Gaussian Mixture Model

1022 Words5 Pages
II. OBJECT DETECTION USING GAUSSIAN MIXTURE MODEL The Gaussian mixture model is a Gaussian probability density function. GMM can assess any form of the density distribution that is the reason it is frequently utilized in image processing recently for better yield [4]. GMMs can be applied as a parametric model of the probability distribution associated with continual measurements or even attributes in the biometric process, for example color based tracking of the objects in video. In many computer related vision technology, it is critical to identify moving objects from a sequence of videos frames. In order to achieve this, background subtraction is applied which mainly identifies moving objects from each portion of video frames. Background…show more content…
Although it's not a quick and easy process to detect or monitor the objects. There are many techniques and papers introduced by many scientists for the backend process in the video surveillance. Different automated software’s are used for the analysis of the video footage. A. Background Subtraction Background subtraction is the process of separating out foreground objects from the background in a sequence of video frames. It’s a pixel-by-pixel subtraction of the current image from a previously known background image [5]. Background subtraction is used in many emerging video applications, such as video surveillance which is one of today's popular applications, traffic monitoring, and gesture recognition for human-machine interfaces. Background Subtraction is based on four important steps which are given below: 1) Preprocessing Preprocessing is used to eliminate noise that occurs due to any factor like different light intensity. 2) Background Modeling Background model will be strong against the environmental changes in the background and over-protecting against identifying all moving objects of…show more content…
BLOB ANALYSIS For image processing, a blob is defined as a region of connected pixels. These regions of images are identified and studied by Blob analysis. The Blob analysis differentiates the foreground (typically pixels with a non-zero value) or the background (pixels with a zero value). Mostly, area, centroid, bounding box and perimeter are the calculated features in blob analysis. The foreground detector is used to segment moving cars from the background model. Normally a binary mask, assign the pixel value of 1 to the foreground and the value of 0 to the background. IV. KALMAN FILTER A Kalman filter is used to estimate the distributed Gaussian state of a linear system. There are two steps involved in Kalman filter, one prediction and other is correction. The prediction step use to predict the new state of the object. The correction step use to update the object’s state. Updated state of the object is also distributed by a Gaussian. V. RELATED WORK To analyze traffic pattern or speed checking of whole traffic can be done by tracking and detecting cars. Many researchers have proposed different techniques to detect cars. Mostly used methods are foreground detector and blob analysis to detect and count cars in a video sequence. The camera is stationary only cars are moving to detect their

More about Gaussian Mixture Model

Open Document