First Aid Tent

New Disaster Shelters Use Edge AI to Manage Pandemics

As we all begin to live with the global aftermath of Covid-19, the challenge is in creating smarter and safer disaster shelters to alleviate the heavy loads put on hospitals when another infectious disease breaks out or when the next natural disaster occurs.

During any given season of the year, countries face the threat of natural disasters – typhoons, tornadoes, hurricanes, flooding, wildfires, earthquakes, and tsunamis.

Pain Points of Traditional Disaster Shelters

As Covid-19 has brought to light, creating and using impromptu shelters for disease control and prevention leaves much to be desired. Here are a few ways that traditional disaster shelters have failed in the wake of Covid-19 and taught us lessons about what we can improve upon:

  • Disaster shelters are typically set up in schools, gyms, public halls or community centers with little or no patient-to-patient or doctor-to-patient separation.
  • They require a large number of front-line staff to manage everything on the ground.
  • Due to the urgency in accommodating large numbers of victims in these places, these makeshift indoor spaces are often overcrowded and utilized as much as possible.

Due to the situations above, an unexpected and innumerable amount of group infections occurred when the first wave of Covid-19 hit. Because of that, however, we can learn how to improve on the idea of how a disaster shelter should be created.

New Disaster Shelter Criteria

In countries like Taiwan and Japan where natural disasters occur frequently, governments have come up with guidelines for new kinds of disaster shelters. For example, the Office of the Cabinet of Japan has formulated the following criteria for the creation of these new disaster shelters: 

  • Prepare more sites in advance for disaster shelters in order to reroute much of the affected population.
  • To avoid gathering in large crowds, it is recommended that affected people seek refuge at the homes of relatives or friends. 
  • All people entering the disaster shelter must be physically examined.
  • Disaster shelters should focus on maintaining good ventilation.
  • Disaster shelters must provide dedicated spaces for people with fevers or other symptoms, and provide a dedicated bathroom to people with abnormal health symptoms if possible.
Field Hospital

In the past and under normal conditions, the government advocated that these items be strictly followed. Yet these key guidelines are quite difficult to maintain in an emergency situation.

Edge AI Solutions in New Disaster Shelters

There are many governments and local authorities in countries that are paying attention to how AI technology can be used in shelters in areas at high risk for natural disasters. For example, using an access control system with contactless temperature detection combined with facial recognition technology can record body temperature and personal data when anyone enters the shelter. That data can then be linked to a specific face identity. Because the system is contactless, it can help avoid transmission between the front-line staff and the admitted person because the device replaces the front-line staff with technology.

Furthermore, physical methods have been used to block people from associating with one another. This is achieved by creating spaces separated by plastic or cardboard dividers. Video analytics can be used to confirm the number of people in each area is within limits. Using AI identification along with personnel management can work together to efficiently strengthen protocols and efficiency.

In addition, mask detection video analytics can be used to identify whether or not an entrant is actually wearing a mask. This cuts the common reliance on staff to check at entry and exit points, thus again reducing the need for front-line staff.

Loss or theft of relief supplies for disaster shelters is often a problem. It’s impossible to allocate people to monitor these materials 24/7. If there are non-staff wandering around the storage area, video analytics can alert the proper personnel and warn them to take necessary action.

Finally, in case an infected person does enter a shelter, focusing on contact tracing methods to effectively locate all the people they have interacted with to prevent further spread of the disease is important as well. Carrying out effective contact tracing, quarantines, and similar measures can benefit from technology and is being closely watched by governments and local administrations.

Covid-19 Alert

Transforming the Traditional with Edge AI

Covid-19 is a pandemic the likes of which the world has never seen and transforming traditional shelters into a more intelligent disaster refuge is an issue that all governments must address. Gorilla’s edge AI technology is leveraged in creating a variety of post-pandemic solutions that governments and countries can adopt. In the future, Gorilla Technology will continue to produce the most advanced video analytics and will continue to develop products that contribute to society.

virus-5308335_1920

5 Ways Video Analytics Can Help Manage Covid-19

Since the pandemic began, wearing face masks and social distancing have been used as preventive guidelines for individuals, organizations, and governments to follow. Many governments have gone so far as to make those guidelines into mandates and laws.

