Detection of wood defects types based on Hu invariant moments and BP neural network

作者：Qi, Dawei^*; Mu, Hongbo

来源：Journal of Southeast University(Natural Science Edition), 2013, 43(SUPPL.I): 63-66.

DOI：10.3969/j.issn.1001-0505.2013.S1.013

• Automatic Detection
• BP (back propagation) neural network
• BP neural networks
• Hu invariant moments
• Rotation invariance
• Scaling invariance
• Translation invariance
• Transmitted intensities

X-ray was adopted as a measure method for wood nondestructive testing. Wood defects were identified by testing X-ray transmitted intensity through the wood. The detected defects were conducted by image processing. Wood defect images were first converted into grayscale images, and then into binary images. With the threshold values determined by some known experience, the wood defects were separated from the background and the clear wood defects edge was extracted. A group of parameters describing shape features were obtained by extending Hu invariant moments. Those parameters not only have translation invariance, scaling invariance and rotation invariance, but also have lower computational complexity. The feature parameters were input into BP (back propagation) neural network after preprocessing, and then the wood defects were recognized. The experimental results show that the recognition ratio is above 86%, indicating that this method is successful for detection and classification of wood defects. This study offers a new method for automatic detection of wood defects.

• 出版日期2013-7
• 单位东北林业大学