Traffic signs and Object detection app

    Traffic signs and Object detection was the main purpose the app was built for. This module is in its development stage. You can fully use it, but we're still working to improve it.

    Right now, the app comes with stop and yield sign detection capabilities. But theoretically you can use it to detect any object, if you upload the corresponding detection file.

    What is a detection file? MyGuardCam app uses OpenCV for android as a computer vision library. The detection is done with the help of xml files. This are HAAR or LBP cascade classifiers. More on what is a cascade classifier.
***Important: HAAR cascade classifiers are more accurately but take more time to obtain and the detection video preview has lower fps. LBP cascade classifiers are less accurate but take less time to obtain and the detection video preview has higher fps.

How can I build a xml cascade classifier? We will show you two ways:

1. You can use a dedicated software like Cascade Trainer GUI. Click here to get it and see the guide.
***Important: keep in mind the differences between HAAR and LBP. We recommend you train for LBP, at least, in the beginning. HAAR training takes a lot of time and processing power.

2. Using OpenCV library. This is a little harder to use because it involves a lot of steps and it's not that intuitive. Click here to see the guide.

We recommend you use option 1. After you obtain the xml file and want to use it, you have to take two steps:

1. Go to , login, my dashboard and My Detection Files. Here you have the option to upload your xml file to the server.

2. After the file is uploaded you can click "SYNC" from the app settings and the file will be downloaded to your device. To use it on the app, from settings, click "Detect custom shape -> DCIM -> GuardCam -> DetectFiles", from here select the desired xml file. And that's it! If you want to use another file, just uncheck Detect custom shape box, check it again and select another file.
***Important: "GuardCam" folder may not be located, on all devices, in the DCIM folder and you have to look for it.