浏览全部资源
扫码关注微信
纸质出版日期： 2022-04-16 ，
录用日期： 2021-01-05
移动端阅览
贾可心, 马正华, 朱蓉, 李永刚. 注意力机制改进轻量SSD模型的海面小目标检测[J]. 中国图象图形学报, 2022,27(4):1161-1175.
Kexin Jia, Zhenghua Ma, Rong Zhu, Yonggang Li. Attention-mechanism-based light single shot multiBox detector modelling improvement for small object detection on the sea surface[J]. Journal of Image and Graphics, 2022,27(4):1161-1175.
目的
2
海面目标检测图像中的小目标数量居多，而基于深度学习的目标检测方法通常针对通用目标数据集设计检测模型，对图像中的小目标检测效果并不理想。使用一般目标检测模型检测海面目标图像的特征时，通常会出现小目标漏检情况，而一些特定的小目标检测模型对海面目标的检测效果还有待验证。为此，在标准的SSD(single shot multiBox detector)目标检测模型基础上，结合Xception深度可分卷积，提出一种轻量SSD模型用于海面目标检测。
方法
2
在标准的SSD目标检测模型基础上，使用基于Xception网络的深度可分卷积特征提取网络替换VGG-16(Visual Geometry Group network-16)骨干网络，通过控制变量来对比不同网络的检测效果；在特征提取网络中的exit flow层和Conv1层引入轻量级注意力机制模块来提高检测精度，并与在其他层引入轻量级注意力机制模块的模型进行检测效果对比；使用注意力机制改进的轻量SSD目标检测模型和其他几种模型分别对海面目标检测数据集中的小目标和正常目标进行测试。
结果
2
为证明本文模型的有效性，进行了多组对比实验。实验结果表明，模型轻量化导致特征表达能力降低，从而影响检测精度。相对于标准的SSD目标检测模型，本文模型在参数量降低16.26%、浮点运算量降低15.65%的情况下，浮标的平均检测精度提高了1.1%，漏检率减小了3%，平均精度均值(mean average precision，mAP)提高了0.51%，同时，保证了船的平均检测精度，并保证其漏检率不升高，在对数据集中的小目标进行测试时，本文模型也表现出较好的检测效果。
结论
2
本文提出的海面小目标检测模型，能够在压缩模型的同时，保证模型的检测速度和检测精度，达到网络轻量化的效果，并且降低了小目标的漏检率，可以有效实现对海面小目标的检测。
Objective
2
Object detection on the sea surface plays a key role in the development and utilization of marine resources. The sea environment is complex and changeable
and there are many kinds of objects. Considering the factors such as safety and obstacle avoidance
the shooting process of sea surface object detection images will target on the amount of small and medium-sized objects in the image majority
which puts forward higher requirements for accurate detection of objects on the sea surface. Although some regular object detection methods with good detection results have been proposed
they still face the problems of low detection accuracy and slow detection speed. With the rapid development of deep learning theory
the feature extraction capability of deep learning model is gradually mature
and it is widely used in object detection technology. Compared with the original object detection methods
deep-learning-based object detection method has its priority in speed and accuracy. Deep-learning-based object detection method focuses on the construction of deeper network to improve the detection accuracy. The network model usually has the difficulties with too large parameters
which leads to the slow detection speed. Most of the good detection network can only run on high-performance graphics processor unit (GPU)
which requires higher computing power equipment. It will also interfere the detection accuracy of the network if the model is compressed. In addition
the initial deep-learning-based object detection method is a detection model designed for the general object dataset. For the small object in the image
the detection effect is not very ideal. In terms of the characteristics of the sea object detection image
the general object detection model will miss the detection of small objects
and the detection effect of some small-targeted object detection models for sea objects needs to be verified.
