[1]卜子渝,杨 哲,刘纯平.基于 EfficientNet 的无锚框目标检测模型[J].计算机技术与发展,2024,34(01):37-43.[doi:10. 3969 / j. issn. 1673-629X. 2024. 01. 006]
　BU Zi-yu,YANG Zhe,LIU Chun-ping.An Anchor-free Object Detection Model Based on EfficientNet[J].,2024,34(01):37-43.[doi:10. 3969 / j. issn. 1673-629X. 2024. 01. 006]
点击复制

基于 EfficientNet 的无锚框目标检测模型()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
34
期数:
2024年01期
页码:
37-43
栏目:
媒体计算
出版日期:
2024-01-10

文章信息/Info

Title:
An Anchor-free Object Detection Model Based on EfficientNet
文章编号:
1673-629X(2024)01-0037-07
作者:
卜子渝1 ; 杨 哲1; 2; 3 ; 刘纯平2; 3
1. 苏州大学 计算机科学与技术学院,江苏 苏州 215006;
2. 江苏省计算机信息处理技术重点实验室,江苏 苏州 215006;
3. 江苏省大数据智能工程实验室,江苏 苏州 215006
Author(s):
BU Zi-yu1 ; YANG Zhe1; 2; 3 ; LIU Chun-ping2; 3
1. School of Computer Science and Technology,Soochow University,Suzhou 215006,China;
2. Provincial Key Laboratory for Computer Information Processing Technology,Suzhou 215006,China;
3. Provincial Key Laboratory for Intelligent Engineering in Big Data,Suzhou 215006,China
关键词:
深度学习; 计算机视觉; 目标检测; 正负样本分配算法; 无锚框
Keywords:
deep learning; computer vision; object detection; positive / negative samples assignment algorithm; anchor-free
分类号:
TP391
DOI:
10. 3969 / j. issn. 1673-629X. 2024. 01. 006
摘要:
目标检测是计算机视觉的热门研究方向之一,包含分类和定位两个任务。 针对单阶段目标检测模型普遍存在的两个问题:训练时正负样本的不均衡以及锚框的设置需要人工干预,提出一
种基于 EfficientNet 的无锚框目标检测模型(Anchor-free Efficientnet-based Object Detector,AEOD)。 AEOD 先筛选出落在目标框中的特征点,再根据特征点所作的预测计算代价矩阵,
在训练时基于代价矩阵为目标动态分配正负样本,从而达到平衡二者数量的目的。 此模型通过特征图中的特征点直接预测目标的位置和形状,不仅省去了人工设置锚框的环节,还提高了
可检出目标的数量。 此外,可缩放的EfficientNet 进一步提高了模型的泛化能力,使之可以接收多尺度的输入。 在 PASCAL VOC07 +12 数据集中,AEOD 最高可以获得 91. 3% 的平均精度(mAP) ,检测速度达到 32. 1 FPS,较其他主流的目标检测模型有显著提升。
Abstract:
Object detection is one of the hot research areas in computer vision,which includes two tasks:classification and location. Dueto the two common problems appearing in one - stage object detector: extreme imbalance between positive / negative samples duringtraining and anchors pre-defined deeply depending on manual settings,an anchor-free efficientnet-based object detector ( AEOD) is proposed. AEOD first selects out the feature points that fall in the target box,then calculates the cost matrix based on values predicted bythese feature points, finally assigns the positive / negative samples to the target dynamically according to the cost matrix during thetraining. Therefore,the number of positive / negative samples is balanced to enhance the performance of the model. AEOD directlypredicts location and shape of the object through the feature points in the feature maps. As a result,not only the step of pre - defininganchors can be skipped, but also the number of objects that successfully detected increases. In addition, the scalable backbone( EfficientNet) improves the generalization ability of AEOD,it can receive multi-scale input. AEOD achieves the highest 91. 3% mAPon PASCAL VOC07+12 at speed of 32. 1 FPS,showing a significant improvement compared to other modern models.

