To solve the problems of the low accuracy and poor real-time performance of traditional strip steel surface defect detection meth-ods,which are caused by the characteristics of many kinds,complex shapes,and different ...To solve the problems of the low accuracy and poor real-time performance of traditional strip steel surface defect detection meth-ods,which are caused by the characteristics of many kinds,complex shapes,and different scales of strip surface defects,a strip steel surface defect detection algorithm based on improved Faster R-CNN is proposed.Firstly,the residual convolution module is inserted into the Swin Transformer network module to form the RC-Swin Transformer network module,and the RC-Swin Transformer module is introduced into the backbone network of the traditional Faster R-CNN to enhance the ability of the network to extract the global feature information of the image and adapt to the complex shape of the strip steel surface defect.To improve the attention of the network to defects in the image,a CBAM-BiFPN network module is designed,and then the backbone network is combined with the CBAM-BiFPN network to realize the de-tection and fusion of multi-scale features.The RoI align layer is used instead of the RoI pooling layer to improve the accuracy of defect loca-tion.Finally,Soft NMS is used to achieve non-maximum suppression and remove redundant boxes.In the comparative experiment on the NEU-DET dataset,the improved algorithm improves the mean average precision by 4.2%compared with the Faster R-CNN algorithm,and also improves the average precision by 6.1%and 6.7%for crazing defect and rolled-in scale defect,which are difficult to detect with the Faster R-CNN algorithm.The experiments show that the improvements proposed in the paper effectively improve the detection accuracy of the algorithm and have certain practical value.展开更多
In order to more accurately detect the accuracy of word-wheel water meter digits, 2000 water meter pictures were produced, and an improved Faster-RCNN algorithm for detecting water meter digits was proposed. The impro...In order to more accurately detect the accuracy of word-wheel water meter digits, 2000 water meter pictures were produced, and an improved Faster-RCNN algorithm for detecting water meter digits was proposed. The improved Faster-RCNN algorithm uses ResNet50 combined with FPN (Feature Pyramid Network) structure instead of the original ResNet50 as the feature extraction network, which can enhance the accuracy of the model for small-sized digit recognition;the use of ROI Align instead of ROI Pooling can eliminate the error caused by the quantization process of the ROI Pooling twice, so that the candidate region is more accurately mapped to the feature map, and the accuracy of the model is further enhanced. The experiment proves that the improved Faster-RCNN algorithm can reach 91.8% recognition accuracy on the test set of homemade dataset, which meets the accuracy requirements of automatic meter reading technology for water meter digital recognition, which is of great significance for solving the problem of automatic meter reading of mechanical water meters and promoting the intelligent development of water meters.展开更多
近年来,异常行为识别算法取得了一定的研究进展,但是针对复杂环境、人体遮挡、动作相似度高等多种挑战,识别算法的适应性、效率、准确性都有待进一步提高。为了解决以上问题,提出了基于特征增强的人体检测与异常行为识别联合算法,首先...近年来,异常行为识别算法取得了一定的研究进展,但是针对复杂环境、人体遮挡、动作相似度高等多种挑战,识别算法的适应性、效率、准确性都有待进一步提高。为了解决以上问题,提出了基于特征增强的人体检测与异常行为识别联合算法,首先将视频序列分别送入人体检测网络和特征加强网络,再采用爱因斯坦求和法将特征加强网络输出的多头卷积注意力特征与人体检测网络输出的热力图特征融合,得到加强融合特征,然后利用检测网络输出的人体目标位置特征信息和ROI Align模块对加强融合特征进行人体ROI(region of interest)区域特征截取,得到人体ROI区域加强融合特征,最后将人体ROI区域加强融合特征送入Transformer时序建模网络模块进行人体行为特征时序建模和识别。所提算法充分利用检测网络中间过程产生的行为主体区域特征,弱化了复杂环境中背景的干扰,同时实现了检测网络的输出特征共享,避免了识别网络的二次特征提取过程,从而提高了网络运行效率,且利用Transformer网络的建模优势,能够充分挖掘人体行为空间特征、时序特征以及之间的跨域特征的优势。实验结果表明:所提算法在提高了网络效率的同时大幅度地提升了网络的识别准确率,达到了预期效果。展开更多
基金supported by the National Natural Science Foundation of China(12002138).