Whether guidelines or laws, it’s still the responsibility of society to maintain these and make sure a second wave of the pandemic does not accelerate out of control. But how can organizations or governments monitor or even execute these rules with limited human resources and no smart methodologies on hand?

We recommend intelligent video analytics — AI technology interpreting IP video sources. Preventive measures like mask wearing and social distancing guidelines can be easily monitored as well as executed. Below are five examples of intelligent video analytics from Gorilla Technology and their use cases that can help you make a proactive difference in responding to the Covid-19 pandemic:

 

1. Facial Recognition

Employing no-touch access control reduces physical contact with devices and people, which can drastically reduce transmission of the virus.

 

2. Mask/No-Mask & Eyewear Facial Recognition

Mask guidelines and mandates are in effect to some degree all over the world and leveraging edge AI to assist in enforcing these regulations decreases required manpower and increases efficacy in areas where personal protective equipment is required.

 

3. Crowd Detection with Occupancy Analysis

As social distancing is part of the new normal, people counting and crowd detection can help keep transmission risks to a minimum.

 

4. Crowd Detection

Once offices reopen, management can ensure break rooms and other gathering spots remain safe and not overcrowded with edge AI zone capacity alerts.

 

5. Heatmap and People Counting

Good air quality in high foot traffic areas is important and heatmap & people counting analytics can empower HVAC systems to circulate air appropriately to keep infection risks low.

We firmly believe in and see the benefits of implementing intelligent video analytics that can adapt to many situations — it saves time, money, and makes it easier to quickly solve new and unforeseen challenges, like what we are experiencing now with Covid-19.

For more info about this topic, refer to our solutions on Post-Pandemic Area Management. Also, if you’d like a deeper dive on the technology or specific cases, our whitepapers offer more details.

DataCenter16x9

Post-Pandemic Situational Monitoring with Big Data

The period of lockdowns around the world is coming to an end. It is inevitable that business and operations need to reopen for economic survival. However, the possibility of new waves of the virus after re-opening is an issue that requires close monitoring.

Balancing Economic Activities and Epidemic Control

The answer to this problem is actively sought out by various government agencies. From gradual re-openings within borders, to the conditional re-openings at borders (such as travel bubbles), these policies and measures are already starting to be implemented in many areas.    

The most important supporting measure after a lockdown – apart from wearing masks, social distancing, and frequent hand-washing – should be epidemic surveillance. This means monitoring those in home quarantines and self-isolation. For example, once a confirmed case is discovered, the history of who they came in contact with needs to be determined in the shortest time possible. Once those potentially infected people are discovered, monitoring them and possibly enforcing home quarantine and self-isolation is a high priority.  

When monitoring is not automated, the staffing required will be massive and it may not be possible to execute quickly. A failure like this could cause a large-scale community-based spread of the virus that cannot be traced, which would then lead to the city needing to return to lockdowns yet again.

Considering automated monitoring, it can be applied via the deployment and use of post-pandemic area management solutions like big data analysis. From monitoring home quarantines and whether the monitoring device of a quarantined person strays from its designated location, to understanding the previous movements of a newly diagnosed person – automated big data analysis can be applied to identify possible connections quickly, accurately, and efficiently.

The monitoring devices used for quarantine and self-isolation are IoT devices that can report back to the big data platform at set times. Once an irregular report from the IoT device is recorded, the big data platform can immediately issue an alarm to notify personnel.

Understanding Contact History with Big Data

To accurately understand the contact history of a newly diagnosed person, data from multiple sources is required. For example, the logs from a telecommunication company’s cell tower, check-in/out times from locations using ID verification, CCTV recordings, credit card spending history, etc. would all be utilized in contact tracing. These data can ultimately be analyzed to find possible contacts with location/time crossovers.

For integrating heterogeneous data from multiple sources, Gorilla’s big data analysis platform provides real-time event notifications and space–time analysis functionality, making it an optimal solution for contact tracing.

