|December 2021January 2022

* Hardware-accelerated color space conversion (Graphics Processing Unit, GPU)* Hardware-accelerated real-time inferencing (Neural Processing Unit, NPU)* Hardware-accelerated video encoding (two streamsVideo Processing Unit, VPU)* Hardware-accelerated video decoding (two streamsVideo Processing Unit, VPU)* Hardware-accelerated GUI (Graphics Processing Unit, GPU).

{| style="background-color: #ededed;padding: 20px"|[[File:ML-TN-005 dual camera system.png|none|thumb|right|170x170px|Dual Camera Module with Machine Learning Processing Unit]]|[[File:OrcaSBC-demo-dual-camera3.png|thumb|center|600px|Test Bed block diagram]]|[[File:ML-TN-005 HDMI GUI module.png|alt=Decoded and Display Module|thumb|left|170x170px|Decoded and Display Module]] |}

*[https://www.baslerweb.com/en/products/cameras/area-scan-cameras/dart/daa2500-60mc-s-mount/ daA2500-60mc]*[https://www.baslerweb.com/en/products/cameras/area-scan-cameras/dart/daa3840-30mc-s-mount/ daA3840-30mc]

*camera module #1 (daa2500daA2500-60mc)

*camera module #2 (daa3840daA3840-30mc)

Camera module #1 stream is the one processed by the Social Distancing (SD) algorithm, which is described in detail in [[ML-TN-002_-_Real-time_Social_Distancing_estimation|this document]]. As such, this stream is fed to an NPU-powered neural network. This stream is also color space converted to fulfill the input requirements of the network. In order to offload the ARM Cortex-A53 cores, the conversion is performed by the GPU.

The output stream of the SD algorithm is hardware -encoded before being streamed over a Gigabit Ethernet connection. On the Orca SBC #2 side, the stream is hardware -decoded and visualized on a HDMI monitor.

Stream #2 originates at the daa3840daA3840-30mc camera module. Then, it is hardware-encoded and streamed by the Orca SBC #1. Finally, it is received by the Orca SBC #2, which hardware decodes and displays it.

A GUI application runs on the Orca SBC #2as well. This application is engineered with [https://www.cranksoftware.com/storyboard Crank Storyboard 6] and shows some parameters of the Social Distancing algorithm while it operates. The GUI application remotely communicates with the SD Social Distancing application — which is written in Python — through [https://resources.cranksoftware.com/cranksoftware/v5.3.0/docs/webhelp/index.html#ww_sbio.html Storyboard IO API]. According to official documentation, ''the Storyboard IO API, formerly known as GREIO, provides a platform independent communication API that allows inter-task and inter-process queued message passing. This is primarily used to allow external communication with a Storyboard application''. For more details, please see [https://support.cranksoftware.com/hc/en-us/articles/360056943752-Receiving-Events-from-a-Storyboard-Application this article].

Functional tests were conducted in an environment that mimics a real-world scenario. The goal is to use the SD algorithm for people detection in an a hazardous area where a cobot operates and thus where only authorized personnel are allowed to enter — the cobot shown below is part of the actual DAVE Embedded Systems manufacturing department's [https://en.wikipedia.org/wiki/Automatic_test_equipment ATE] test-benchesof DAVE Embedded Systems manufacturing department.

Orca SBC #1 in is installed in the proximity of the cobot, while Orca SBC #2 is located in a different room where human operators can monitor remotely the test-benchremotely. Of course, more advance advanced notification mechanisms could be also implemented in order to alarm operators if people are detected in the hazardous area.

TBD inserire {| class="wikitable" | width="100%"| {{#ev:youtube|EMvh6gMmpXQ|600|center|Simultaneous dual video e fotocapturing and encoding plus Machine Learning People detector on NXP i.MX8MP|frame}}|}