Currently, although there are some bamboo slice defect detection schemes based on image processing techniques, these schemes detect fewer types of defects, are less practical, and are difficult to deploy on machines. For this reason, an improved defect detection model for bamboo slice is proposed. Therefore, we propose an improved model for bamboo slice defect detection. The model proposed in this paper is an improved Deformable-DETR model, which firstly replaces the original backbone extraction network ResNet with InternImage, which is stacked with DCNv3 convolution as the core. This network retains the a priori properties of the traditional CNN and captures the long-range dependencies, making the extracted feature spatial semantics richer. Then, after feature extraction, a new sampling module is added, which abstracts the image feature mapping into fine a fine foreground target feature vectors and a small number of coarse background context feature vectors, which can not only remove redundant backgroud feature information but also extract high-semantic foreground.Finally, a novel collaborative hybrid allocation training scheme is introduced, which supervises the training of multiple parallel auxiliary heads through one-to-many label allocation, to easily improve the encoder's learning capability in an end-to-end detector. In addition, data augmentation is used to extend the dataset and migration learning is used to enhance the detection of bamboo slice defects. The experimental results show that the method proposed in this paper improves the defective feature extraction and parsing ability of the model, achieves 85.7% of mAP50 on the test dataset, the inference time for a single image is 0.28 seconds, and the detection accuracy is better than other mainstream target detection models, which provide a new method for detecting defects in bamboo slices.
在进行模型试验时,因为本研究的竹片缺陷检测问题数据集规模较小、缺陷语义信息不丰富,所以使用迁移学习进行模型训练,在预训练权重模型的基础上再进行初步测试。测试过程为:首先选取不同架构的骨干特征提取网络在COCO(Common Dbjects in Context)数据集上进行测试,得到特征热力图;其次根据特征热力图选取性能较好的骨干特征网络;最后进行模型后续模块性能测试的试验。在挑选测试原图时,图片要求有较少物体且物体能有明显个性化特征,在COCO数据集中挑选图7(a)作为测试原图,图7(b)为ResNet50提取的特征热力图,图7(c)为ResNet101提取的特征热力图,图7(d)为SwinTransformer提取的特征热力图,图7(e)为InternImage提取的特征热力图。
ZHANGY X, YAOS B.Inheritance of bamboo culture and development of bamboo industry-based on Yiyang district of Hunan Province,the origin of bamboo culture in China[J].Journal of Beijing Forestry University (Social Science),2011,10(4):7-13.
NIUH, WUX Z, RENG Q,et al.Image detection of bamboo chip defects based on OTSU and CANNY algorithms[J].Forest Engineering,2022,38(6):75-81.
[5]
YANGJ, YANGG.Modified convolutional neural network based on dropout and the stochastic gradient descent optimizer[J].Algorithms,2018,11(3):28.
[6]
BERGMANNP, LÖWES, FAUSERM,et al.Improving unsupervised defect segmentation by applying structural similarity to autoencoders[J].arXiv preprint arXiv:1807. 02011,2018.
[7]
YUL, CHENH, DOUQ,et al.Automated melanoma recognition in dermoscopy images via very deep residual networks[J].IEEE Transactions on Medical Imaging,2016,36(4):994-1004.
[8]
XUEY, LIY.A fast detection method via region-based fully convolutional neural networks for shield tunnel lining defects[J].Computer-Aided Civil and Infrastructure Engineering,2018,33(8):638-654.
[9]
LEIJ, GAOX, FENGZ,et al.Scale insensitive and focus driven mobile screen defect detection in industry[J].Neurocomputing,2018,294:72-81.
[10]
LECUNY, BOTTOUL, BENGIOY,et al.Gradient-based learning applied to document recognition[J].Proceedings of the IEEE,1998,86(11):2278-2324.
[11]
HEY, SONGK, MENGQ,et al.An end-to-end steel surface defect detection approach via fusing multiple hierarchical features[J].IEEE Transactions on Instrumentation and Measurement,2019,69(4):1493-1504.
[12]
TAOX, WANGZ, ZHANGZ,et al.Wire defect recognition of spring-wire socket using multitask convolutional neural networks[J].IEEE Transactions on Components,Packaging and Manufacturing Technology,2018,8(4):689-698.
[13]
HOSANGJ, BENENSONR, SCHIELEB.Learning non-maximum suppression[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.Honolulu,HI,USA,IEEE,2017:4507-4515.
MUH B, WANGS W, QID W,et al.Wood defects recognition based on gray-level co-occurrence matrix and fuzzy BP neural network[J].Forest Engineering,2017,33(4):40-43,54.
[16]
FERGUSONM, RONAYA, LEE TINAY T,et al.Detection and segmentation of manufacturing defects with convolutional neural networks and transfer learning[J].Smart and Sustainable Manufacturing Systems,2018,2(1):137-164.
[17]
HEK, GKIOXARIG, DOLLÁRP,et al.Mask R-CNN[C]//Proceedings of the IEEE International Conference on Computer Vision.Venice,Italy,IEEE,2017:2961-2969.
WANGZ, JIANGY, YANF.Research on wood defect detection model wood-Net based on YOLOv7[J].Journal of Forestry Engineering,2024,9(1):132-140.
[20]
WANGW, DAIJ, CHENZ,et al.Internimage:Exploring large-scale vision foundation models with deformable convolutions[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition.2023:14408-14419.
[21]
WANGT, YUANL, CHENY,et al.Pnp-DETR:Towards efficient visual analysis with transformers[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision.2021:4661-4670.
DAIJ, QIH, XIONGY,et al.Deformable convolutional networks[C]//Proceedings of the IEEE International Conference on Computer Vision.Venice,Italy,IEEE,2017:764-773.
[24]
ZHAOH, SHIJ, QIX,et al.Pyramid scene parsing network[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition,2017:2881-2890.
[25]
ZONGZ, SONGG, LIUY.DETRs with collaborative hybrid assignments training[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision.Paris,France,IEEE,2023:6748-6758.