Method
2
The original data of this demonstration is based on the marine obstacle detection dataset 2 (MODD 2)
which is mainly composed of boats
buoys and other sea objects. Total 5 050 images of them are used in the illustrated data. To construct the sea surface object dataset
the boats and buoys are calibrated by calibration software called LabelImg
and processed in accordance with the format of visual object class 2007 (VOC2007) dataset. First
on the basis of standard single shot MultiBox detector (SSD) object detection model
Visual Geometry Group network-16 (VGG-16) backbone network is substituted via depth wise separable convolution feature extraction network based on Xception network. The detection effect of different network models is compared based on variables application
including VGG-16-based SSD network
Mobilenet-based SSD network and Xception-based SSD network. In the process of training
the size of the input image is scaled to the RGB image of 300 × 300 pixels. The following input images are normalized. The trained model is based on the Xception pre-trained model on common objects in context(COCO) dataset. Next
the SSD + Xception object detection model is used as lightweight SSD model based on Xception feature extraction network. The lightweight attention mechanism module is evolved into exit flow layer and Conv1 layer in feature extraction network to improve the detection accuracy
and the detection effect is compared with the model of lightweight attention mechanism module in other layers. The model parameters (params)
floating-point operations per second (FLOPs) and the quantity of images can be processed via frames per second (FPS). Precision rate and miss rate are used to evaluate the model. The mean average precision (mAP) are used to evaluate the performance of the model. At last
the small object and normal object in the sea object detection dataset are tested via the lightweight SSD object detection model with improved attention mechanism and other models.
Result
2
In order to prove the effectiveness of this model
a quantity of comparative experiments are conducted. Firstly
the parameters and floating-point operation of each model are compared
and the reason of network lightweight is analyzed. The demonstrated illustration analyzed that the model can improve the memory reading and writing speed to achieve the network lightweight effect via Mul deduction and operations adding. But
the compression model will lead to the reduction of network feature expression ability
thus affecting the detection accuracy to a certain extent. The SSD object detection model with Mobilenet as feature extraction network is the lightest
but its detection accuracy is interfered at most
and the mAP is reduced by 2.28%. The SSD + Xception object detection model is opted as the lightweight SSD sea object detection model based on Xception feature extraction network. The model transforms a certain amount of Mul and Add operations only
which reduces the parameter amount by 19.01% and the floating-point operation amount by 18.40%. It garantees the feature expression ability of the model
and maintains the amount of images processed per second based on cutting parameters and floating-point operations. The quantity is basically unchanged
and the detection accuracy is reduced less
which achieves the effect of lightening the network under the condition of a certain detection accuracy. In order to improve the detection accuracy of the lightweight SSD sea object detection model
a lightweight attention mechanism module is issued in the lower layer of the model to focus on some significant or interesting information
which facilitates the illustration of the feature semantic information of small objects. In comparison of the standard SSD target detection model
the analyses demonstrate that the average accuracy of the buoy is increased by 1.1%
the miss detection rate is reduced by 3%
and the mAP is increased by 0.51%
the parameters are reduced by 16.26% and the floating-point operation is reduced by 15.65%. Simultaneously
the average detection accuracy of the boat is guaranteed
and the miss detection error is not generated more. The illustrated model also shows qualified detection effect based on small objects in the dataset verification.
Conclusion
2
For the small object detection in the sea image
this small object detection model can identify the detection speed of the model and guarantee the detection accuracy of the model
and achieve the effect of network lightweight. Moreover
this model reduces the rate of missing detection of small objects to realize the detection of small sea objects effectively as well.
深度学习目标检测注意力机制深度可分卷积SSD海面小目标检测
deep learningobject detectionattention mechanismdepthwise separable convolutionsingle shot multiBox detector (SSD)small object detection on the sea surface
