved. Developers say that mask recognition software in theory bypasses privacy is- sues because the programs don’t actu- ally identify the people. Such software is trained on two sets of images: one to teach the algorithm how to recognize a face (“face detection”) and a second to figure out how to recognize a mask on a face (“mask recognition”). The machine learning algorithm doesn’t identify the faces in any way that can link a face to a specific person, be-


cause it doesn’t use a training set— the set of examples used to train such programs—with faces that are linked to identities. Tryolabs’ software has two main com- ponents: The ‘pose’ algorithms deci- pher different parts of the body before the ‘classification’ algorithms decide if the facial region includes a mask. Companies that have developed mask recognition software say that they ul- timately want this technology to be used in broad ways that help people set policy or enhance awareness cam- paigns. “If we can compute the num- ber [of people who are complying with the mask mandates], people can make policies and monitor on whether or not they need to do another campaign to push mask usage,” says Alan Des- coins, the chief technology officer of Tryolabs, a company based in Monte- video, Uruguay. LeewayHertz’s algorithm, for example, could be used in real time and inte- grated


with closed-circuit television


(CCTV) cameras. From a given frame in a video, it isolates images and or- ganises them into two categories, peo- ple who are wearing masks and those who are not. Currently, this recogni- tion software is being used in ‘stealth mode’ in multiple settings in Europe. Restaurants and hotels are using it to make sure the staff is complying with wearing masks.


A team including England rugby star Mike Tindall and his wife Zara Phillips is promot- ing a rapid antibody test leading to a mobile phone based digital passport which allows those who pass to enter sports and other mass participation events.


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24