Hand Gesture Recognition based on Depth map


In this Hand Gesture Recognition based on Depth map paper a proposed method for gesture recognition using depth map image using Opencv is presented.  Using feature extraction method based on Radon transform will identify the hand posture recognition.

This method reduces the feature vector of Radon transform by averaging its values. For the classification of features vectors, the support vector machine is used.

Progress in depth map calculation of later year’s leads to exploitation then for several objects of research, one of depth map usage is gesture recognition since it provides information about shape of captured hand as well as position in frame. Depth maps have several advantages over traditional color pictures.

Existing Work:

In the existing the work, in order to capture the image  they have used an external hardware i.e Microsoft Kinect system, using the implementation will on higher cost side as well this system only has done all the preprocessing of the images ,no exact algorithm was implemented to detect the hand.

Proposed Work:

Proposed Hand Gesture Recognition work will be based on a Open Computer Vision library, using an Linux based real time device, will capture the images using a USB Camera connected to the device. And with our algorithms will detect the Open palm and Closed Palm in the capture images and draw a rectangle around it.




ARM11, USB Camera,  Power supply.

Software: OS:

Embedded Linux, Language: C/ C++, IDE: Qt Creator.


 Computer Vision system, Gesture control devices


  • Open source algorithm are used to implement the concept
  • Low cost and has future scope of controlling the appliances on gesture. 

Leave a Reply

Your email address will not be published. Required fields are marked *