Contact us to learn more about post-pandemic area management solutions like our big data platform. If you’d like to read more in-depth on the topic, please check out the whitepaper we recently published here: https://www.gorilla-technology.com/Edge-AI/whitepapers

architecture-buildings-business-city-325185

Post-Pandemic Workplace Wellness & Intelligent Building Automation Systems

The Pandemic

The world was caught unprepared for Covid-19 and its profound consequences on public health and the economy. In just the first six months following its discovery, the highly contagious virus was able to infect over ten million people globally, and has been responsible for at least half of a million deaths around the world in as much time.

Since the outbreak, the pursuit of workplace wellness has dramatically accelerated. Workplace wellness encompasses the physical health and safety of its employees, also how they can successfully perform their jobs and how they interact with one another.

Workforce Vulnerabilities

Most jobs (office, engineering, construction, government, etc..) can be broken down into daily tasks that have at least some element of human interaction – this may act as a channel for the virus to spread. Jobs that involve these tasks not only expose employees to risk of infection, they can also increase the spread of the virus by acting as a distribution network for it.

The challenge is in reformatting these jobs to keep our workforce healthy and productive. Checking temperatures at entrances, maintaining social distancing during working hours, assisting health professionals with contact tracing – these are just a few examples of what offices will need to do in order to reopen safely.

Enhancing Building Automated Systems with Edge AI

Edge AI driven video analytics from Gorilla provide reliable solutions to slow the spread of the virus. Integrating and deploying facial recognition access control, attendance systems, and smart parking solutions with current building automation systems to automatically monitor compliance in many ways are now possible. To name a few:

  1. Performing screening for fevers and masks at entrances and other key zones.
  2. Using crowd detection in cafeterias and break rooms to ensure proper social distancing is followed.
  3. Assisting health authorities with link analysis & contact tracing when positive cases come to light.

Post-Pandemic Area Management solutions from Gorilla can also help maintain public safety and order – something that will prove to be a challenge as lockdown measures result in closed businesses and greater economic hardship for many people.

Gorilla’s offerings of behavior analytics, intrusion detection, loitering detection, direction violation detection, fighting detection — these can all help minimize the prospect of looting, vandalism, robberies. Gorilla is reaching out to assist law enforcement in preserving public order during this trying time.

While the health crisis is far from over and the full impact of the virus and global lockdown is still to be seen, we know we will get through this. We will return to our workplaces as they are re-tooled into safer and smarter environments – and Gorilla will help get us there.

Airport Terminal

Improving Public Spaces at Transportation Hubs with Digital Transformation

As cities and borders find ways to reopen, students, business people, and travelers are preparing to travel and catch up for time lost. However, we can’t ignore the risk of another outbreak before a vaccine is discovered. Airports, train stations, and metro stations are critical areas as we expect massive foot traffic. This article will look into how video analytics can assist people in getting adjusted to the new norms.

Staff at transportation hubs are not able to respond promptly to people not wearing their masks if there’s congestion or a long queue at the counter – if they even get noticed at all. Traditional CCTV systems can’t deliver actionable events. By utilizing the video source on site and implementing Gorilla post-pandemic area management solutions, we become capable of assisting transportation operators in keeping the area safe by ensuring passengers are wearing masks and are not gathering or getting jammed together in large groups in one area.

Common Challenges Facing Transportation Operations

Train cars, platforms, station halls, ticket booths and check-in counters –  these are the areas where we expect congestion taking place when borders reopen. With a lack of human resources and an influx of heavy traffic, we won’t get clear visuals when a space becomes overcrowded. We have to rely on past experiences most of the time. Motion sensors are known as a common way to monitor how many people enter or exit. Unfortunately, with the lack of visual input, we can only rely on the number statistics sent by the motion sensor. However, with crowd detection video analytics, we can clearly tell which area is crowded and where we should send staff to assist in dealing with the queue in a timely manner. Gorilla’s post-pandemic area management solution offers an all-in-one solution in response to the current situation.

