You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Copy file name to clipboardExpand all lines: articles/cognitive-services/Custom-Vision-Service/iot-visual-alerts-tutorial.md
+9-9Lines changed: 9 additions & 9 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -1,5 +1,5 @@
1
1
---
2
-
title: "Tutorial: IoT Visual Alert sample"
2
+
title: "Tutorial: IoT Visual Alerts sample"
3
3
titleSuffix: "Azure Cognitive Services"
4
4
description: In this tutorial, ...
5
5
services: cognitive-services
@@ -13,31 +13,31 @@ ms.date: 09/11/2019
13
13
ms.author: pafarley
14
14
---
15
15
16
-
# Tutorial: IoT Visual Alert sample
16
+
# Tutorial: IoT Visual Alerts sample
17
17
18
18
This sample app illustrates how to use Azure Custom Vision to train a device with a camera to detect visual states. You can run this detection scenario on an IoT device by using an ONNX model exported from the Custom Vision service.
19
19
20
20
A visual state describes the content of an image: an empty room or a room with people, an empty driveway or a driveway with a truck, and so on. In the image below, you can see the app detect when a banana or an apple is placed in front of the camera.
21
21
22
-

22
+

23
23
24
24
This tutorial will show you how to:
25
25
> [!div class="checklist"]
26
-
> * Configure the sample app to use your Custom Vision and IoT Hub resources.
26
+
> * Configure the sample app to use your own Custom Vision and IoT Hub resources.
27
27
> * Use the app to train your Custom Vision project.
28
-
> * Use the app to score new images in real time and send results to Azure.
28
+
> * Use the app to score new images in real time and send the results to Azure.
29
29
30
30
If you don't have an Azure subscription, create a [free account](https://azure.microsoft.com/free/) before you begin.
31
31
32
32
## Prerequisites
33
33
34
-
*[Visual Studio 2015 or later](https://www.visualstudio.com/downloads/)
* You'll also need to create an IoT Hub resource on Azure.
35
+
* You'll also need to [create an IoT Hub resource](https://ms.portal.azure.com/#create/Microsoft.IotHub) on Azure.
36
+
*[Visual Studio 2015 or later](https://www.visualstudio.com/downloads/)
37
37
* Optionally, an IoT device running Windows 10 IoT Core version 17763 or higher. You can also run the app directly from your PC.
38
38
* For Raspberry Pi 2 and 3, you can set up Windows 10 directly from the IoT Dashboard app. For other devices such as DrangonBoard, you'll need to flash it using the [eMMC method](https://docs.microsoft.com/windows/iot-core/tutorials/quickstarter/devicesetup#flashing-with-emmc-for-dragonboard-410c-other-qualcomm-devices). If you need help setting up a new device, see [Setting up your device](https://docs.microsoft.com/windows/iot-core/tutorials/quickstarter/devicesetup) in the Windows IoT documentation.
39
39
40
-
## About the app
40
+
## About the Visual Alerts app
41
41
42
42
The IoT Visual Alerts app runs in a continuous loop, switching between four different states as appropriate:
43
43
@@ -95,7 +95,7 @@ people, an empty desk, a desk with a toy truck, and so on).
95
95
96
96
Once the app has finished capturing images, it will upload them and then switch to the **Waiting For Trained Model** state. At this point, you need to go to the [Custom Vision portal](https://www.customvision.ai/) and build a model based on the new training images. The following animation shows an example of this process.
97
97
98
-

98
+

99
99
100
100
To repeat this process with your own scenario:
101
101
1. Sign in to the [Custom Vision portal](http://customvision.ai).
0 commit comments