Improved GMM‐based method for target detection

Abstract

The Gaussian mixture model (GMM) is prone to large-scale misdetection in the static case where the background and foreground have similar colours. This study presents an improved GMM method to solve this problem. First, the principal component analysis is used to transform the high-dimensional space into the low-dimensional one with three colour channels, which aims to reduce runtime. Then, the images are processed by GMM to obtain the foreground areas. At the same time, the mean and difference of pixel features in red, green and blue and hue, saturation and value (HSV) colour models are calculated to analyse the similar-coloured components. The areas with similar colours are classified in this process, and thus the foreground is separated from similar-coloured background. Then, all of the extracted areas are targets. The experimental results show that the presented method identified the target areas with high accuracy and high efficiency. Compared with conventional GMM, the presented method obtained the detection-accuracy improvement by about 6% on average.

A Novel User-Friendly Application for Foreground Detection with Post-Processing in Surveillance Video Analytics

“…The proposed method enhanced the convergence speed and classification performance. Zhao et al [2] combined Principal Component Analysis (PCA) with the GMM for target detection. They proved that the method identified the targets with high accuracy and high efficiency.…”

“…[5]. Many algorithms exist for object detection [2][3] [11] [17]. The main objective of this system is to provide the simplest benchmark method for foreground detection to be applicable in real-time.…”

An intelligent joint filter for vector tracking loop considering noise interference

“…As a surprised Machine learning theory, GMM clustering is a distribution-based algorithm. In GMM, the probability density distribution of samples can be determined by the weighted sum of several Gaussian distribution functions [17,18]. Compared with the existing works, the main contributions of this letter could be summarized as follows:…”