Inicio  /  Applied Sciences  /  Vol: 9 Par: 22 (2019)  /  Artículo
ARTÍCULO
TITULO

Automated Identification of Wood Veneer Surface Defects Using Faster Region-Based Convolutional Neural Network with Data Augmentation and Transfer Learning

Augustas Urbonas    
Vidas Raudonis    
Rytis Maskeliunas and Robertas Dama?evicius    

Resumen

In the lumber and wood processing industry, most visual quality inspections are still done by trained human operators. Visual inspection is a tedious and repetitive task that involves a high likelihood of human error. Currently, new automated solutions with high-resolution cameras and visual inspection algorithms are being tested, but they are not always fast and accurate enough for real-time industrial applications. This paper proposes an automatic visual inspection system for the location and classification of defects on the wood surface. We adopted a faster region-based convolutional neural network (faster R-CNN) for the identification of defects on wood veneer surfaces. Faster R-CNN has been successfully used in medical image processing and object tracking before, but it has not yet been applied for wood panel surface quality assurance. To improve the results, we used pre-trained AlexNet, VGG16, BNInception, and ResNet152 neural network models for transfer learning. The results of the experiments using a synthetically augmented dataset are presented. The best average accuracy of 80.6% was obtained using the pretrained ResNet152 neural network model. By combining all the defect classes, a 96.1% accuracy of finding wood panel surface defects was achieved.

 Artículos similares

       
 
Noor Ul Ain Tahir, Zuping Zhang, Muhammad Asim, Junhong Chen and Mohammed ELAffendi    
Enhancing the environmental perception of autonomous vehicles (AVs) in intelligent transportation systems requires computer vision technology to be effective in detecting objects and obstacles, particularly in adverse weather conditions. Adverse weather ... ver más
Revista: Algorithms

 
Fajia Zheng, Bin Zhang, Yuqiong Zhao, Jiakun Li, Fei Long and Qibo Feng    
Key errors of machine tools have a significant impact on their accuracy, however accurately and quickly measuring the geometric errors of machine tools is essential for key error identification. Fortunately, a quick and direct laser measurement method an... ver más
Revista: Applied Sciences

 
Munir Ahmad, Sagheer Abbas, Areej Fatima, Ghassan F. Issa, Taher M. Ghazal and Muhammad Adnan Khan    
The importance of accurate livestock identification for the success of modern livestock industries cannot be overstated as it is essential for a variety of purposes, including the traceability of animals for food safety, disease control, the prevention o... ver más
Revista: Applied Sciences

 
Jens Dörpinghaus, David Samray and Robert Helmrich    
The German labor market relies heavily on vocational training, retraining, and continuing education. In order to match training seekers with training offers and to make the available data interoperable, we present a novel approach to automatically detect... ver más
Revista: Information

 
Miro Petkovic, Igor Vujovic, Zvonimir Lu?ic and Jo?ko ?oda    
Automated surveillance systems based on machine learning and computer vision constantly evolve to improve shipping and assist port authorities. The data obtained can be used for port and port property surveillance, traffic density analysis, maritime safe... ver más