文摘To solve the problems of the low accuracy and poor real-time performance of traditional strip steel surface defect detection meth-ods,which are caused by the characteristics of many kinds,complex shapes,and different scales of strip surface defects,a strip steel surface defect detection algorithm based on improved Faster R-CNN is proposed.Firstly,the residual convolution module is inserted into the Swin Transformer network module to form the RC-Swin Transformer network module,and the RC-Swin Transformer module is introduced into the backbone network of the traditional Faster R-CNN to enhance the ability of the network to extract the global feature information of the image and adapt to the complex shape of the strip steel surface defect.To improve the attention of the network to defects in the image,a CBAM-BiFPN network module is designed,and then the backbone network is combined with the CBAM-BiFPN network to realize the de-tection and fusion of multi-scale features.The RoI align layer is used instead of the RoI pooling layer to improve the accuracy of defect loca-tion.Finally,Soft NMS is used to achieve non-maximum suppression and remove redundant boxes.In the comparative experiment on the NEU-DET dataset,the improved algorithm improves the mean average precision by 4.2%compared with the Faster R-CNN algorithm,and also improves the average precision by 6.1%and 6.7%for crazing defect and rolled-in scale defect,which are difficult to detect with the Faster R-CNN algorithm.The experiments show that the improvements proposed in the paper effectively improve the detection accuracy of the algorithm and have certain practical value.
文摘In order to more accurately detect the accuracy of word-wheel water meter digits, 2000 water meter pictures were produced, and an improved Faster-RCNN algorithm for detecting water meter digits was proposed. The improved Faster-RCNN algorithm uses ResNet50 combined with FPN (Feature Pyramid Network) structure instead of the original ResNet50 as the feature extraction network, which can enhance the accuracy of the model for small-sized digit recognition;the use of ROI Align instead of ROI Pooling can eliminate the error caused by the quantization process of the ROI Pooling twice, so that the candidate region is more accurately mapped to the feature map, and the accuracy of the model is further enhanced. The experiment proves that the improved Faster-RCNN algorithm can reach 91.8% recognition accuracy on the test set of homemade dataset, which meets the accuracy requirements of automatic meter reading technology for water meter digital recognition, which is of great significance for solving the problem of automatic meter reading of mechanical water meters and promoting the intelligent development of water meters.
文摘近年来,异常行为识别算法取得了一定的研究进展,但是针对复杂环境、人体遮挡、动作相似度高等多种挑战,识别算法的适应性、效率、准确性都有待进一步提高。为了解决以上问题,提出了基于特征增强的人体检测与异常行为识别联合算法,首先将视频序列分别送入人体检测网络和特征加强网络,再采用爱因斯坦求和法将特征加强网络输出的多头卷积注意力特征与人体检测网络输出的热力图特征融合,得到加强融合特征,然后利用检测网络输出的人体目标位置特征信息和ROI Align模块对加强融合特征进行人体ROI(region of interest)区域特征截取,得到人体ROI区域加强融合特征,最后将人体ROI区域加强融合特征送入Transformer时序建模网络模块进行人体行为特征时序建模和识别。所提算法充分利用检测网络中间过程产生的行为主体区域特征,弱化了复杂环境中背景的干扰,同时实现了检测网络的输出特征共享,避免了识别网络的二次特征提取过程,从而提高了网络运行效率,且利用Transformer网络的建模优势,能够充分挖掘人体行为空间特征、时序特征以及之间的跨域特征的优势。实验结果表明:所提算法在提高了网络效率的同时大幅度地提升了网络的识别准确率,达到了预期效果。
