Aiming at the problem of insufficient image segmentation accuracy caused by fuzzy edge and texture interference of surface defects in automatic production of wood processing, an improved ivy algorithm (IIVY) is proposed for multi-threshold segmentation of wood defect images. Firstly, the population diversity is enhanced by the wind direction growth mechanism, and a nonlinear dynamic balance factor is designed to dynamically coordinate the global exploration and local development capabilities. Secondly, the elite-oriented regeneration strategy is introduced to improve the ability of the algorithm to jump out of the local optimum. Then, the fitness function is designed based on symmetric cross entropy, and the wood defect image is segmented by IIVY. Compared with four classical algorithms (grey wolf optimizer (GWO), whale optimization algorithm (WOA), sparrow search algorithm (SSA) and sand cat swarm optimization (SCSO), the evaluation indexes include optimal fitness value, peak signal-to-noise ratio, feature similarity and subjective visual evaluation. The results show that the IIVY fitness convergence curve is significantly better than the comparison algorithm; in terms of peak signal-to-noise ratio and feature similarity index, the number of test groups in which IIVY achieved the optimal value accounted for 83.33% and 91.67% of the total number of test groups, respectively. IIVY is more accurate in the segmentation of the edge of the defect area, and the segmentation results completely retain the wood texture details. The IIVY algorithm can accurately segment the wood surface defects, retain the texture features of the wood surface, and provide reliable technical support for wood defect detection.
然而随着迭代次数的增加,多阈值分割方法的计算复杂度呈指数级增长。为提升计算效率和分割精度,近年来许多学者将群智能优化算法通过与多阈值分割方法结合。Ma等[16]提出了一种基于鲸鱼优化算法的改进多阈值图像分割方法(Adaptive weighting strategy,and vertical crossover strategy with whale optimization algorithm,RAV-WOA),以大津法为目标函数,在对灰度图像和彩色图像进行分割时,该算法能够在保证高效率和高质量的前提下选择满意的最优阈值。Bei等[17]采用基于领导和自吞噬机制的黏菌算法(Slime mould algorithm with mechanism of leadership and self-phagocytosis,SMA-MLS)确定图像的最优阈值,其次,提出了带领导机制的位置更新公式,提高了收敛速度和精度,并以Kapur为目标函数,对彩色图像进行了多阈值图像分割试验,证明了SMA-MLS算法在图像MT(Multilevel thresholding of color image)分割中的有效性。Wang等[18]提出了一种改进金枪鱼优化算法(Improved tuna swarm optimization,ITSO),并与对称交叉熵结合应用于森林冠层图像分割,在分割质量、一致性和准确性方面取得了优异的分割效果。李圣涵等[19]使用蛇优化算法(Snake optimization,SO)对OTSU方法进行改进,提出了一种SO-OTSU方法,试验结果表明,该方法拥有更快的计算速度,分割的精确度也更高,是一种有效的图像分割方法。陈光伟等[20]提出了一种基于多策略融合未来搜索算法(IFSA)的多阈值林火图像分割方法,分割精度和效率显著优于传统优化算法,为森林火灾监测系统提供了可靠的技术支撑。
近年,诸多新兴智能优化算法展现出优异的寻优性能。例如,Ghasemi等[21]在2024年提出了一种模仿自然界常春藤植物行为的智能启发式算法,即常春藤优化算法(ivy algorithm,IVY),IVY算法的简单和灵活,使其易于修改和扩展。王合彬等[22]将IVY与灰狼优化算法(grey wolf optimizer,GWO)的深度融合改进优化天文干涉阵列;Zhang等[23]引入自适应扰动因子等策略,进一步证明了IVY在工程优化领域的潜力。然而,该算法仍存在收敛精度不足、易陷入局部最优等固有缺陷。为此,本研究提出一种改进的常春藤优化算法(improved ivy algorithm,IIVY),将其应用到木材缺陷图像分割领域。其核心思路是利用IIVY对对称交叉熵函数搜索寻优,以获取最优阈值组合,最终提升木材缺陷图像分割的精度。
YANGY T, WANGH H, JIANGD,et al.Surface detection of solid wood defects based on SSD improved with ResNet[J].Forests,2021,12(10):1419.
[2]
EHTISHAMR, QAYYUMW, CAMPC V,et al.Computing the characteristics of defects in wooden structures using image processing and CNN[J].Automation in Construction,2024,158:105211.
