Nonparametric background modelling and segmentation to detect micro air vehicles using RGB-D sensor

Abstract

A novel approach to detect micro air vehicles in GPS-denied environments using an external RGB-D sensor is presented. The nonparametric background subtraction technique incorporating several innovative mechanisms allows the detection of high-speed moving micro air vehicles by combining colour and depth information. The proposed method stores several colour and depth images as models and then compares each pixel from a frame with the stored models to classify the pixel as background or foreground. To adapt to scene changes, once a pixel is classified as background, the system updates the model by finding and substituting the closest pixel to the camera with the current pixel. The background model update presented uses different criteria from existing methods. Additionally, a blind update model is added to adapt to background sudden changes. The proposed architecture is compared with existing techniques using two different micro air vehicles and publicly available datasets. Results showing some improvements over existing methods are discussed.

Moving Object Detection Using Adaptive Blind Update and RGB-D Camera

“…Using color and depth data in nonparametric methods such as ViBe to detect moving object is not new and this approach has previously been applied in [10] [11][12] [1]. However, the vision system that the authors used in [10] and [11] consist of a separate TOF (Time of flight) sensor and RGB camera.…”

“…Some other authors such as [12] tried a standard stereo camera. Recently, a new approach proposed in [1] which incorporates innovative mechanisms to detect micro air vehicles (MAVs) in GPS-denied environments using an external RGB-D sensor. This method stores several color and depth frames as a model and then compares individual pixel from a frame with the stored models to identify the pixel as a foreground or background.…”

“…In the last few years low cost RGB-D sensors such as Microsoft Kinect attracted many researchers to detect and recognize objects and behaviors such as motion [13], Suicide [14] ,drone [1] and obstacle detection [15]. These devices are able to produce well calibrated RGB and depth frames and normally capture 30 frames per second which is appropriate for motion detection algorithms.…”

“…The proposed method in this paper is an improved method introduced in [1]. The main novelties and contributions of this research are as following; 1) Adding adaptive blind update method which changes the frequently of blind update based on the speed of moving object.…”

“…n the past few years background subtraction techniques have shown a growing interest among researchers in video surveillance applications [1]. Typically, the main goal of these applications are to separate the moving object known as foreground from other objects in the scene called background [2].…”