Timber moisture detection using wavelet packet decomposition and convolutional neural network

Abstract

As timber structures are vulnerable to degradation due to the tendency to trap moisture, the present study proposed a new percussion-based method to replace the existing constant contact between structures and sensors. A total of two approaches have been proposed to automated detect the moisture content (MC) of timber: (a) the random forest classifier (machine learning-based) was employed to classify the wavelet packet decomposition (WPD) features extracted from excitation-induced sound signals (WPD + RF); and (b) the 2D-CNN framework (deep learning-based) was employed to classify the Mel frequency cepstral coefficient (MFCC) features extracted from excitation-induced sound signals (MFCC + 2DCNN). The proposed automatic detection methods are covered from 1D time-domain signal classification to 2D image classification. To verify the effectiveness of both two approaches, an experimental study was conducted. The MC of two types of timber specimens (i.e. softwood and hardwood) was gradually increased from 0% to 60% with 10% increments. The change of MC of timber material caused different material properties, resulting in a measurable differential in forced vibration among the various specimens used. The results demonstrated that MFCC + 2DCC outperformed the RF + WPD in MC classification of timber material. Overall, the percussion-based method proposed in this study can provide an outstanding classification performance.

Near real‐time bolt‐loosening detection using mask and region‐based convolutional neural network

“…These drawbacks make conventional SHM methods not convenient for large‐scale industrial applications. It is found that machine learning shows great potential in structural health monitoring (SHM) 11–17 . For simplification, Kong et al 18 proposed a percussion‐based nondestructive method to detect the health condition of bolted joints by using the machine learning method.…”

A novel intelligent inspection robot with deep stereo vision for three-dimensional concrete damage detection and quantification

“…[23][24][25] However, the application of sensorbased acquisition networks can be prohibitively timeconsuming and labour-intensive, and long-term monitoring may prove impractical due to the need to store massive amounts of data. [26][27][28][29][30] Machine learning in computer vision demonstrated rapidity and reliability for conducting image-based inspection of the concrete surface. 8,[31][32][33][34][35][36] Deep learning with robustness against noise disturbance has been applied to accurately interpret images and sensing data for crack detection.…”

“…23–25 However, the application of sensor-based acquisition networks can be prohibitively time-consuming and labour-intensive, and long-term monitoring may prove impractical due to the need to store massive amounts of data. 26–30…”

An innovative deep neural network–based approach for internal cavity detection of timber columns using percussion sound

“…With the rapid improvement of engineered wood products, timber has also been widely constructed for tall buildings and long-span structures which offers more functions in human society. [1][2][3] However, timber is a biomass construction material, resulting that its mechanical properties are highly influenced by environmental variations and biological actions. Internal cavity eroded by termites is a common degradation pattern, that may cause unexpected structural failures of timber structures.…”