DINGA N, HEC G, DUOH Q,et al.Research review of wood defect recognition based on digital images[J].Chinese Journal of Wood Science and Technology,2022,36(1):9-16,28.
[5]
VIDHYAK, REVATHIS, A-SHWINIS S,et al.Review on digital image segmentation techniques[J].International Research Journal of Engineering and Technology,2016,3(2):618-619.
FUX, ZHUL K, HUANGJ P,et al.Multi-threshold image segmentation based on improved northern goshawk optimization algorithm[J].Computer Engineering,2023,49(7):232-241.
[10]
RAMESHK K D, KUMARG K, SWAPNAK,et al.A review of medical image segmentation algorithms[J].EAI Endorsed Transactions on Pervasive Health and Technology,2021,7(27):1-9.
[11]
LUOW, SUNL P.An improved binarization algorithm of wood image defect segmentation based on non-uniform background[J].Journal of Forestry Research,2019,30:1527-1533.
BAIX B, XUJ T, GUOJ Q,et al.Segmentation of wood surface defect image based on improved C-V model[J].Journal of Beijing Forestry University,2015,37(12):108-115.
[16]
MINAEES, BOYKOVY, PORIKLIF,et al.Image segmentation using deep learning:A survey[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2022,44(7):3523-3542.
CHENGY, LIB J.Image segmentation technology and its application in digital image processing[C]//2021 IEEE Asia-Pacific Conference on Image Processing,Electronics and Computers (IPEC).April 14-16,2021.Dalian,China.IEEE,2021:1174-1177.
[19]
HANB, WUY Q.A novel active contour model based on modified symmetric cross entropy for remote sensing river image segmentation[J].Pattern Recognition,2017,67:396-409.
[20]
GONGQ Y, ZHAOX, BIC Y,et al.Maximum entropy multi-threshold image segmentation based on improved particle swarm optimization[J].Journal of Physics:Conference Series,2020,1678(1):012098.
[21]
MAG Y, YUEX F.An improved whale optimization algorithm based on multilevel threshold image segmentation using the Otsu method[J].Engineering Applications of Artificial Intelligence,2022,113:104960.
[22]
BEIJ L, WANGJ Q, SONGH H,et al.Slime mould algorithm with mechanism of leadership and self-phagocytosis for multilevel thresholding of color image[J].Applied Soft Computing,2024,163:111836.
[23]
WANGJ Y, ZHUL K, WUB W,et al.Forestry canopy image segmentation based on improved tuna swarm optimization[J].Forests,2022,13(11):1746.
CHENG W, XUL, FANGL,et al.Forest fire image segmentation based on multi-strategy fusion future search algorithm[J].Forest Engineering,2023,39(4):134-144.
[28]
GHASEMIM, ZAREM, TROJOVSKÝP,et al.Optimization based on the smart behavior of plants with its engineering applications:Ivy algorithm[J].Knowledge-Based Systems,2024,295:111850.
WANGH B, WANGG H, YANGD H,et al.Application of improved ivy optimization algorithm in the design of astronomical interference arrays[J].Application of IC,2024,41(11):78-79.
[31]
ZHANGC Q, LINW Z, HUG.An enhanced ivy algorithm fusing multiple strategies for global optimization problems[J].Advances in Engineering Software,2025,203:103862.
General Administration of Quality Supervision,Inspection and Quarantine of the People′s Republic of China,Standardization Administration of the People′s Republic of China. Defects in logs:GB/T 155—2017 [S].Beijing:China Standards Press,2017.
[37]
REKAR, KARTHICKR, SARAVANARAM R,et al.Multi head self-attention gated graph convolutional network based multi-attack intrusion detection in MANET[J].Computers & Security,2024,136:103526.
[38]
YANZ P, ZHANGJ Z, YANGZ W,et al.Kapur′s entropy for underwater multilevel thresholding image segmentation based on whale optimization algorithm[J].IEEE Access,2020,9:41294-41319.
[39]
MIRJALILIS, MIRJALILIS M, LEWISA.Grey wolf optimizer[J].Advances in Engineering Software,2014,69:46-61.
[40]
MIRJALILIS, LEWISA.The whale optimization algorithm[J].Advances in Engineering Software,2016,95:51-67.
SEYYEDABBASIA, KIANIF.Sand Cat swarm optimization:A nature-inspired algorithm to solve global optimization problems[J].Engineering with Computers,2023,39(4):2627-2651.