面向水下环境的海洋生物轻量化目标检测模型

王哲; 田生伟; 王博; 周铁军

doi:10.13568/j.cnki.651094.651316.2021.10.31.0001

您当前的位置：

首页 >

文章列表页 >

面向水下环境的海洋生物轻量化目标检测模型

信息科学与技术 | 更新时间：2026-01-21

- 面向水下环境的海洋生物轻量化目标检测模型
- 面向水下环境的海洋生物轻量化目标检测模型
- 新疆大学学报（自然科学版中英文） 2022年39卷第5期页码：598-607
- 作者机构：
  
  1. 新疆大学软件学院
  2. 国家计算机网络与信息安全管理中心新疆分中心
- 作者简介：
- 基金信息：
  
  国家自然科学基金（U2003208）;新疆维吾尔自治区重点研发项目（2021B01002;2020E0234）
- DOI：10.13568/j.cnki.651094.651316.2021.10.31.0001
  中图分类号： Q178.53;TP391.41
- 纸质出版：2022
- 稿件说明：
移动端阅览
[1]王哲,田生伟,王博,等.面向水下环境的海洋生物轻量化目标检测模型[J].新疆大学学报(自然科学版)(中英文),2022,39(05):598-607.
[1]王哲,田生伟,王博,等.面向水下环境的海洋生物轻量化目标检测模型[J].新疆大学学报(自然科学版)(中英文),2022,39(05):598-607. DOI： 10.13568/j.cnki.651094.651316.2021.10.31.0001.

DOI：10.13568/j.cnki.651094.651316.2021.10.31.0001.

摘要

针对传统目标检测模型面对物体种类较少、背景相对单一的水下海洋生物数据集时，存在训练时间长、推理速度慢等问题，提出了一种快速轻量化的目标检测模型FL-Net(Fast and Lightweight Network).(1)采用Res Net18作为骨干网络，减少训练时间和计算量；(2)采用空洞卷积替换普通卷积，提高卷积核的感受野；(3)采用动态激活函数(MetaAcon)替换线性整流函数(Re LU)，增强骨干网络的特征提取能力；(4)采用单阶段的Generalized Focal Loss(GFL)方法作为网络头部，提高推理速度和准确率．实验结果表明：FL-Net在Brackish数据集上的准确率达到了79.3%AP(96.6%AP50)，平均推理速度为65.2 FPS.

Abstract

Aiming at the problem that the SOTA object detection method takes a long time on the underwater marine life dataset with fewer object types and a relatively single background

a fast and lightweight network FLNet(Fast and Lightweight Network) was proposed.(1) Use Res Net18 as the backbone to reduce training time and computing burden.(2) Use dilation convolution to replace ordinary convolution to improve the field-of-view of the convolution kernel.(3) Use MetaAcon to replace the Re LU activation function to improve the feature extraction capability of the backbone.(4) Use single-stage Generalized Focal Loss(GFL) method as head to improve inference speed and accuracy. The experimental results show that the accuracy of FL-Net on the Brackish dataset reaches79.3% AP(96.6% AP50)

and the average inference speed is 65.2 FPS.

关键词

Keywords

references

CAO X,WU C,YAN P,et al.Linear SVM classification using boosting HOG features for vehicle detection in low-altitude airborne videos:2011 18th IEEE International Conference on Image Processing[C].Brussels:IEEE,2011.

KRIZHEVSKY A,SUTSKEVER I,HINTON G E.Image Net classification with deep convolutional neural networks[J].Communications of the ACM,2017,60(6):84-90.

HE K,ZHANG X,REN S,et al.Deep residual learning for image recognition:2016 IEEE Conference on Computer Vision and Pattern Recognition(CVPR)[C].Las Vegas:IEEE,2016.

GAO S,CHENG M,ZHAO K,et al.Res2Net:a new multi-scale backbone architecture[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2021,43(2):652-662.

XIE S,GIRSHICK R,DOLL′AR P,et al.Aggregated residual transformations for deep neural networks:2017 IEEE Conference on Computer Vision and Pattern Recognition(CVPR)[C].Honolulu:IEEE,2017.

LIN T Y,DOLL′AR P,GIRSHICK R,et al.Feature pyramid networks for object detection:2017 IEEE Conference on Computer Vision and Pattern Recognition(CVPR)[C].Honolulu:IEEE,2017.

GIRSHICK R.Fast R-CNN:2015 IEEE International Conference on Computer Vision(ICCV)[C].Santiago:IEEE,2015.

REN S,HE K,GIRSHICK R,et al.Faster R-CNN:towards real-time object detection with region proposal networks[J].IEEETransactions on Pattern Analysis and Machine Intelligence,2017,39(2):1137-1149.

PANG J,CHEN K,SHI J,et al.Libra R-CNN:towards balanced learning for object detection:2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR)[C].Long Beach:IEEE,2019.

