Hi Samuel, Great Job. Is there a way to read multiple rows from a device like a Blood Monitor Pression which has three rows of seven segment digits (Sys, Dia, HeartRate). I can't figure it out how we can achive that. I Guess that implementing like a Text Detection Algorithm or something like that but I've no idea on how to do it. Can you show us how to do that? Thanks
Hey, you can also modify the code in the computer vision essentials repository -- I'm not actively maintaining it anymore but the code is open and free to use anyway. It should be relatively easy to experiment with once you get the dependencies installed. I also run a consultancy (supertype.ai) if that's more of what you're looking for -- we do these kind of computer vision work a lot and might be an option too!
Yes I used to teach this in workshops! All of the materials are open source on my GitHub! github.com/onlyphantom/cvessentials I might try to revive it through my mentorship and peer learning platform (no definite dates though, doing a lot at the moment 😁) fellowship.supertype.ai
Yes; When I teach this course I just ask my students to bring a laptop with webcam, instead of having to buy a fancy camera. Most of the work is done in OpenCV. Another low cost alternative is the raspberry Pi camera module which is US$15-20.
Detecting decimal points should be even easier than the digits. Use the findContours() then if detected, calculate size to confirm it is in fact a decimal point. You can tweak the code from here dear friend: github.com/onlyphantom/cvessentials/blob/master/digitrecognition/digit_01.py
If I understand you correctly, you should be looking at cv2.VideoCapture(0). If you work through the materials on my GitHub you will see sample code - too much to include in a single TH-cam comment :) GitHub.com/onlyphantom/cvessentials
You can find code from the video in commit b4e1df2ed84c87fdc65e18c2fd46b98a25ac94d1 in master on Jan 28, 2020 In latest commit to this date file digit_01.py missing setup of FONT, CYAN and canvas variables
![[DEMO] Headshot Tracking || OpenCV | Arduino](http://i.ytimg.com/vi/f2TUxoaKIsA/mqdefault.jpg)
Samuel, thank you for the hint with threshold filter. When applied, it is much easier for OCR to see the digits with such font.
Big thank you. Greatly implemented course.
Hi Samuel, Great Job. Is there a way to read multiple rows from a device like a Blood Monitor Pression which has three rows of seven segment digits (Sys, Dia, HeartRate). I can't figure it out how we can achive that. I Guess that implementing like a Text Detection Algorithm or something like that but I've no idea on how to do it. Can you show us how to do that?
Thanks
Hey, you can also modify the code in the computer vision essentials repository -- I'm not actively maintaining it anymore but the code is open and free to use anyway. It should be relatively easy to experiment with once you get the dependencies installed.
I also run a consultancy (supertype.ai) if that's more of what you're looking for -- we do these kind of computer vision work a lot and might be an option too!
How can i get the code of this video
Do you have a workshop reading lcd digits with A.I. ? Nice video!!
Yes I used to teach this in workshops! All of the materials are open source on my GitHub!
github.com/onlyphantom/cvessentials
I might try to revive it through my mentorship and peer learning platform (no definite dates though, doing a lot at the moment 😁)
fellowship.supertype.ai
This code can use detect digit via Camera?
Yes; When I teach this course I just ask my students to bring a laptop with webcam, instead of having to buy a fancy camera. Most of the work is done in OpenCV. Another low cost alternative is the raspberry Pi camera module which is US$15-20.
can this load it to raspberry?
I actually used a raspberry pi camera module v2 mounted on an Nvidia Jetson for the live workshop!
The code is on github.com/onlyphantom/cvessentials
calm
How to add decimal points if we want to extract digits from electricity meters?
Detecting decimal points should be even easier than the digits. Use the findContours() then if detected, calculate size to confirm it is in fact a decimal point.
You can tweak the code from here dear friend: github.com/onlyphantom/cvessentials/blob/master/digitrecognition/digit_01.py
hey i want to ask about the using camera I tried try a lot of thigs where should i change in the code to make it use camera and thanks :)
If I understand you correctly, you should be looking at cv2.VideoCapture(0).
If you work through the materials on my GitHub you will see sample code - too much to include in a single TH-cam comment :)
GitHub.com/onlyphantom/cvessentials
what if the image is flipped
you can use the cv2.flip() method to cancel out / undo the flip?
You can find code from the video in commit b4e1df2ed84c87fdc65e18c2fd46b98a25ac94d1 in master on Jan 28, 2020
In latest commit to this date file digit_01.py missing setup of FONT, CYAN and canvas variables
Sir I want to learn all this.
Download the materials on GitHub and get started! If you have a question, open an Issue or leave a comment. Good luck!