Automatic Analysis System for Graphite Morphology of Grey Cast Iron
Author Information
Author(s): Hong Jiang, Yiyong Tan, Junfeng Lei, Libo Zeng, Zelan Zhang, Jiming Hu
Primary Institution: Wuhan University
Hypothesis
Can an automatic image-processing system improve the classification and quantitative analysis of graphite morphology in grey cast iron?
Conclusion
The developed software system successfully applies artificial neural network technology to achieve quantitative analysis of grey cast iron's graphite morphology.
Supporting Evidence
- The system achieved a checkout correct rate of 92% after training with 5000 iterations.
- Using a combined feature vector significantly increased the precision of recognition compared to using a single feature.
- The method for determining the number of hidden nodes in the ANN was empirically validated with experimental data.
Takeaway
This study created a computer program that can automatically analyze and classify different types of graphite in cast iron, making it easier and faster than doing it by hand.
Methodology
The study used an artificial neural network (ANN) for classification based on texture features extracted from images of graphite morphology.
Limitations
The segmentation regions were not very smooth, and the precision of the classification depends on the amount of expert knowledge available.
Statistical Information
P-Value
0.001
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website