基于卷积神经网络的复杂场景目标检测算法
|
摘要
海上石油平台监控环境复杂,采油工作平台摄像头监控角度不同,海上环境复杂多变,雨雾等天气下,摄像头图片模糊不清.针对上述增加了目标检测的难度的问题,提出了一种基于卷积神经网络的复杂场景目标检测算法(简称ODCS)来检测图像中的特定对象.该方法结合不同分辨率的特征图预测来自然处理各种尺寸的对象,消除了特征重新采样阶段,并将所有计算封装在单个网络中,这样易于训练且可以直接集成到需要检测组件的系统中.实验结果表明,相对于传统的方法,该方法检测在准确率和召回率上明显提高,且检测效率能够满足实时应用的要求.
The monitoring environment of offshore oil platforms is complex, the monitoring angle of the oil production working platform is different, the marine environment is complex and changeable, and the camera pictures are blurred in the weather such as fog and rain. To solve the above problem of increasing the difficulty of object detection, the object detection algorithm based on Convolutional Neural Network(CNN) in complicated scenario(ODCS) is proposed to detect specific objects in the image. This method integrates feature map prediction with different resolutions to naturally process objects of various sizes, eliminates the feature re-sampling phase, and encapsulates all calculations in a single network. This is easy to train and can be integrated directly into the system that needs to detect components. The experimental results show that compared with the traditional methods, the detection accuracy of this method and the recall rate are significantly improved, and the detection efficiency can meet the requirements of real-time applications.
The monitoring environment of offshore oil platforms is complex, the monitoring angle of the oil production working platform is different, the marine environment is complex and changeable, and the camera pictures are blurred in the weather such as fog and rain. To solve the above problem of increasing the difficulty of object detection, the object detection algorithm based on Convolutional Neural Network(CNN) in complicated scenario(ODCS) is proposed to detect specific objects in the image. This method integrates feature map prediction with different resolutions to naturally process objects of various sizes, eliminates the feature re-sampling phase, and encapsulates all calculations in a single network. This is easy to train and can be integrated directly into the system that needs to detect components. The experimental results show that compared with the traditional methods, the detection accuracy of this method and the recall rate are significantly improved, and the detection efficiency can meet the requirements of real-time applications.
引文
1 Girshick R, Donahue J, Darrell T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation. Proceedings of 2014 IEEE Conference on Computer Vision and Pattern Recognition. Columbus, OH,USA. 2014. 580-587.
2 He KM, Zhang XY, Ren SQ, et al. Spatial pyramid poolingin deep convolutional networks for visual recognition.Proceedings of the 13th European Conference on Computer Vision. Zurich, Switzerland. 2014. 346-361.
3 Girshick R. Fast R-CNN. Proceedings of 2015 IEEE International Conference on Computer Vision. Santiago,Chile. 2015. 1440-1448.
4 Ren SQ, He KM, Girshick R, et al. Faster R-CNN:Towards real-time object detection with region proposal networks.Proceedings of the 28th International Conference on Neural Information Processing Systems. Montreal, Canada. 2015.91-99.
5 Redmon J, Divvala S, Girshick R, et al. You only look once:Unified, real-time object detection. Proceedings of 2016IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, NV, USA. 2016. 779-788.
6 Liu W,Anguelov D,Erhan D, et al. SSD:Single shot multibox detector. Proceedings of the 14th European Conference on Computer Vision. Amsterdam, the Netherlands. 2016. 21-37.
7 Felzenszwalb P, Mc Allester D. Ramanan D. A discriminatively trained, multiscale,deformable part model.Proceedings of 2008 IEEE Conference on Computer Vision and Pattern Recognition. Anchorage, AK, USA. 2008. 1-8.
8 Uijlings JRR, van de Sande KEA, Gevers T, et al. Selective search for object recognition. International Journal of Computer Vision,2013,104(2):154-171.[doi:10.1007/s11263-013-0620-5]
9 Erhan D, Szegedy C, Toshev A, et al. Scalable object detection using deep neural networks. Proceedings of 2014IEEE Conference on Computer Vision and Pattern Recognition. Columbus, OH, USA. 2014. 2155-2162.
