Median and Morphological Specialized Processors for a Real-Time Image Data Processing
- PDF / 569,147 Bytes
- 7 Pages / 600 x 792 pts Page_size
- 41 Downloads / 168 Views
edian and Morphological Specialized Processors for a Real-Time Image Data Processing Kazimierz Wiatr Institute of Electronics, AGH Technical University of Cracow, Mickiewicza 30, 30-059 Krakow, Poland Email: [email protected] Received 29 July 2001 and in revised form 12 October 2001 This paper presents the considerations on selecting a multiprocessor MISD architecture for fast implementation of the vision image processing. Using the author’s earlier experience with real-time systems, implementing of specialized hardware processors based on the programmable FPGA systems has been proposed in the pipeline architecture. In particular, the following processors are presented: median filter and morphological processor. The structure of a universal reconfigurable processor developed has been proposed as well. Experimental results are presented as delays on LCA level implementation for median filter, morphological processor, convolution processor, look-up-table processor, logic processor and histogram processor. These times compare with delays in general purpose processor and DSP processor. Keywords and phrases: image processing, median specialized processor, morphological specialized processor, real-time vision system.
1.
Operations sequence
TASKS OF REAL-TIME IMAGE ANALYSIS
The vision signal real-time processing for the needs of control systems require high computation powers. Therefore, methods for fast implementation of the processing algorithms have to be looked for. In the bibliography, many attempts can be found to formulate the algorithms so that their implementation time is as short as possible. However, the author took an effort to search qualitatively different solutions which would be on the one hand related to using specialized hardware structures for implementing various operations and on the other hand to use such intercommunication of those that the architecture is as effective as possible. Here below, the structure of the vision system is presented and its tasks are singled out. In the algorithms of image analysis, several levels of image processing can be singled out [1, 2]. Most often, three levels are provided (Figure 1). The lowest level of image analysis (I), called the vision signal preprocessing is aimed at: eliminating the interference, drawing the object out of its background, edge detection, adjusting the object greyness level from the histogram, histogram balancing, and so forth. The middle level of image analysis (II) performs the image segmentation, the object localization, recognizes the image shape and singles out the shape specific features. The highest level (III) is the analysis of the complicated scene: the object movement detection, the object current control, presetting the parameters for low and middle level image processing and analysis.
I-MISD ARCHITECTURE A/D
P
P
P
II
III
Algorithm Parameters OUT
Object control
Figure 1: A real-time image processing levels.
The vision system structure in Figure 1 shows the feedbacks between various levels of the vision system. The results
Data Loading...
