The global spread of Coronavirus has significantly disrupted many industries across the world. Though many organizations are working remotely, eventually, there will be a need to resume operations from factories and offices. Once the operations resume, organizations will have to implement strict Covid protocols and ensure their workforce adheres to them. This will require constant monitoring and vigil, which technology can enable.
Governments have made social distancing mandatory by laying down Standard Operating Procedures for offices, workplaces, factories, and other establishments. Organizations can ensure compliance with protocols with the help of surveillance cameras and mobile applications.
However, manually monitoring these conditions is a huge challenge and may not prove effective. Implementing Machine Learning algorithms will make tracking a lot more effective and less tedious. ML algorithms will have to be implemented on the Edge for quick decision-making and require high-performance processors that can rapidly process large amounts of data. NPU (Neural Processing Unit) computes Machine Learning tasks 10K times faster with low power consumption and improved resource utilization than GPUs and CPUs.
NXP i.MX 8M Plus application processor has a dedicated neural network accelerator and supports advanced neural network processing. The processor is equipped with a Neural Processing Unit (NPU) as well as dual image signal processors (ISPs), and Graphic Processing Units (GPUs). In addition, the ML accelerator for ISP enhances ML performance by extracting maximum image details in high contrast scenes. The capability of real-time image processing can be used in applications such as surveillance, smart retail, robot vision, and home health monitors.
i.MX 8M Plus can produce industry-leading audio, voice, and video output for applications that range from consumer home audio to industrial building automation and mobile computing, and the i.MX8M Plus can be used for building control, consumer electronics, healthcare, instrument clusters, multimedia devices, voice assistance/control, automotive electronic cockpit. This versatility, combined with the processing capabilities, makes it an ideal processor to perform edge monitoring and detection effectively.
Companies are coming up with numerous ML models to measure the distance between people in public or private areas. The system comprises ML algorithms that can identify whether people keep safe gaps by measuring their distance from one another.
NXP, EBV, and Dave Embedded Systems have jointly designed and developed an open-source stand-alone camera system based on the NXP i.MX 8M Plus processor to monitor social distancing between individuals. The processor has an NPU to accelerate the inference process, the ARM Cortex for video pre-processing, and the Graphical User Interface (GUI) supports human interaction.
While several individuals are going about their work, surveillance cameras monitor them to see if Covid protocols are followed. First, the i.MX 8M Plus-based camera and ML algorithm detect the individual’s presence by reconstructing their skeletons and calculating the physical distance from the person near them. Then the Graphical User Interface shows whether the individuals maintain the required social distance by imposing various colored circles around them based on the gap between them.
This processor can be incorporated into surveillance camera systems at various workplaces and public areas. However, organizations would need a method to inform individuals when they disregard social distancing measures. One possible option is an alert that sounds when employees come near each other or ping them on a device they have with them, such as their mobile phones. It can help in creating a safer work environment and keeping the business compliant with regulatory requirements. The video feeds can also be used further as datasets for training the algorithm for more precision. Recurring patterns can be studied from the data, and effective guidelines can also be produced to ensure that employees always follow the protocols, reducing the chances of being infected.
Companies and factories will need technology partners with extensive knowledge of processors and proven expertise in Machine Learning to implement solutions like these.
Being an NXP Connect technology partner, eInfochips comes with the knowledge and experience to enable and implement total system solutions, developing customized evaluation kits, reference designs, and next-generation, fully-featured products on NXP platforms.
Connect with us to know more about our offerings and our expertise on i.MX 8M Plus and Machine Learning.
This post was last modified on July 6, 2021 7:05 am
With the urban areas witnessing a population rise and industries becoming technologically advanced, the need…
The automotive manufacturers now focus on some major capabilities like real-time vehicle-to-vehicle communication, vehicle-to-infrastructure, telematics,…
The future of medical imaging: beyond CT scans The traditional medical imaging systems primarily include…
The global supply chain has been witnessing tough times for the past few years. The…
According to Allied Market Research, the worldwide wireless charging market was valued at $6.51 billion…