10 Szegedy C, Reed S, Erhan D, et al. Scalable, high-quality object detection. arXiv:1412.1441,2014.
11 Sermanet P, Eigen D, Zhang X, et al. OverFeat:Integrated recognition, localization and detection using convolutional networks. arXiv:1312.6229, 2013.
12 Shelhamer E, Long J, Darrell T. Fully convolutional networks for semantic segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017,39(4):640-651.[doi:10.1109/TPAMI.2016.2572683]
13 Hariharan B, Arbelaez P, Girshick R, et al. Hypercolumns for object segmentation and fine-grained localization.Proceedings of 2015 IEEE Conference on Computer Vision and Pattern Recognition. Boston, MA, USA. 2015. 447-456.
14 Liu W, Rabinovich A, Berg AC. ParseNet:Looking wider to see better. arXiv:1506.04579, 2016.
15 Zhou BL, Khosla A, Lapedriza A, et al. Object detectors emerge in deep scene CNNs. arXiv:1412.6856, 2014.
2 He KM, Zhang XY, Ren SQ, et al. Spatial pyramid poolingin deep convolutional networks for visual recognition.Proceedings of the 13th European Conference on Computer Vision. Zurich, Switzerland. 2014. 346-361.
3 Girshick R. Fast R-CNN. Proceedings of 2015 IEEE International Conference on Computer Vision. Santiago,Chile. 2015. 1440-1448.
4 Ren SQ, He KM, Girshick R, et al. Faster R-CNN:Towards real-time object detection with region proposal networks.Proceedings of the 28th International Conference on Neural Information Processing Systems. Montreal, Canada. 2015.91-99.
5 Redmon J, Divvala S, Girshick R, et al. You only look once:Unified, real-time object detection. Proceedings of 2016IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, NV, USA. 2016. 779-788.
6 Liu W,Anguelov D,Erhan D, et al. SSD:Single shot multibox detector. Proceedings of the 14th European Conference on Computer Vision. Amsterdam, the Netherlands. 2016. 21-37.
7 Felzenszwalb P, Mc Allester D. Ramanan D. A discriminatively trained, multiscale,deformable part model.Proceedings of 2008 IEEE Conference on Computer Vision and Pattern Recognition. Anchorage, AK, USA. 2008. 1-8.
8 Uijlings JRR, van de Sande KEA, Gevers T, et al. Selective search for object recognition. International Journal of Computer Vision,2013,104(2):154-171.[doi:10.1007/s11263-013-0620-5]
9 Erhan D, Szegedy C, Toshev A, et al. Scalable object detection using deep neural networks. Proceedings of 2014IEEE Conference on Computer Vision and Pattern Recognition. Columbus, OH, USA. 2014. 2155-2162.
10 Szegedy C, Reed S, Erhan D, et al. Scalable, high-quality object detection. arXiv:1412.1441,2014.
11 Sermanet P, Eigen D, Zhang X, et al. OverFeat:Integrated recognition, localization and detection using convolutional networks. arXiv:1312.6229, 2013.
12 Shelhamer E, Long J, Darrell T. Fully convolutional networks for semantic segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017,39(4):640-651.[doi:10.1109/TPAMI.2016.2572683]
13 Hariharan B, Arbelaez P, Girshick R, et al. Hypercolumns for object segmentation and fine-grained localization.Proceedings of 2015 IEEE Conference on Computer Vision and Pattern Recognition. Boston, MA, USA. 2015. 447-456.
14 Liu W, Rabinovich A, Berg AC. ParseNet:Looking wider to see better. arXiv:1506.04579, 2016.
15 Zhou BL, Khosla A, Lapedriza A, et al. Object detectors emerge in deep scene CNNs. arXiv:1412.6856, 2014.