Lonely Woman Wearing a Mask on Train

To adapt to the new normal, passengers in most areas are requested to wear a mask when taking mass transportation. It has become a must. In some countries, they have strictly advised their people to have their masks on throughout the entire journey. While at the turnstiles in stations and other hubs, it’s challenging for staff to monitor multiple lanes of passengers for masks.  Gorilla’s mask detection solution is able to monitor if passengers are wearing masks before entering the facility. The intelligent video analytics does the detailed and challenging work of examining dozens of gates from different entrances/exits.

Post-Pandemic Digital Transformations Are The New Normal

Intelligent video analytics maximizes the use of video footage on-site. By running real-time video analytics, meaningful insight and action can be defined by users. Our post-pandemic area management solutions enhance workplace efficiency in many places – transportation hubs included – and assist in minimizing health risks for staff. For a more in-depth look, you should read the whitepaper we recently published regarding post-pandemic area management solutions.

Gorilla has built up a strong community and connection with global partners in many fields, e.g. system integrators, managed service providers, and hardware makers. We collaborate to find out the challenges our customers face, offer virtual training for certified partners, and deliver actionable end-to-end solutions.

woman-holding-face-mask-3902883

A Tense Encounter at the Grocery Store

Any parent will understand that your kids are at home with you all day and every day. Every chore and every errand you have to run, your child is with you – there are no schools and no childcare available. This includes errands like grocery runs.

Last Wednesday morning, I geared up to go to the grocery store with my child. Masks? Check. Sanitizing wipes? Check. Hand sanitizer? Check. Snacks for cranky kiddo? Check. We got to the store and I noticed there wasn’t a security guard standing outside to ensure everyone abided by the ‘no face mask, no entry’ rule. I figured it would be ok because the store looked pretty empty – I had strategically chosen the morning to conduct this task because this shop is usually slow before lunch.

We entered the store, I grabbed a shopping cart and wiped it down with an alcohol wipe before putting my child in it – then went about my shopping.

Once I completed our shopping, we got in the checkout line and stood on the strip of red tape six feet behind the shopper ahead of us to follow the store’s social distancing policy – there was another strip six feet behind me, and so on.

There were only two of the ten cashier lines open, which is normal for mornings. Yet as I looked at the impulse buys in front of the register, the line behind us grew. My strategic plan of going shopping in the morning to avoid a crowd was crumbling.

I turned from looking at gum and noticed that the man behind us wasn’t wearing a mask and was mumbling to himself. Visibly agitated, he was muttering profanities about the slightly longer wait and had moved to within whispering distance of me. Did he even notice the red tape?! I think all parents, everyone for that matter, can imagine the fear and panic snowballing in the pit of my stomach as I hovered closer and closer to my child – hoping to shield her from harm with my body. Enough was enough.

At that point, I took a breath from through my mask, turned, and firmly told the man behind us that we’re all in this together and he should kindly withdraw to the red strip of tape six feet behind us and offered him an extra mask I had packed. He momentarily balked at my comment, but then begrudgingly took the mask and moved back to the strip of red tape. We got home without further incident, but it got me thinking – was that whole episode really necessary?

Since I work for an edge AI and computer vision company, I know there are better ways to manage and control situations such as the one I experienced. Firstly, deploying mask detection at store entrances can avoid the issue of not having security personnel on duty. If the store wants to invest further, they can include crowd density detection throughout the store as well as at the checkout. Computer vision is quick to implement. It isn’t complicated and there is no breach in privacy. In any case, if it can help protect my child without me having to step in and confront an agitated stranger, I really don’t mind a little bit of surveillance. 

As an active member of the community, I do my civic duty whenever I can. I hope grocers and other retail shops will do the same by implementing extra measures to protect their shoppers.

To read more, click on what interests you below:

Post-Pandemic Area Management

Video Analytics

Edge AI

Whitepapers

Reopening Service & Retail Business Safely in the Post-Pandemic World

Reopening Service & Retail Business Safely in the Post-Pandemic World

Lifting the Lockdown with Edge AI Technologies

Countries are gradually lifting their lockdown policies with the hope that life will go back to what we knew before as normal. However, we are all conscious of the fact that the impact of Covid-19 has redefined our understanding of normalcy forever.

