Obstacle Avoidance At Night

Computer Vision Solutions

In this blog, we demonstrate how to run our obstacle detection algorithm using raspberry pi and a Sphero RVR rover. Obstacle detection is involves classifying meaningful pixels as obstacles and identifying traversable paths. Sample images were obtained from Raspberry Pi camera. The setup is shown below.

We run a python code to obtain the output, the API source code can be found below.

You’ll need to have a subscription key to use this, you can check out the intro blog on how to get your free subscription key.

We input the sample image as a base64 image. The output will give back a base64 obstacle map image if the processing is successful. A sample input image and the response is shown below.

The output shows connected obstacles, the black region is where no obstacles were detected and the pixels with colour represent obstacles. The colors represent the connected obstacles. A single obstacle can be represented with multiple colors. In our next blog, we show how to use the above obstacle map to move while avoiding obstacles.