Detection of Wood Surface Defects Based on Improved YOLOv3 Algorithm

Baogang Wang, Chunmei Yang, Yucheng Ding, Guangyi Qin


For the detection of wood surface defects, a convolutional neural network has a low detection efficiency and insufficient generalization ability, so it does not meet the requirements of online detection. Aiming to solve the above problems, the YOLOv3 baseline model, which has the advantage of multi-objective dynamic detection, was improved and applied to the online detection of wood surface defects. To solve the problem of the poor generalization ability of the network, GridMask was used to enhance the data and improve the robustness of the network. In order to solve the problem of the considerable amount of network parameter calculations and insufficient real-time performance, the residual block of the backbone network was changed to a Ghost block structure to achieve a lightweight model. Finally, the confidence loss function of the network was improved to reduce the influence of simple samples and negative samples on model convergence. The experimental results showed that, compared with the original network, the improved algorithm increased the mean average precision by 5.73% and the detection speed was increased to 28 frames per second (an increase of 11), which met the requirements for real-time industrial detection.


Deep learning; Wood defects detection; You Only Look Once; Convolution neural network; Data enhancement

Full Text:


Welcome to BioResources! This online, peer-reviewed journal is devoted to the science and engineering of biomaterials and chemicals from lignocellulosic sources for new end uses and new capabilities. The editors of BioResources would be very happy to assist you during the process of submitting or reviewing articles. Please note that logging in is required in order to submit or review articles. Martin A. Hubbe, (919) 513-3022,; Lucian A. Lucia, (919) 515-7707, URLs:; ISSN: 1930-2126