Cost-Effective Video Filtering Solution for Real-Time Vision Systems

Viktor Fischer,Rastislav Lukac,Karl Martin

doi:10.1155/asp.2005.2026

Abstract

This paper presents an efficient video filtering scheme and its implementation in a field-programmable logic device (FPLD). Since the proposed nonlinear, spatiotemporal filtering scheme is based on order statistics, its efficient implementation benefits from a bit-serial realization. The utilization of both the spatial and temporal correlation characteristics of the processed video significantly increases the computational demands on this solution, and thus, implementation becomes a significant challenge. Simulation studies reported in this paper indicate that the proposed pipelined bit-serial FPLD filtering solution can achieve speeds of up to 97.6 Mpixels/s and consumes 1700 to 2700 logic cells for the speed-optimized and area-optimized versions, respectively. Thus, the filter area represents only 6.6 to 10.5% of the Altera STRATIX EP1S25 device available on the Altera Stratix DSP evaluation board, which has been used to implement a prototype of the entire real-time vision system. As such, the proposed adaptive video filtering scheme is both practical and attractive for real-time machine vision and surveillance systems as well as conventional video and multimedia applications.

Highlights

Computer vision methods are becoming increasingly important for the development of novel commercial devices such as wireless phones, vision-based pocket devices, sensor networks, and surveillance and automotive apparatus [1, 2, 3, 4]
The basic structures and blocks utilized in the reduced nonlinear adaptive video filtering (NAVF) have been described in very high-speed integrated circuit hardware description language (VHDL)
Its simple structure suggests the possibility of implementation as a cost-effective field-programmable logic device (FPLD) solution, keeping the majority of available resources unused for the implementation of a compact, modern, integrated computer vision system

Summary

Introduction

Computer vision methods are becoming increasingly important for the development of novel commercial devices such as wireless phones, vision-based pocket devices, sensor networks, and surveillance and automotive apparatus [1, 2, 3, 4] This increases the demand for hardware-based implementations of new, relatively complex video processing algorithms [5]. Due to the complex nature of the noise process, the overall acquisition noise is usually modeled as a zero mean white Gaussian noise [9, 10] Aside from this type of noise, image imperfections resulting from impulsive noise are generated during transmission through a communication channel [11, 12, 13], with sources ranging from human-made sources (switching and interference) to signal representation (bit errors) and natural (atmospheric lightning) ones. Image filtering is of paramount importance [6, 7, 14]

Objectives

Methods

Results

Conclusion