改进CaffeNet模型在水面垃圾识别中的应用
|
摘要
为了提高水面垃圾识别的准确率,提出一种改进CaffeNet的卷积神经网络模型对水面垃圾进行识别。模型改进了卷积核的大小、卷积核的数量以及增加了一层稀疏结构,进而增强了网络模型特征提取的能力,降低了网络复杂度。实验结果证明:改进的CaffeNet模型将水面垃圾的识别率提高到95. 75%,能减少水面波纹、物体倒影和桥梁等复杂环境对水面垃圾识别的影响,具有较好的水面垃圾识别效果。
In order to improve the accuracy of water surface garbage identification,a modified convolutional neural network model of Caffe Net is proposed to identify water surface garbage. The model improves the size of the convolution kernel,the number of convolution kernels,and adds a sparse structure,which enhances the ability of network model feature extraction and reduces network complexity. The experimental results show that the improved Caffe Net model improves the recognition rate of water surface garbage to 95.75 %,which can reduce the influence of water surface ripple,object reflection and bridges on the recognition of water surface garbage,and has better water surface garbage recognition effect.
In order to improve the accuracy of water surface garbage identification,a modified convolutional neural network model of Caffe Net is proposed to identify water surface garbage. The model improves the size of the convolution kernel,the number of convolution kernels,and adds a sparse structure,which enhances the ability of network model feature extraction and reduces network complexity. The experimental results show that the improved Caffe Net model improves the recognition rate of water surface garbage to 95.75 %,which can reduce the influence of water surface ripple,object reflection and bridges on the recognition of water surface garbage,and has better water surface garbage recognition effect.
引文
[1]王斌.海面目标红外检测方法研究[D].大连:大连海事大学,2018.
[2]叶晓杰,赵巨峰,公晓丽,等.基于边缘权重分析的水面弱目标红外增强[J].激光与红外,2018(1):119-123.
[3]李昀峰,周东,阮雅端,等.抑制船尾拖纹的船舶显著性视频检测方法[J].北京邮电大学学报,2017,40(S1):72-76.
[4]李双双.内河视频监控中运动船舶检测算法研究[D].大连:大连海事大学,2018.
[5]薛萍.基于超像素特征表示的图像前景背景分割算法[J].西安科技大学学报,2017(5):731-735.
[6]方晶,冯顺山,冯源.水面航行体对舰船目标的图像检测方法[J].北京理工大学学报,2017(12):1235-1240.
[7]魏建荣.背景纹理模型在海面舰船目标检测中的应用研究[J].舰船科学技术,2017(20):159-161.
[8] AREL I,ROSE D C,KARNOWSKI T P. Deep machine learning—A new frontier in artificial intelligence research[Research Frontier][J]. Computational Intelligence Magazine IEEE,2010,5(4):13-18.
[9] DAHL G E,YU D,DENG L,et al. Context-dependent pre-trained deep neural networks for large-vocabulary speech recognition[J].IEEE Transactions on Audio Speech&Language Processing,2012,20(1):30-42.
[10]余乐,李阳光,陈岩,等.基于卷积神经网络的软硬触觉感知方法研究[J].传感器与微系统,2017,36(6):35-37.
[11]尹宝才,王文通,王立春.深度学习研究综述[J].北京工业大学学报,2015(1):48-59.
[12]纪野,李玉惠,王蒙,等.基于卷积神经网络的车型识别方法研究[J].传感器与微系统,2017,36(11):42-43.
[13]秦子亮,李朝锋.基于卷积神经网络的胸片肺野自动分割[J].传感器与微系统,2017,36(10):64-66.
[14] KRIZHEVSKY A,SUTSKEVER I,HINTON G E. Image Net classification with deep convolutional neural networks[C]∥International Conference on Neural Information Processing Systems,Curran Associates Inc,2012:1097-1105.
[15] HINTON G E,SRIAVSTAVA N,KRIZHEVSKY A,et al. Improving neural networks by preventing co-adaptation of feature detectors[J].Computer Science,2012,3(4):212-223.
[16] LIU F,SHEN C. Learning deep convolutional features for MRIbased Alzheimer’s disease classification[EB/OL].[2017—04—30]. https:∥www. researchgate. net/publication/261636125_Learning_Deep_Convolutional_Features_for_MRI_Based_Alzheimer's_Disease_Classification.
[2]叶晓杰,赵巨峰,公晓丽,等.基于边缘权重分析的水面弱目标红外增强[J].激光与红外,2018(1):119-123.
[3]李昀峰,周东,阮雅端,等.抑制船尾拖纹的船舶显著性视频检测方法[J].北京邮电大学学报,2017,40(S1):72-76.
[4]李双双.内河视频监控中运动船舶检测算法研究[D].大连:大连海事大学,2018.
[5]薛萍.基于超像素特征表示的图像前景背景分割算法[J].西安科技大学学报,2017(5):731-735.
[6]方晶,冯顺山,冯源.水面航行体对舰船目标的图像检测方法[J].北京理工大学学报,2017(12):1235-1240.
[7]魏建荣.背景纹理模型在海面舰船目标检测中的应用研究[J].舰船科学技术,2017(20):159-161.
[8] AREL I,ROSE D C,KARNOWSKI T P. Deep machine learning—A new frontier in artificial intelligence research[Research Frontier][J]. Computational Intelligence Magazine IEEE,2010,5(4):13-18.
[9] DAHL G E,YU D,DENG L,et al. Context-dependent pre-trained deep neural networks for large-vocabulary speech recognition[J].IEEE Transactions on Audio Speech&Language Processing,2012,20(1):30-42.
[10]余乐,李阳光,陈岩,等.基于卷积神经网络的软硬触觉感知方法研究[J].传感器与微系统,2017,36(6):35-37.
[11]尹宝才,王文通,王立春.深度学习研究综述[J].北京工业大学学报,2015(1):48-59.
[12]纪野,李玉惠,王蒙,等.基于卷积神经网络的车型识别方法研究[J].传感器与微系统,2017,36(11):42-43.
[13]秦子亮,李朝锋.基于卷积神经网络的胸片肺野自动分割[J].传感器与微系统,2017,36(10):64-66.
[14] KRIZHEVSKY A,SUTSKEVER I,HINTON G E. Image Net classification with deep convolutional neural networks[C]∥International Conference on Neural Information Processing Systems,Curran Associates Inc,2012:1097-1105.
[15] HINTON G E,SRIAVSTAVA N,KRIZHEVSKY A,et al. Improving neural networks by preventing co-adaptation of feature detectors[J].Computer Science,2012,3(4):212-223.
[16] LIU F,SHEN C. Learning deep convolutional features for MRIbased Alzheimer’s disease classification[EB/OL].[2017—04—30]. https:∥www. researchgate. net/publication/261636125_Learning_Deep_Convolutional_Features_for_MRI_Based_Alzheimer's_Disease_Classification.