孟欣欣,阿里甫·库尔班,吕情深,等.基于迁移学习的自然环境下香梨目标识别研究[J].新疆大学学报(自然科学版),2019,36(4):461-467.

韩文轩,阿里甫·库尔班,黄梓桐.基于改进SSD算法的遥感影像小目标快速检测[J].新疆大学学报(自然科学版)(中英文),2020,37(2):163-169.

VILLON S,CHAUMONT M,SUBSOL G,et al.Coral reef fish detection and recognition in underwater videos by supervised machine learning:comparison between deep learning and HOG+SVM methods:Advanced Concepts for Intelligent Vision Systems[C].Cham:Springer,2016.

SIDDIQUI S A,SALMAN A,MALIK M I,et al.Automatic fish species classification in underwater videos:exploiting pre-trained deep neural network models to compensate for limited labelled data[J].ICES Journal of Marine Science,2018,75(1):374-389.

RASMUSSEN C,ZHAO J,FERRARO D,et al.Deep census:AUV-based scallop population monitoring:2017 IEEE International Conference on Computer Vision Workshops(ICCVW)[C].Venice:IEEE,2017.

PEDERSEN M,BRUSLUND HAURUM J,GADE R,et al.Detection of marine animals in a new underwater dataset with varying visibility:2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR)[C].Long Beach:IEEE,2019.

LIN T Y,MAIRE M,BELONGIE S,et al.Microsoft COCO:common objects in context:Computer Vision-ECCV 2014[C].Cham:Springer,2014.

EVERINGHAM M,VAN GOOL L,WILLIAMS C K I,et al.The PASCAL visual object classes(VOC) challenge[J].International Journal of Computer Vision,2010,88(2):303-338.

NAIR V,HINTON G E.Rectified linear units improve restricted boltzmann machines:Proceedings of the 27th International Conference on International Conference on Machine Learning[C].Madison:ICML,2010.

BODLA N,SINGH B,CHELLAPPA R,et al.Soft-NMS:improving object detection with one line of code:2017 IEEE International Conference on Computer Vision(ICCV)[C].Venice:IEEE,2017.

YANG Z,LIU S,HU H,et al.Rep Points:point set representation for object detection:2019 IEEE/CVF International Conference on Computer Vision(ICCV)[C].Seoul:IEEE,2019.

ZHANG S,CHI C,YAO Y,et al.Bridging the gap between anchor-based and anchor-free detection via adaptive training sample selection:2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR)[C].Seattle:IEEE,2020.

IOFFE S,SZEGEDY C.Batch normalization:accelerating deep network training by reducing internal covariate shift:Proceedings of the 32nd International Conference on Machine Learning[C].Lille:ICML,2015.

MA N,ZHANG X,LIU M,et al.Activate or not:learning customized activation:2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR)[C].Virtual:IEEE,2021.

LIN T Y,GOYAL P,GIRSHICK R,et al.Focal loss for dense object detection:2017 IEEE International Conference on Computer Vision(ICCV)[C].Venice:IEEE,2017.

LI X,WANG W,HU X,et al.Generalized focal loss V2:learning reliable localization quality estimation for dense object detection:2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR)[C].Virtual:IEEE,2021.

LI X,WANG W,WU L,et al.Generalized focal loss:learning qualified and distributed bounding boxes for dense object detection[J].Advances in Neural Information Proccessing Systems,2020,33:21002-21012.

PASZKE A,GROSS S,MASSA F,et al.Py Torch:an imperative style,high-performance deep learning library[J].Advances in Neural Information Processing Systems,2019,32:8026-8037.

CAI Z,VASCONCELOS N.Cascade R-CNN:delving into high quality object detection:2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR)[C].Salt Lake City:IEEE,2018.

KIM K,LEE H S.Probabilistic anchor assignment with I o U prediction for object detection:Computer Vision-ECCV 2020[C].Glasgow:ECCV,2020.

ZHANG X,WAN F,LIU C,et al.Free Anchor:learning to match anchors for visual object detection[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2022,44(6):3096-3109.

LU X,LI B,YUE Y,et al.Grid R-CNN:2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR)[C].Long Beach:IEEE,2019.

ZHU C,HE Y,SAVVIDES M.Feature selective anchor-free module for single-shot object detection:2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR)[C].Long Beach:IEEE,2019.

TIAN Z,SHEN C,CHEN H,et al.FCOS:fully convolutional one-stage object detection:2019 IEEE/CVF International Conference on Computer Vision(ICCV)[C].Seoul:IEEE,2019.

CHEN Q,WANG Y,YANG T,et al.You only look one-level feature:2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR)[C].Virtual:IEEE,2019.

浏览量

1389

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于改进SSD算法的遥感影像小目标快速检测

温度引导的CNN-Transformer红外与可见光图像融合方法

遗传算法在瞬变电磁深度学习反演中的优化策略

皮肤癌病变组织分割的多通道并行U型网络

基于锯齿空洞残差卷积的单幅图像超分辨率重建研究