Image difference threshold strategies and shadow detection
BMVC '95 Proceedings of the 1995 British conference on Machine vision (Vol. 1)
Real-time thresholding with Euler numbers
Pattern Recognition Letters
Thresholding for Change Detection
ICCV '98 Proceedings of the Sixth International Conference on Computer Vision
Synthesis of Arithmetic Circuits: FPGA, ASIC and Embedded Systems
Synthesis of Arithmetic Circuits: FPGA, ASIC and Embedded Systems
Optimal binarization of images by neural networks for morphological analysis of ductile cast iron
Pattern Analysis & Applications
Embedded active vision system based on an FPGA architecture
EURASIP Journal on Embedded Systems
A new binarization algorithm based on maximum gradient of histogram
ICIG '07 Proceedings of the Fourth International Conference on Image and Graphics
Local Properties of Binary Images in Two Dimensions
IEEE Transactions on Computers
A parallel hardware architecture for connected component labeling based on fast label merging
ASAP '08 Proceedings of the 2008 International Conference on Application-Specific Systems, Architectures and Processors
Digital image thresholding, based on topological stable-state
Pattern Recognition
Performance Comparison of Graphics Processors to Reconfigurable Logic: A Case Study
IEEE Transactions on Computers
Image change detection algorithms: a systematic survey
IEEE Transactions on Image Processing
Hi-index | 0.00 |
The stable Euler-number-based image binarization has been shown to give excellent visual results for images containing high amount of image noise. Being computationally expensive, its applications are limited mostly to general-purpose processors and in application specific integrated circuits. In this paper a modified stable Euler-number-based algorithm for image binarization is proposed and its real-time hardware implementation in a Field Programmable Gate Array with a pipelined architecture is presented. The proposed modifications to the algorithm facilitate hardware implementation. The end result is a design that out-performs known software implementations. The amount of noisy pixels introduced during the binarization process is also minimized. Despite the stable Euler-number-based image binarization being computationally expensive, our simulations show that the proposed architecture gives accurate results and this in real time and without consuming all chip resources.