Home // ICONS 2015, The Tenth International Conference on Systems // View article

IPOL - A Domain Specific Language for Image Processing Applications

Authors:
Christian Hartmann
Marc Reichenbach
Dietmar Fey

Keywords: DSL, design flow, image processing

Abstract:
Image processing applications are common in every field of our daily life. However, most of them are very complex and contain several tasks with different complexities which result in varying requirements for computing architectures. Nevertheless, a general processing scheme in every image processing application has a similar structure, called image processing pipeline: (1) capturing an image, (2) pre-processing using local operators, (3) processing with global operators and (4) post-processing using complex operations. Therefore, application-specialized hardware solutions combined in a heterogeneous system are used for image processing. To archive this, finding an optimal heterogeneous hardware architecture to meet the image processing application requirements is the central problem and still unsolved. Instead, engineers use languages like VHDL, Verilog, C/C++ and Cuda for designing such systems. But, these kind of languages are not suitable for system analysis - they provide a hardware specific solution for a specific algorithm. Therefore, a holistic modeling of a complete image processing pipeline, with automatic optimization and assignment to different heterogeneous computing cores is not possible. To overcome this problem, we propose in this paper a new domain specific language, called Image Processing Operator Language (IPOL). This description language contain all needed components hardware components like Sensors, Displays, execution units and software parts like image processing algorithms.

Pages: 40 to 43

Copyright: Copyright (c) IARIA, 2015

Publication date: April 19, 2015

Published in: conference

ISSN: 2308-4243

ISBN: 978-1-61208-399-5

Location: Barcelona, Spain

Dates: from April 19, 2015 to April 24, 2015