The Gaussian Mixture Model

Automatic vehicle detection and classification is a resource for automatically determining and transmitting the geographic location of a vehicle movement and counting the number of vehicles passed in a particular road. It is an important observation for managing passes of vehicles such as service vehicles, emergency vehicles and especially valuable production equipment and also public transportation vehicles. It is also used to track mobile resources, such as non-wheeled construction tools, non-motorized trailers. The other application of automatic vehicle detection is to manage a large driver and crewing staff effectively. The calculation of number of vehicles passed is mainly applied to determine the traffic system in a particular lane.

The adaptive thresholding concept is applied on the frames for noise removal process and binarization of the image. The binarization is defined as the process of converting a gray level image into bi-level image with the threshold value. The noise free frames which are in the gray scale are forwarded for the image binarization completion. The pixel values are calculated for each segment of the image in the frame. An optimal threshold value is determined by using adaptive thresholding for background subtraction and foreground detection. Depending upon the threshold value, the gray level image is converted into binary image which only contains background and foreground

The noise free frames are forwarded for the image binarization accomplishment where the separation of pixels is carried out. From the pixel separation, an optimum threshold is calculated for background and foreground separation. Depending upon the threshold value, the gray level image is converted into binary image which only contains background and foreground element. Optical flow algorithm is proposed for the feature selection process in a video. It introduces image motion and 2D optical flow assessment. The motion estimation is applied to the sequence of images of both dissipated image and distinct image. Once the image is selected by the optical flow algorithm, feature extraction is carried out by using the PSO algorithm (Particle Swam Optimization). The parameters such as position and velocity of the particle are used in the extraction process algorithm. From these parameters, the feature is extracted for the further analysis. Image classification is an important issue in the analysis of moving image particle. In this proposed concept, ANFIS algorithm is used in the feature classification process although this work emphasizes on ANFIS as a classifier, in addition, it can be used for other different tasks like de-noising and