Blob detection is a common task in vision-based applications. Most existing algorithms are aimed at execution on general purpose computers; while very few can be adapted to the computing restrictions present in embedded platforms. This paper focuses on the design of an algorithm capable of real-time blob detection that minimizes system memory consumption. The proposed algorithm detects objects in one image scan; it is based on a linked-list data structure tree used to label blobs depending on their shape and node information. An example application showing the results of a blob detection co-processor has been built on a low-powered field programmable gate array hardware as a step towards developing a smart video surveillance system. The detection method is intended for general purpose application. As such, several test cases focused on character recognition are also examined. The results obtained present a fair trade-off between accuracy and memory requirements; and prove the validity of the proposed approach for real-time implementation on resource-constrained computing platforms.

This paper introduces a statistical background pixel classifier intended for real-time and low-resource implementation. The algorithm works within a smart video surveillance application aimed to detect unattended objects in images with fixed backgrounds. The algorithm receives an input image and builds an initial background model based on image statistics. Using this information, the algorithm identifies new objects that do not belong to the original image. The algorithm categorizes image pixels in four possible classes: shadows, midtones, highlights and foreground pixels. The classification stage produces a binary mask where only objects of interest are shown. The pixel classifier processes Quarter VGA (320 x 240) gray-scale images at a nomial processing rate of 30 frames per second. Higher resolutions such as VGA (640 x 480) have been also tested. We compare results with traditional statistical background modeling methods. Our experiments demonstrate that our approach achieves successful background segmentation at a minimal resource consumption while maintaining realtime execution.