Introduction: Intel Edison Opecv Image Recognition With AfricasTalking SMS Gateway (node Mailer)
In this instruct-able we are using Opencv Image recognition such that when a specific image is recognized a text will be sent to a user using AfricasTalking Sms gateway.This tech can be applied in security such that when an intruder is recognised an sms will be sent to specific authority.How cool is that :-)
Step 1: Step 1: Replacing Yocto Linux With Ubilinux
The Intel Edison comes with Yocto Linux pre-installed. Yocto is a small version of Linux which is suited for many IOT project, but it does not include the Linux sendmail program and Yocto also does not include an apt-get capability to install sendmail, so I chose the alternate route of installing Ubilinux for the Intel Edison. I will not go into details on how to do this since it is documented here: Getting Started with ubilinux in Intel edison
https://www.instructables.com/id/Ubilinux-Installat... and also here: https://www.instructables.com/id/Ubilinux-Installat... Once ubilinux is installed, it is simple to install sendmail using apt-get
apt-get install sendmail
Note: Someone with better Linux expertise may be able to get sendmail to work on Yocto linux, but that can be their Instructable ;)
download ubilinux from Ubilinux Download
Step 2: Step 2: Updating the Mraa Library
The mraa library is used to provide an interface to the Edison I/O. Most of the mraa I/O functions worked when I installed Ubilinux, but the LCD display did not. I followed these steps to update the mraa library
https://learn.sparkfun.com/tutorials/installing-li... Note: This instructable used Ubilinux version: ubilinux-edison-150309. If a later version of Ubilinux includes updates to the mraa library, this step may no longer be needed.
for more visit http://iotdk.intel.com/docs/master/mraa/
Step 3: Step 3: the Node.js Source Code
The code can be found here: Mukira Node-Mailer
After you clone this github repository, you should see the following files and sub-directories:
config/ logs/ LICENSE main.js package.json
main.js: This file contains the main program. package.json: This file contains version information. config/default.json: This is the configuration file. You will need to modify this file to add your email account information. Note: Since file will contain your email account name and password information, you should be careful with the security of this file. (don't allow access to it by unauthorized users). logs: This sub-directory will contain a log of email sent. LICENSE: MIT open-source license. http://en.wikipedia.org/wiki/MIT_License
Step 4: Step 4: Install the Required Node Modules Using Npm Install
The following node.js modules need to be installed (run the following command from within the directory that contains your code)
npm install nodemailer
npm install winston
npm install configAfter this step, your code directory should have a new sub-directory called node_modules config/ logs/ node_modules/ LICENSE main.js package.json
Step 5: Step 5: Update Config/default.json
You will need to update this file with your email account information.
"user": "your_email_address_goes_here", "password": "your_email_password_goes_here", "service": "your_email_service_goes_here",user: This is your email address (the full address with the .com extension) password: Your email password service: "Yahoo" or "Gmail" (other services may be possible but I have not tried them) Note: Since file contains your email account name and password information, you should be careful with the security of this file. (don't allow access to it by unauthorized users).
Step 6: Step 6: Connect the Ip Camera
Follow the instructions for setting up the image recognition in intel edison from my github account https://github.com/mukira/face-detection-node-opencv
Step 7: Step 7: Run the Program
In the code directory, run this command:
If all previous steps were successful, then each time an image is recognized the Edison should send a sms/email to the test address that you configured in the config.default.json file.
Step 8: Step 8: Acknowledgements
Acknowledgment to Graham from Aficastalking with whom the project would be a success