Special issue (part II) on parallel computing for real-time image processing

Mohamed Akil, Laurent Perroton
2011 Journal of Real-Time Image Processing  
The performance requirements of image processing applications have continuously increased, especially when they are executed under real-time constraints. We have organized this special issue on Parallel Computing for Real-Time Image Processing to present the current state-of-the-art in the field of parallel programming and the future trends in realtime image and video processing as related to parallel computing or real-time implementation of embedded image processing applications on parallel
more » ... hitectures including multi-core platforms, graphics processors units (GPUs), and dedicated parallel architectures. Due to the overwhelming number and their wide scope of submissions received for this special issue and thus the difficulty associated with finding expert reviewers, it was decided to offer this special issue in three parts as stated in the first part of this special issue. The papers that are currently under review will appear in the third part. We are very grateful to the reviewers who provided valuable comments and suggestions to improve the quality of the accepted papers. This second part of this special issue on Parallel Computing for Real-Time Image Processing presents five papers addressing different parallel applications including 2-D digital curve approximation, object detection, Bayesian real-time perception, real-time medical video application and improvement of image quality. The real-time implementation of these applications uses GPU and FPGA platforms. Brief outlines of these papers are stated below: The first paper by Damiand and Coeurjolly presents a generic topological and geometrical framework, which allows defining and controlling several parallel algorithms for 2-D digital curve approximation. The proposed technique is based on combinatorial map simplifications guided by geometrical criteria. The genericity of the framework is illustrated by defining three contour simplification methods: a polygonal approximation one based on area deviation computation; a digital straight segments reconstruction, which guarantees to obtain a loss-less representation; and a moment preserving simplification, which simplifies the contours while preserving geometrical moments of the image regions. Thanks to a complete experimental evaluation, the authors demonstrate that the proposed technique can be efficiently implemented in a multi-thread environment to simplify labelled image contours. The second paper by Herout, Jošth, Juránek, Havel, Hradiš and Zemčík presents the acceleration of object detection in images and video sequences using GPU. It includes algorithmic modifications and adjustments, constructing variants of efficient implementations and evaluation comparing to implementations on the CPU's. This article focuses on detection by statistical classifiers based on boosting. The implementation and the necessary algorithmic alterations are described, followed by experimental measurements of the created object detector and discussion of the results. The final solution outperforms the reference-efficient CPU/SSE implementation *6-89 for high-resolution videos using nVidia GeForce 9800GTX and Intel Core2 Duo E8200. The third paper by Ferreira, Lobo and Dias presents the real-time implementation of a Bayesian framework for robotic multisensory perception on a graphics processing unit (GPU) using the compute unified device architecture M. Akil (
doi:10.1007/s11554-011-0210-0 fatcat:hwmsn52suzakdhvfemgo3asnim