Now that your biped is up and mobile, is able to find barriers, and knows how to plan its path, you can now start to have it move around autonomously. However, you may want your robot to follow a color or motion.
In this chapter, you will be learning:
How to add a webcam to your biped robot
How to add RaspiCam to your biped robot
How to install and use OpenCV, an open source vision package
How to follow motion with your biped robot
Having vision capability is a real advantage for your biped robot; you'll use this functionality in lots of different applications. Fortunately, adding hardware and software for vision is both easy and inexpensive. There are two choices as far as vision hardware is concerned. You can add a USB webcam to your system, or you can add RaspiCam, a camera designed specifically for Raspberry Pi.
Connecting a USB camera is very easy. Just plug it into the USB slot. To make sure that your device is connected, type lsusb. You should see the following:
This shows a Creative Webcam located at Bus 001 Device 004: ID 041e:4095. To make sure that the system sees this as a video device, type ls /dev/v* command and you should see something like the following:
The /dev/video0 is the webcam device. Now that your device is connected, let's actually see if you can capture images and video. There are several tools that can allow you to...
Now that you have your camera connected, you can begin to access some amazing capabilities that have been provided by the open source community. Open a terminal window and type the following commands:
sudo apt-get update: You're going to download a number of new software packages, so it is good to make sure that everything is up to date.
sudo apt-get install build-essential: Although you may have done this earlier, this library is essential to build OpenCV.
sudo apt-get install libavformat-dev: This library provides a way to code and decode audio and video streams.
sudo apt-get install ffmpeg: This library provides a way to transcode audio and video streams.
sudo apt-get install libcv2.4 libcvaux2.4 libhighgui2.4: This command shows the basic OpenCV libraries. Note the number in the command. This will almost certainly change as new versions of OpenCV become available. If 2.4 does not work, either try 3.0 or search on Google...
Fortunately, one of the examples in the OpenCV Python set is a program named edge.py. The following is that file (with blank lines removed):
This program uses the Canny image detection algorithm implemented by OpenCV to find the edges in any image. For more on the Canny edge algorithm, refer to http://dasl.mem.drexel.edu/alumni/bGreen/www.pages.drexel.edu/_weg22/can_tut.html or http://opencv-python-tutroals.readthedocs.org/en/latest/py_tutorials/py_imgproc/py_canny/py_canny.html. You captured an image earlier; you can use this program to look at the edges and to also see how setting a different threshold can show more/less edges. Run the program with the image captured earlier and you will see the following:
You will notice that there is a threshold slide bar setting at the top. If you adjust this threshold up, it will find fewer edges—the edges that have a larger threshold. The picture for a setting of 30 is as follows:
OpenCV and your webcam can also track colored objects. This will be useful if you want your biped to follow a colored object. OpenCV makes this amazingly simple by providing some high-level libraries that can help us with this task. To accomplish this, you'll edit a file to look something like what is shown in the following screenshot:
Let's look specifically at the code that makes it possible to isolate the colored ball:
hue_img = cv.CvtColor(frame, cv.CV_BGR2HSV): This line creates a new image that stores the image as per the values of hue (color), saturation, and value (HSV), instead of the red, green, and blue (RGB) pixel values of the original image. Converting to HSV focuses our processing more on the color, as opposed to the amount of light hitting it.
threshold_img = cv.InRangeS(hue_img, low_range, high_range): The low_range, high_range parameters determine the color range. In this case, it is an orange ball, so you want to detect the color orange. For a good...
© 2015 Packt Publishing Limited All Rights Reserved
