Image Reconstruction of Internal Defects in Wood Based on Segmented Propagation Rays of Stress Waves

Du, Xiaochen; Li, Jiajie; Feng, Hailin; Chen, Shengyong

doi:10.3390/app8101778

Cited by 38 publications

(26 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On this basis, the cross-sectional image of the tested tank can be reproduced. The main problem of UST is the difficulty of building a physical model that would be able to take into account the full complexity of acoustic phenomena occurring in a relatively small space [8]. This causes problems in implementing classic tomographic methods.…”

Section: Tomographic Methods and Algorithmsmentioning

confidence: 99%

Maintenance of industrial reactors supported by deep learning driven ultrasound tomography

Kłosowski

Rymarczyk

Kania

et al. 2020

Eksploatacja i Niezawodność – Maintenance and Reliability

View full text Add to dashboard Cite

Maintenance of industrial reactors supported by deep learning driven ultrasound toMography eksploatacja reaktorów przeMysłowych ze wspoMaganieM toMografii ultradźwiękowej i algorytMów głębokiego uczeniaMonitoring of industrial processes is an important element ensuring the proper maintenance of equipment and high level of processes reliability. The presented research concerns the application of the deep learning method in the field of ultrasound tomography (UST). A novel algorithm that uses simultaneously multiple classification convolutional neural networks (CNNs) to generate monochrome 2D images was developed. In order to meet a compromise between the number of the networks and the number of all possible outcomes of a single network, it was proposed to divide the output image into 4-pixel clusters. Therefore, the number of required CNNs has been reduced fourfold and there are 16 distinct outcomes from single network. The new algorithm was first verified using simulation data and then tested on real data. The accuracy of image reconstruction exceeded 95%. The results obtained by using the new CNN clustered algorithm were compared with five popular machine learning algorithms: shallow Artificial Neural Network, Linear Support Vector Machine, Classification Tree, Medium k-Nearest Neighbor classification and Naive Bayes. Based on this comparison, it was found that the newly developed method of multiple convolutional neural networks (MCNN) generates the highest quality images.

show abstract

Section: Tomographic Methods and Algorithmsmentioning

confidence: 99%

Maintenance of industrial reactors supported by deep learning driven ultrasound tomography

Kłosowski

Rymarczyk

Kania

et al. 2020

Eksploatacja i Niezawodność – Maintenance and Reliability

View full text Add to dashboard Cite

show abstract

“…3 GM-Mask R-CNN is glance network and multiple channels mask region convolution neural network. 4 FCN is fully convolutional neural network. Figure 12 shows the detective results for each image including the types and locations of the defects.…”

Section: Classification Performance Evaluationmentioning

confidence: 99%

“…A fast wood defect detection method is necessary for modern wood veneer processing industries to improve their wood use rate and increase their revenue. Currently, many kinds of technologies have been used to detect defects on wood veneers, including air-coupled ultrasonic technology [3], stress wave technology [4], 3D laser technology [5], computed tomography [6], and computer vision technology [7]. Air-coupled ultrasonic technology is a contactless ultrasonic measurement method that detects defects inside the wood veneer based on defects' densities variations inside the wood.…”

Section: Introductionmentioning

confidence: 99%

Defect Detection of Industry Wood Veneer Based on NAS and Multi-Channel Mask R-CNN

Shi

Zhu

et al. 2020

Sensors

View full text Add to dashboard Cite

Wood veneer defect detection plays a vital role in the wood veneer production industry. Studies on wood veneer defect detection usually focused on detection accuracy for industrial applications but ignored algorithm execution speed; thus, their methods do not meet the required speed of online detection. In this paper, a new detection method is proposed that achieves high accuracy and a suitable speed for online production. Firstly, 2838 wood veneer images were collected using data collection equipment developed in the laboratory and labeled by experienced workers from a wood company. Then, an integrated model, glance multiple channel mask region convolution neural network (R-CNN), was constructed to detect wood veneer defects, which included a glance network and a multiple channel mask R-CNN. Neural network architect search technology was used to automatically construct the glance network with the lowest number of floating-point operations to pick out potential defect images out of numerous original wood veneer images. A genetic algorithm was used to merge the intermediate features extracted by the glance network. Multi-Channel Mask R-CNN was then used to classify and locate the defects. The experimental results show that the proposed method achieves a 98.70% overall classification accuracy and a 95.31% mean average precision, and only 2.5 s was needed to detect a batch of 50 standard images and 50 defective images. Compared with other wood veneer defect detection methods, the proposed method is more accurate and faster.

show abstract

“…On this basis, it is possible to reproduce the cross-sectional image of the tested tank. The main problem of UTT is the difficulty of building a physical model, which would be able to take into account the full complexity of acoustic phenomena occurring in a relatively small space (Bartusek et al , 2008; Du et al ,2018; Gudra and Opieliński, 2006; Herman, 1980; Kak and Slaney, 1999; Kaczmarz, 1937; Nityananda and Samuel, 1992; Polakowski and Sikora, 2016). This causes problems in the implementation of classical tomographic methods.…”

Section: Introductionmentioning

confidence: 99%

Image reconstruction by solving the inverse problem in ultrasonic transmission tomography system

Rymarczyk

Kania²,

Gołąbek³

et al. 2020

COMPEL

View full text Add to dashboard Cite

Purpose The purpose of this study is to develop a reconstruction and measurement system for data analysis using ultrasonic transmission tomography. The problem of reconstruction from the projection is encountered in practical implementation, which consists in reconstructing an image that is an estimation of an unknown object from a finite set of projection data. Reconstructive algorithms used in transmission tomography are based on linear mathematical models, which makes it necessary to process non-linear data into estimates for a finite number of projections. The application of transformation methods requires building a mathematical model in which the projection data forming the known and unknown quantities are functions with arguments from a continuous set of real numbers, determining the function describing the unknown quantities sought in the form of inverse relation and adapting it to operate on discrete and noisy data. This was done by designing a tomographic device and proprietary algorithms capable of reconstructing two-dimensional images regardless of the size, shape, location or number of inclusions hidden in the examined object. Design/methodology/approach The application consists of a device and measuring sensors, as well as proprietary algorithms for image reconstruction. Ultrasonic transmission tomography makes it possible to analyse processes occurring in an object without interfering with the examined object. The proposed solution uses algorithms based on ray integration, the Fermat principle and deterministic methods. Two applications were developed, one based on C and implemented on the embedded device, while the other application was made in Matlab. Findings Research shows that ultrasonic transmission tomography provides an effective analysis of tested objects in closed tanks. Research limitations/implications In the presented technique, the use of ultrasonic absorption wave has been limited. Nevertheless, the effectiveness of such a solution has been confirmed. Practical implications The presented solution can be used for research and monitoring of technological processes. Originality/value Author’s tomographic system consisting of a measuring system and image reconstruction algorithms.

show abstract

Image Reconstruction of Internal Defects in Wood Based on Segmented Propagation Rays of Stress Waves

Cited by 38 publications

References 29 publications

Maintenance of industrial reactors supported by deep learning driven ultrasound tomography

Maintenance of industrial reactors supported by deep learning driven ultrasound tomography

Defect Detection of Industry Wood Veneer Based on NAS and Multi-Channel Mask R-CNN

Image reconstruction by solving the inverse problem in ultrasonic transmission tomography system

Contact Info

Product

Resources

About