Testing and improving <cascade>
by Michael Beyeler, Prateek Joshi, Joseph Howse
OpenCV: Computer Vision Projects with Python
- OpenCV: Computer Vision Projects with Python
- Table of Contents
- OpenCV: Computer Vision Projects with Python
- OpenCV: Computer Vision Projects with Python
- Credits
- Preface
- 1. Module 1
- 1. Setting up OpenCV
- Choosing and using the right setup tools
- Running samples
- Finding documentation, help, and updates
- Summary
- 2. Handling Files, Cameras, and GUIs
- 3. Filtering Images
- 4. Tracking Faces with Haar Cascades
- 5. Detecting Foreground/Background Regions and Depth
- A. Integrating with Pygame
- B. Generating Haar Cascades for Custom Targets
- 1. Setting up OpenCV
- 2. Module 2
- 1. Detecting Edges and Applying Image Filters
- 2. Cartoonizing an Image
- 3. Detecting and Tracking Different Body Parts
- 4. Extracting Features from an Image
- Why do we care about keypoints?
- What are keypoints?
- Detecting the corners
- Good Features To Track
- Scale Invariant Feature Transform (SIFT)
- Speeded Up Robust Features (SURF)
- Features from Accelerated Segment Test (FAST)
- Binary Robust Independent Elementary Features (BRIEF)
- Oriented FAST and Rotated BRIEF (ORB)
- Summary
- 5. Creating a Panoramic Image
- 6. Seam Carving
- 7. Detecting Shapes and Segmenting an Image
- 8. Object Tracking
- 9. Object Recognition
- 10. Stereo Vision and 3D Reconstruction
- 11. Augmented Reality
- 3. Module 3
- 1. Fun with Filters
- 2. Hand Gesture Recognition Using a Kinect Depth Sensor
- 3. Finding Objects via Feature Matching and Perspective Transforms
- 4. 3D Scene Reconstruction Using Structure from Motion
- 5. Tracking Visually Salient Objects
- 6. Learning to Recognize Traffic Signs
- 7. Learning to Recognize Emotions on Faces
- A. Bibliography
- Index
<cascade> is an XML file that is compatible with the constructor for OpenCV's CascadeClassifier class. For
an example of how to use CascadeClassifier, refer back to our implementation of FaceTracker in Chapter 4, Tracking Faces with Haar Cascades. By copying and modifying FaceTracker and Cameo, you should be able to create a simple test application that draws rectangles around tracked instances of your custom target.
Perhaps in your first attempts at cascade training, you will not get reliable tracking results. To improve your training results, do the following:
- Consider making your classification problem more specific. For example, a
bald, shaven, male face without glassescascade might be easier to train than a generalfacecascade. Later, as your results improve, you can try to expand your problem again. - Gather more training images, many more!
- Ensure that
<negative_description>contains all the negative training images and only the negative training images. - Ensure that
<positive_description>contains all the positive training images and only the positive training images. - Ensure that the sub-rectangles specified in
<positive_description>are accurate. - Review and experiment with the optional flags to
<opencv_createsamples>and<opencv_traincascade>. The flags are described in the official documentation at http://docs.opencv.org/doc/user_guide/ug_traincascade.html.
Good luck and good image-hunting!
-
No Comment
..................Content has been hidden....................
You can't read the all page of ebook, please click here login for view all page.