Most image-based object detection methods employ horizontal bounding boxes(HBBs)to capture objects in tunnel images.However,these bounding boxes often fail to effectively enclose objects oriented in arbitrary directio...Most image-based object detection methods employ horizontal bounding boxes(HBBs)to capture objects in tunnel images.However,these bounding boxes often fail to effectively enclose objects oriented in arbitrary directions,resulting in reduced accuracy and suboptimal detection performance.Moreover,HBBs cannot provide directional information for rotated objects.This study proposes a rotated detection method for identifying apparent defects in shield tunnels.Specifically,the oriented region-convolutional neural network(oriented R-CNN)is utilized to detect rotated objects in tunnel images.To enhance feature extraction,a novel hybrid backbone combining CNN-based networks with Swin Transformers is proposed.A feature fusion strategy is employed to integrate features extracted from both networks.Additionally,a neck network based on the bidirectional-feature pyramid network(Bi-FPN)is designed to combine multi-scale object features.The bolt hole dataset is curated to evaluate the efficacyof the proposed method.In addition,a dedicated pre-processing approach is developed for large-sized images to accommodate the rotated,dense,and small-scale characteristics of objects in tunnel images.Experimental results demonstrate that the proposed method achieves a more than 4%improvement in mAP_(50-95)compared to other rotated detectors and a 6.6%-12.7%improvement over mainstream horizontal detectors.Furthermore,the proposed method outperforms mainstream methods by 6.5%-14.7%in detecting leakage bolt holes,underscoring its significant engineering applicability.展开更多
Object detection in Remote Sensing(RS)has achieved tremendous advances in recent years,but it remains challenging for rotated object detection due to cluttered backgrounds,dense object arrangements and the wide range ...Object detection in Remote Sensing(RS)has achieved tremendous advances in recent years,but it remains challenging for rotated object detection due to cluttered backgrounds,dense object arrangements and the wide range of size variations among objects.To tackle this problem,Dense Context Feature Pyramid Network(DCFPN)and a powerα-Gaussian loss are designed for rotated object detection in this paper.The proposed DCFPN can extract multi-scale information densely and accurately by leveraging a dense multi-path dilation layer to cover all sizes of objects in remote sensing scenarios.For more accurate detection while avoiding bottlenecks such as boundary discontinuity in rotated bounding box regression,a-Gaussian loss,a unified power generalization of existing Gaussian modeling losses is proposed.Furthermore,the properties ofα-Gaussian loss are analyzed comprehensively for a wider range of applications.Experimental results on four datasets(UCAS-AOD,HRSC2016,DIOR-R,and DOTA)show the effectiveness of the proposed method using different detectors,and are superior to the existing methods in both feature extraction and bounding box regression。展开更多
We propose a method for detecting the symmetry of rotating patterns based on the rotational Doppler effect(RDE)of light.The basic mechanisms of the RDE are introduced,and the spiral harmonic distribution of rotating p...We propose a method for detecting the symmetry of rotating patterns based on the rotational Doppler effect(RDE)of light.The basic mechanisms of the RDE are introduced,and the spiral harmonic distribution of rotating patterns is analyzed.By irradiating the rotating pattern using a superimposed optical vortex and analyzing the amplitude of the RDE signal,the spiral harmonic distribution of the pattern can be measured,and then its symmetry can be detected.We demonstrate this method experimentally by using patterns with different symmetries and shapes.As the method does not need to receive the scattered light completely and accurately,it promises potential application in detecting symmetrical rotating objects at a long distance.展开更多
基金support from the National Natural Science Foundation of China(Grant Nos.52025084 and 52408420)the Beijing Natural Science Foundation(Grant No.8244058).
文摘Most image-based object detection methods employ horizontal bounding boxes(HBBs)to capture objects in tunnel images.However,these bounding boxes often fail to effectively enclose objects oriented in arbitrary directions,resulting in reduced accuracy and suboptimal detection performance.Moreover,HBBs cannot provide directional information for rotated objects.This study proposes a rotated detection method for identifying apparent defects in shield tunnels.Specifically,the oriented region-convolutional neural network(oriented R-CNN)is utilized to detect rotated objects in tunnel images.To enhance feature extraction,a novel hybrid backbone combining CNN-based networks with Swin Transformers is proposed.A feature fusion strategy is employed to integrate features extracted from both networks.Additionally,a neck network based on the bidirectional-feature pyramid network(Bi-FPN)is designed to combine multi-scale object features.The bolt hole dataset is curated to evaluate the efficacyof the proposed method.In addition,a dedicated pre-processing approach is developed for large-sized images to accommodate the rotated,dense,and small-scale characteristics of objects in tunnel images.Experimental results demonstrate that the proposed method achieves a more than 4%improvement in mAP_(50-95)compared to other rotated detectors and a 6.6%-12.7%improvement over mainstream horizontal detectors.Furthermore,the proposed method outperforms mainstream methods by 6.5%-14.7%in detecting leakage bolt holes,underscoring its significant engineering applicability.
文摘Object detection in Remote Sensing(RS)has achieved tremendous advances in recent years,but it remains challenging for rotated object detection due to cluttered backgrounds,dense object arrangements and the wide range of size variations among objects.To tackle this problem,Dense Context Feature Pyramid Network(DCFPN)and a powerα-Gaussian loss are designed for rotated object detection in this paper.The proposed DCFPN can extract multi-scale information densely and accurately by leveraging a dense multi-path dilation layer to cover all sizes of objects in remote sensing scenarios.For more accurate detection while avoiding bottlenecks such as boundary discontinuity in rotated bounding box regression,a-Gaussian loss,a unified power generalization of existing Gaussian modeling losses is proposed.Furthermore,the properties ofα-Gaussian loss are analyzed comprehensively for a wider range of applications.Experimental results on four datasets(UCAS-AOD,HRSC2016,DIOR-R,and DOTA)show the effectiveness of the proposed method using different detectors,and are superior to the existing methods in both feature extraction and bounding box regression。
基金supported by the National Natural Science Foundation of China(Nos.11772001 and 61805283)Key Research Projects of Foundation Strengthening Program(No.2019-JCJQ-ZD)。
文摘We propose a method for detecting the symmetry of rotating patterns based on the rotational Doppler effect(RDE)of light.The basic mechanisms of the RDE are introduced,and the spiral harmonic distribution of rotating patterns is analyzed.By irradiating the rotating pattern using a superimposed optical vortex and analyzing the amplitude of the RDE signal,the spiral harmonic distribution of the pattern can be measured,and then its symmetry can be detected.We demonstrate this method experimentally by using patterns with different symmetries and shapes.As the method does not need to receive the scattered light completely and accurately,it promises potential application in detecting symmetrical rotating objects at a long distance.
