This article from http://blog.csdn.net/hellogv/, reference must indicate the source!
Recently, more and more people are looking at object motion recognition (for example, in the sixth sense), and I am also a little idle recently ..... Most people use opencv, so I will not repeat the work. I will change to another open-source class library ~~~ Aforge.
About aforge from Baidu: aforge. net is designed for developers and researchers based on the C # framework. It covers computer vision and AI, image processing, neural networks, genetic algorithms, machine learning, robotics, and other fields. This framework consists of a series of class libraries and examples. The features include:
Aforge. Imaging-some daily image processing and filters
Aforge. Vision-computer vision application library
Aforge. Neuro-neural network computing Library
Aforge. Genetic-Evolutionary Algorithm Programming Library
Aforge. machinelearning-Machine Learning class library
Aforge. Robotics-provides some machine learning tool Libraries
Aforge. Video-a series of video processing class libraries (very convenient)
PS: The algorithm in aforge is enough for me to play for a long time .....
The example code modified from the http://www.aforgenet.com/articles/step_to_stereo_vision/, cut part of the official code, added a more concise object recognition function... For example:
At the beginning, I haven't told the program which is the object to be recognized. After clicking the mouse on the green bottle cap, the program will know that the green bottle cap is the object to be recognized ...... Why should we use green? Because the green finger covers are the easiest to do, it is okay to find two leaves to stick together. It is not recommended to use red because the human body is too red for identification.
The code for this example can be downloaded here in: http://download.csdn.net/source/2281943, aforge Class Library Please download here in http://code.google.com/p/aforge/downloads/list.