A hybrid morphology processing units architecture for real-time video segmentation systems

In this paper, we propose a hybrid morphology processing units architecture for real-time video segmentation systems. It contains two parts: a gray-level part and a binary part. A partial-result-reuse technique is applied to reduce the hardware cost of gray-level part. For the target of high through...

Full description

Saved in:
Bibliographic Details
Main Authors: Shao-Yi Chien, Yu-Wen Huang, Shyh-Yih Ma, Liang-Gee Chen
Format: Conference Proceeding
Language:English
Subjects:
Background noise
Change detection algorithms
Costs
Filters
Hardware
Morphological operations
Morphology
Noise cancellation
Real time systems
Throughput
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper, we propose a hybrid morphology processing units architecture for real-time video segmentation systems. It contains two parts: a gray-level part and a binary part. A partial-result-reuse technique is applied to reduce the hardware cost of gray-level part. For the target of high throughput and flexibility, the binary part is implemented with a programmable PE array. Simulation shows the proposed hardware architecture is efficient in both hardware complexity and memory organization.
DOI:10.1109/ISCAS.2001.922038