First we gather a data set of all the hand-shapes we wish to recognise. A naive approach to recognizing a new image D would be to simply compare it with all the images stored in the data set and find the target image T with the closest match. But because there are so many images in the data set this will take far too long. We can reduce the time by using a multi-scale approach. We divide up the data set into groups of images, which are similar to one another by blurring the images at different levels so that small differences between similar images will be eroded.
This post has been scheduled for ages since i posted an?announcement in a previous post to write a review of openCV of detection and recognition, particularly in perspectives of hand detections and recognitions. The announcement post is here as a reference.
About openCV
A library of programming functions for real time computer vision,released under a BSD?license, and free for academic and commercial use (a). Current version is openCV 2.1, this version has been installed and compiled under my VS 2008 environment but with some errors, may not be very easy to configure this version. While the stabilities of openCV 1.0 have been firmly proved already and that is easier to install and configure. Install guides can be found as ?openCV install guide in openCV wiki,?installing openCV in windows, vc2008 install openCV 2.0/2.1(in Chinese)?and get started with openCV.
Using media player control in MFC is not as easy as dragging in a button control. Usually there are some high frequent errors such as ‘cannot convert from LPDISPATCH to CWMPMedia/CWMPControls’, and the worse thing is that even Google is not able to provide proper solutions to address these errors except dozens of posts raising similar problems waiting for somebody to provide positive solutions.
Popular tutorials of using openCV to detect objects and showing recognised results are based on a black window with rapid command lines flushing up. As MFC has more powerful functions in interface design than pure interfaces provided by openCV, it is attractive to move openCV application from console to MFC. Here is an tutorial of introducing using openCV in MFC, check it out.
Using the function cvsnakeimage(), this was an idea i tried to recognise hand gestures. The thought was that as to set a threshold to filter the image firstly, based on the image then contours were able to be extracted (as blue curves shown in the picture on the left), then passing contours points to ‘cvsnakeimage’ function to calculate the green curves which were snake curves.
HCI, Quote, Research
Comment now First we gather a data set of all the hand-shapes we wish to recognise. A naive approach to recognizing a new image D would be to simply compare it with all the images stored in the data set and find the target image T with the closest match. But because there are so many images in the data set this will take far too long. We can reduce the time by using a multi-scale approach. We divide up the data set into groups of images, which are similar to one another by blurring the images at different levels so that small differences between similar images will be eroded.
