A Marine Object Detection Algorithm Based on SSD and Feature Enhancement

Autonomous detection and fishing by underwater robots will be the main way to obtain aquatic products in the future; sea urchins are the main research object of aquatic product detection. When the classical Single-Shot MultiBox Detector (SSD) algorithm is applied to the detection of sea urchins, it also has disadvantages of being inaccurate to small targets and insensitive to the direction of the sea urchin. Based on the classic SSD algorithm, this paper proposes a feature-enhanced sea urchin detection algorithm. Firstly, according to the spiny-edge characteristics of a sea urchin, a multidirectional edge detection algorithm is proposed to enhance the feature, which is taken as the 4th channel of image and the original 3 channels of underwater image together as the input for the further deep learning. Then, in order to improve the shortcomings of SSD algorithm’s poor ability to detect small targets, resnet 50 is used as the basic framework of the network, and the idea of feature cross-level fusion is adopted to improve the feature expression ability and strengthen semantic information. The open data set provided by the National Natural Science Foundation of China underwater Robot Competition will be used as the test set and training set. Under the same training and test conditions, the AP value of the algorithm in this paper reaches 81.0%, 7.6% higher than the classic SSD algorithm, and the confidence of small target analysis is also improved. Experimental results show that the algorithm in this paper can effectively improve the accuracy of sea urchin detection.

[1]  Bingliang Hu,et al.  Method for enhancing visibility of hazy images based on polarimetric imaging , 2014 .

[2]  Codruta O. Ancuti,et al.  Color Balance and Fusion for Underwater Image Enhancement , 2018, IEEE Transactions on Image Processing.

[3]  Xu Zhang,et al.  Dilated residual attention network for load disaggregation , 2019, Neural Computing and Applications.

[4]  Liguo Weng,et al.  Cloud/snow recognition for multispectral satellite imagery based on a multidimensional deep residual network , 2018, International Journal of Remote Sensing.

[5]  Ross B. Girshick,et al.  Fast R-CNN , 2015, 1504.08083.

[6]  Ying-Ching Chen,et al.  Underwater Image Enhancement by Wavelength Compensation and Dehazing , 2012, IEEE Transactions on Image Processing.

[7]  Shifeng Zhang,et al.  Single-Shot Refinement Neural Network for Object Detection , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[8]  Trevor Darrell,et al.  Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation , 2013, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[9]  Mehdi Kaytoue-Uberall,et al.  FSSD - A Fast and Efficient Algorithm for Subgroup Set Discovery , 2019, 2019 IEEE International Conference on Data Science and Advanced Analytics (DSAA).

[10]  Emanuele Trucco,et al.  Detecting man-made objects in unconstrained subsea videos , 2002, BMVC.

[11]  Yu Zhang,et al.  Reduced Complexity Channel Models for IMT-Advanced Evaluation , 2009, EURASIP J. Wirel. Commun. Netw..

[12]  Kaiming He,et al.  Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[13]  Xun Wang,et al.  Texture filtering based physically plausible image dehazing , 2016, The Visual Computer.

[14]  Zhizhong Li,et al.  Task complexity: A review and conceptualization framework , 2012 .

[15]  Jian Sun,et al.  Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition , 2015, IEEE Trans. Pattern Anal. Mach. Intell..

[16]  Bai Ying Lei,et al.  Accurate Segmentation of Cervical Cytoplasm and Nuclei Based on Multiscale Convolutional Network and Graph Partitioning , 2015, IEEE Transactions on Biomedical Engineering.

[17]  Min Xia,et al.  Parallel knowledge acquisition algorithms for big data using MapReduce , 2018, Int. J. Mach. Learn. Cybern..

[18]  Sanghyun Park,et al.  Take me to SSD: a hybrid block-selection method on HDFS based on storage type , 2016, SAC.

[19]  Adrian Galdran,et al.  Automatic Red-Channel underwater image restoration , 2015, J. Vis. Commun. Image Represent..

[20]  Jia Liu,et al.  Portfolio trading system of digital currencies: A deep reinforcement learning with multidimensional attention gating mechanism , 2020, Neurocomputing.

[21]  David G. Lowe,et al.  Distinctive Image Features from Scale-Invariant Keypoints , 2004, International Journal of Computer Vision.

[22]  Min Xia,et al.  Water Areas Segmentation from Remote Sensing Images Using a Separable Residual SegNet Network , 2020, ISPRS Int. J. Geo Inf..

[23]  Ying Chen,et al.  M2Det: A Single-Shot Object Detector based on Multi-Level Feature Pyramid Network , 2018, AAAI.

[24]  Min Xia,et al.  Weighted Densely Connected Convolutional Networks for Reinforcement Learning , 2020, Int. J. Pattern Recognit. Artif. Intell..

[25]  Tiegen Liu,et al.  Polarimetric image recovery method combining histogram stretching for underwater imaging , 2018, Scientific Reports.