Bell S, Zitnick C L, Bala K and Girshick R. 2016. Inside-outside net: detecting objects in context with skip pooling and recurrent neural networks//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas, USA: IEEE: 2874-2883 [DOI: 10.1109/CVPR.2016.314http://dx.doi.org/10.1109/CVPR.2016.314]
Bovcon B, Mandeljc R, Perš J and Kristan M. 2018. Stereo obstacle detection for unmanned surface vehicles by IMU-assisted semantic segmentation. Robotics and Autonomous Systems, 104: 1-13 [DOI: 10.1016/j.robot.2018.02.017]
Cao G M, Xie X M, Yang W Z, Liao Q, Shi G M and Wu J J. 2018. Feature-fused SSD: fast detection for small objects//Proceedings of SPIE 10615 9th International Conference on Graphic and Image Processing (ICGIP 2017). Qingdao, China: SPIE: 106151E [DOI: 10.1117/12.2304811http://dx.doi.org/10.1117/12.2304811]
Chen C Y, Liu M Y, Tuzel O and Xiao J X. 2016. R-CNN for small object detection//Proceedings of the 13th Asian Conference on Computer Vision. Taipei, China: Springer: 214-230 [DOI: 10.1007/978-3-319-54193-8_14http://dx.doi.org/10.1007/978-3-319-54193-8_14]
Cheng P, Liu W, Zhang Y F and Ma H D. 2018. Loco: local context based faster R-CNN for small traffic sign detection//Proceedings of the 24th International Conference on Multimedia Modeling. Bangkok, Thailand: Springer: 329-341 [DOI: 10.1007/978-3-319-73603-7_27http://dx.doi.org/10.1007/978-3-319-73603-7_27]
Chollet F. 2017. Xception: deep learning with depthwise separable convolutions//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu, USA: IEEE: 1800-1807 [DOI: 10.1109/CVPR.2017.195http://dx.doi.org/10.1109/CVPR.2017.195]
Hu J, Shen L and Sun G. 2018. Squeeze-and-excitation networks//Proceedings of 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City, USA: IEEE: 7132-7141 [DOI: 10.1109/CVPR.2018.00745http://dx.doi.org/10.1109/CVPR.2018.00745]
Hu P Y and Ramanan D. 2017. Finding tiny faces//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu, USA: IEEE: 1522-1530 [DOI: 10.1109/CVPR.2017.166http://dx.doi.org/10.1109/CVPR.2017.166]
Kisantal M, Wojna Z, Murawski J, Naruniec J and Cho K. 2019. Augmentation for small object detection [EB/OL]. [2021-08-20]https://arxiv.org/pdf/1902.07296.pdfhttps://arxiv.org/pdf/1902.07296.pdf
Li J N, Liang X D, Wei Y C, Xu T F, Feng J S and Yan S C. 2017. Perceptual generative adversarial networks for small object detection//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu, USA: IEEE: 1951-1959 [DOI: 10.1109/CVPR.2017.211http://dx.doi.org/10.1109/CVPR.2017.211]
Lin T Y, Dollár P, Girshick R, He K M, Hariharan B and Belongie S. 2017. Feature pyramid networks for object detection//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu, USA: IEEE: 936-944 [DOI: 10.1109/CVPR.2017.106http://dx.doi.org/10.1109/CVPR.2017.106]
Liu W, Anguelov D, Erhan D, Szegedy C, Reed S, Fu C Y and Berg A C. 2016. SSD: single shot MultiBox detector//Proceedings of the 14th European Conference on Computer Vision. Amsterdam, the Netherlands: Springer: 21-37 [DOI: 10.1007/978-3-319-46448-0_2http://dx.doi.org/10.1007/978-3-319-46448-0_2]
Menikdiwela M, Nguyen C, Li H D and Shaw M. 2017. CNN-based small object detection and visualization with feature activation mapping//Proceedings of 2017 International Conference on Image and Vision Computing New Zealand (IVCNZ). Christchurch, New Iealand: IEEE: 1-5 [DOI: 10.1109/IVCNZ.2017.8402455http://dx.doi.org/10.1109/IVCNZ.2017.8402455]
Redmon J, Divvala S, Girshick R and Farhadi A. 2016. You only look once: unified, real-time object detection//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas, USA: IEEE: 779-788 [DOI: 10.1109/CVPR.2016.91http://dx.doi.org/10.1109/CVPR.2016.91]
Ren S Q, He K M, Girshick R and Sun J. 2017. Faster R-CNN: towards real-time object detection with region proposal networks. IEEE Transactions on Pattern Analysis and Machine Intelligence, 39(6): 1137-1149 [DOI: 10.1109/TPAMI.2016.2577031]
Ren Y, Zhu C R and Xiao S P. 2018. Small object detection in optical remote sensing images via modified Faster R-CNN. Applied Sciences, 8(5): #813 [DOI: 10.3390/app8050813]
Singh B, Najibi M and Davis L S. 2018. SNIPER: efficient multi-scale training//Proceedings of the 32nd International Conference on Neural Information Processing Systems. New York, USA: Curran Associates Inc. : 9333-9343
Wang F, Jiang M Q, Qian C, Yang S, Li C, Zhang H G, Wang X G and Tang X O. 2017. Residual attention network for image classification//Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu, USA: IEEE: 6450-6458 [DOI: 10.1109/CVPR.2017.683http://dx.doi.org/10.1109/CVPR.2017.683]
Yang F, Choi W and Lin Y Q. 2016. Exploit all the layers: fast and accurate CNN object detector with scale dependent pooling and cascaded rejection classifiers//Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas, USA: IEEE: 2129-2137 [DOI: 10.1109/CVPR.2016.234http://dx.doi.org/10.1109/CVPR.2016.234]
Zhao R, Hynes S and He G S. 2014. Defining and quantifying China's ocean economy. Marine Policy, 43: 164-173 [DOI: 10.1016/j.marpol.2013.05.008]
相关作者
相关机构
京公网安备11010802024621