One of the most significant consequences of the pandemic is the acceleration in the adoption of technology to optimize and automate operations. We have seen a steep increase in the demand for edge AI solutions over the past months in preparation for the world after COVID-19. Gorilla Technology has developed a dedicated solution that is not only flexible and scalable, but also market-ready. The solution taps existing products which have been customized to focus on helping businesses achieve the following goals:

  • Monitor Social Distancing
  • Optimize Service Efficiency
  • Automate Operational Processes
  • Improve Overall Store Hygiene

The New Normal for Service and Retail Business

It has become increasingly important during lock-down to monitor social distancing. Inside brick and mortar stores such as supermarkets, governments have restricted the number of people allowed within any given area at any one time.

Entrances and Access Control

Currently, staff has been diverted to manage people flow at the entrance of stores – a process that comes wrought with unavoidable human error and inefficiency. To combat this, when more businesses reopen their doors we can deploy and see edge AI video analytics performing automated people counting at the door, with minimal staff engagement required. Screens at entrances will show the exact number of people currently in the store, as well as the max occupancy allowed. The queue outside the store will be separated by marks on the floor. Once inside the store, automatic alcohol dispensers will be placed at various points for customers to use as they like.

For businesses that don’t need to keep their doors open during business hours, an access control device can be deployed at the entrance to measure peoples’ temperatures and grant or deny access accordingly. These devices can be equipped with Gorilla’s facial recognition technology in IVAR™, which includes mask detection and can integrate with temperature sensors. In order to provide flexibility, solutions need to be scalable, cost-efficient, and contactless. The choice will naturally be to deploy edge devices able to run AI analytics and convert this into real-time data. The key tool behind it will be computer vision.

Analytics On the Sales Floor

Inside the store, behavior analytics based on video footage will be used to monitor trends over time. The importance of gaining a bird’s eye view of customer’s behavior in the store is crucial for further action.

Businesses want to understand which areas in the store were frequented more in order to do proper sanitation. This data can also reveal customer preferences, which in turn can lead to more effective inventory management as well as an improvement in in-store product placement. Popular products can be placed in more accessible areas, but also, this can benefit the business owners by allowing them to better manage the shelf-space fee from vendors.

Having precise information about how much time people are spending in certain areas (dwell time) could prove valuable in assisting health organizations perform contact tracing. The contact history of the infected person can then be accurately traced, to inform people that have unknowingly been in contact with them. In addition to this, with the same dwell time information, retailers can correlate the data to uncover the exact product that people are spending more time looking at and tweak their supply chain accordingly.

Previous to the pandemic, most of the sourcing in retail chains was done through personal relationships between vendors and buyers. Decision making was largely made by buyers who relied on their own experience, a system that is time consuming and not quickly replicated. The pressing need to automate and make decisions based on precise data will become indispensable for businesses reopening in the post-pandemic era.

POS and Check-Out

The need for social distancing will not subside for a long time. Some businesses estimate that such measures will need to be kept in place for the next two to three years. In view of this, the areas where people are likely to concentrate within brick and mortar stores are crucial. The check-out area is one of these high risk areas where edge AI video analytics can help manage the amount of people queuing up in order to reduce wait times.

Crowd detection analytics in computer vision solutions like IVAR can help detect the number of people in a determined zone and send a notification to appropriate staff when the amount of people exceeds a pre-established limit. Once staff is notified, they can reduce check-out time by opening more counters to serve more people.

Reopening Through Digital Transformation

If COVID-19 has taught us anything, it is to pause and reflect on our business practices. This forced halt in our daily activities has no doubt impacted us negatively, but has also allowed the world to reset itself and find new best practices.

It has made us humble about our own limitations and brought us together to collaborate as a community. It has put into question the way we consume, and striped our purchases to the bare necessities. It has also pushed organizations that were hesitant about digitalization to quickly pivot and adapt. In the wake of the pandemic, the service and retail industries will go back to business transformed – empowered by AI-driven technology and enhanced by data-driven decision making.

Contact us for more on digital transformations.