样本限制场景下金属板带材表面缺陷分类研究

深度学习在金属板带材表面缺陷检测中取得良好的检测效果,但随着网络层数的增加 ,针对较小样 本的金属板带材表面缺陷数据集训练数据容易出现过拟合现象的问题,为此将残差网络与迁 移学习结合提出 了一种融合多层次缺陷特征的图像分类算法。该算法采用残差网络模块逐层提取金属表面缺 陷特征,获得丰 富的位置信息和语义信息缺陷特征的特征图,后续利用分类网络基于该融合特征图得到最终 分类结果,同时 对特征提取网络进行迁移学习,增加网络泛化能力,优化分类精度。利用钢带表面缺陷检测 数据集评估本文 算法性能,实验结果表明,提出的算法具有较好的分类效果,优于其他缺陷分类算法,分类 准确率可达到 99.07%,同时本文所提算法具有良好的抗噪性和泛化性,在金属板带材表面缺陷智能检测中 具有较好的应用价值。

Classification of surface defect of sheet metal strip in sample restricted scene

Deep learning has achieved good detection results in the detection that surface defect detection of Sheet Metal Strip.However,as the number of network layers increases,the train ing data of strip surface defect data set for smaller samples is prone to overfitting.For this reason,combining the residual network and transfer learning,an image classification algorithm combining multi-level defect featur es is proposed.The algorithm uses the residual network module to extract the metal surface defect features la yer by layer,obtains a feature map of rich location information and semantic information defect features,and t hen uses the classification network to obtain the final classification result based on the fusion feature ma p,at the same time,migration learning is performed on the feature extraction network to increase the generali zation ability of the network and optimize the classification accuracy.Using the steel strip surface defect detec tion data set to evaluate the performance of the algorithm in this paper,the experimental results show that t he proposed algorithm has a better classification effect than other defect classification algorithms,and th e classification accuracy rate can reach 99.07%.Simultaneously,the algorithm has good noise resistance and genera lization,and has good application value in the intelligent detection of surface defects of metal plate s and strips.