Automation of Vehicle Theft Identification System Essay

1483 Words 6 Pages
In recent years, vehicle thefts are most important crux in the world. At the present time, the available anti theft vehicular systems are much more expensive. Many people are installed the vehicle theft control system in their vehicle for avoiding the theft. Here, the main goal is to reduce the vehicle theft with the developing of simple and low cost control scheme. The main components of this scheme are microcontroller, one hidden camera, a GPS and one mobile. The mobile makes the communication between the vehicle owner and the control module placed in the vehicle. The MMS is the one of the most and popular application in the mobile. Everyone has to be used that application easily. In this scheme the MMS helps to identify the thieves’ …show more content…
The vehicle location information is sent to the owner’s mobile. So, the vehicle owner easily identifies the theft vehicle. Here ARM 7 controller has been used.

In the ICU section, the first stage of recognition starts with face detection module. The face detection helps to crop only the face region. The cropped face region has the normalized intensity size and shape of the original face image. The granulation process and feature helps to match the face image before and after the face detection process.The Gaussian operator performs the convolution of each of the constituent images iteratively with a 2-D Gaussian kernel and generates the low pass filtered image sequences. Then, DOG pyramid will be formed from successive iterations of Gaussian images. To provide the edge information, noise, smoothness, blurriness and noise present in a face image, the facial features are segregated at different resolutions by this granulation approach. In features extraction stage, WLD descriptor represents an image as a histogram of differential excitations and gradient orientations, and has several properties like noise and illumination changes, detection of edges and powerful image representation. These features are useful to distinguish the maximum number of samples accurately and it is matched with already stored original face samples for identification. The simulated result will be compared with the data base image features. Finally the

Related Documents