|Product Performed to Expectations:||10|
|Specifications were sufficient to design with:||10|
|Demo Software was of good quality:||10|
|Product was easy to use:||10|
|Support materials were available:||10|
|The price to performance ratio was good:||10|
|TotalScore:||60 / 60|
The Raspberry Pi 3A road test project is to build a “Welcome to our Robotics Club” face recognition system on the Raspberry Pi 3A.
The project should serve to inspire people, either in a future STEM career or career change, or just to be creative and thinking out of the box, and to show off Raspberry Pi capabilities at exhibits, robotics and coding meetings, and raspberry pi Jams.
I added workscope creep, by installing the facial recognition software on both the Raspberry Pi 3A and 3B, and comparing the experience.
To summarize the performance, both the Raspberry Pi 3A and Raspberry Pi 3B were surprisingly up to the challenge of a sophisticated and lengthy install and compute task.
When tasked with some heavy hitting of SW building using all 4 cores, RPi3A got pretty warm, but the RPi3B got hot and had some temperature alarms.
The Raspberry Pi 3B had heatsinks installed on its chips, but the 3A was as it is out of the box. I considered this as pretty fantastic accolade for the RPi3A.
The Raspberry Pi 3A has no physical RJ45 ethernet jack, and only one USB port, compared to the RaspberryPi 3B with the RJ45 and qty 4 USB ports.
95% of the setup and install of the facial recognition software was done using SSH, I surprisingly hardly missed those 4 USB ports.
All in all, the RPI3A is an incredible value and can do so many things, it is truly impressive.
The NOIR Pi camera used in the project cost more than the Raspberry Pi 3A.
In the interest of your sanity, I will be brief on the rest of the project details, because the point of the report is the Raspberry Pi 3A :-)
I proposed the facial recognition project because: it was a monumental challenge, way past my comfort zone, and was just so intriguingly cool and amazing.
Both Raspberry Pis started with 8Gb uSD cards, then quickly replaced with 16Gbs cards, then finally 32 Gb uSDs which gave the PIs plenty of headroom room for data and extra application programs.
I had searched online for open source help on raspberry Pi facial recognition.
Most hand waved the important details, after trying a couple and getting stuck, and nearly frying the Raspberry Pi 3B CPU,.........
I finally stuck with Pyimagesearch. https://www.pyimagesearch.com/2018/06/25/raspberry-pi-face-recognition/ .
The PyimageSearch team did reply to questions, and is genuinely interested and adept in their AI research, very kind in sharing and helpful to those who they have provided free expertise.
I will also say in my introduction to the topic, I came away very impressed with the technology ( and astounded), the PyimageSearch team, its potential applications, and even somewhat creeped out by the potential.
I see facial recognition everywhere now.
The U scan at the grocery store knows its me putting in my plastic card into the checkout register.
That same U scan looks at my face when I get something needing to be over 21 to buy, and assures the cashier I’m old, its OK.
The U scan sees my face and mentions I usually get ice cream, but none this trip, I better go get some chocolate because I look in a chocolatey mood.
Back to the topic, I also learned a ton about nano, vi, SSH, VCN, video processing, openCV, virtual workspaces and the Raspberry Pi system, just by doing, failing, and redoing.
The Raspberry Pi 3A is a fabulous value, especially for doing projects with IOT components, cloud interfaces, and in places needing a compact footprint.