相似文献/References:

[1]黄艳 赵越.3D靶标的摄像机三步标定算法与实现[J].计算机技术与发展,2010,(01):135.
　HUANG Yan,ZHAO Yue.Algorithm and Realization of Three-step Camera Calibration Based on 3D-Target[J].,2010,(01):135.
[2]付海洋 牛连强 刘守琳.一种基于平面模板的单应矩阵求解方法[J].计算机技术与发展,2010,(04):69.
　FU Hai-yang,NIU Lian-qiang,LIU Shou-lin.A Solving Homography Matrix Method Based on Planar Pattern[J].,2010,(01):69.
[3]张铖伟 王彪 徐贵力.摄像机标定方法研究[J].计算机技术与发展,2010,(11):174.
　ZHANG Cheng-wei,WANG Biao,XU Gui-li.A Study on Classification of Camera Calibration Methods[J].,2010,(01):174.
[4]毛雁明 杨慧玲.一种新的立体匹配算法[J].计算机技术与发展,2011,(03):105.
　MAO Yan-ming,YANG Hui-ling.A New Stereo Matching Algorithm[J].,2011,(01):105.
[5]杨晟,李学军,王珏,等.连续尺度复合分析核线重排列影像准稠密匹配[J].计算机技术与发展,2013,(04):111.
　YANG Sheng,LI Xue-jun,WANG Jue,et al.Continuous Scale Multi-change Detecting Quasi-dense Matching for Epipolar Resample Images[J].,2013,(01):111.
[6]卢振宇,郭星,魏赛,等.基于计算机视觉的虚拟安全空间预警技术[J].计算机技术与发展,2014,24(02):237.
　LU Zhen-yu,GUO Xing,WEI Sai,et al.A Surveillance Technology for Virtual Security Space Based on Computer Vision[J].,2014,24(01):237.
[7]李孟,周波,孟正大,等. 三目立体相机的标定研究[J].计算机技术与发展,2015,25(02):69.
　LI Meng,ZHOU Bo,MENG Zheng-da,et al. Study on Trinocular Stereo Camera Calibration[J].,2015,25(01):69.
[8]陈强锐,谢世朋.基于深度学习的肺部肿瘤检测方法[J].计算机技术与发展,2018,28(04):201.[doi:10.3969/ j. issn.1673-629X.2018.04.043]
　CHEN Qiang-rui,XIE Shi-peng.Lung Cancer Detection Method Based on Deep Learning[J].,2018,28(01):201.[doi:10.3969/ j. issn.1673-629X.2018.04.043]
[9]黄法秀,张世杰,吴志红,等.数据增广下的人脸识别研究[J].计算机技术与发展,2020,30(03):67.[doi:10. 3969 / j. issn. 1673-629X. 2020. 03. 013]
　HUANG Fa-xiu,ZHANG Shi-jie,WU Zhi-hong,et al.Research on Face Recognition Based on Data Augmentation[J].,2020,30(01):67.[doi:10. 3969 / j. issn. 1673-629X. 2020. 03. 013]
[10]陈浩翔,蔡建明,刘铿然,等. 手写数字深度特征学习与识别[J].计算机技术与发展,2016,26(07):19.
　CHEN Hao-xiang,CAI Jian-ming,LIU Keng-ran,et al. Deep Learning and Recognition of Handwritten Numeral Features[J].,2016,26(01):19.
[11]施泽浩,赵启军.基于全卷积网络的目标检测算法[J].计算机技术与发展,2018,28(05):55.[doi:10.3969/j.issn.1673－629X.2018.05.013]
　SHI Ze-hao,ZHAO Qi-jun.Object Detection Algorithm Based on Fully Convolutional Neural Network[J].,2018,28(01):55.[doi:10.3969/j.issn.1673－629X.2018.05.013]
[12]许必宵,宫 婧,孙知信.基于卷积神经网络的目标检测模型综述[J].计算机技术与发展,2019,29(12):87.[doi:10. 3969 / j. issn. 1673-629X. 2019. 12. 016]
　XU Bi-xiao,GONG Jing,SUN Zhi-xin.A Survey of Object Detection Models Based on Convolutional Neural Networks[J].,2019,29(01):87.[doi:10. 3969 / j. issn. 1673-629X. 2019. 12. 016]
[13]张誉馨,张索非,王文龙,等.面向行人重识别的多域批归一化问题研究[J].计算机技术与发展,2022,32(01):91.[doi:10. 3969 / j. issn. 1673-629X. 2022. 01. 016]
　ZHANG Yu-xin,ZHANG Suo-fei,WANG Wen-long,et al.Research on Batch Normalization for Multi-domain PersonRe-identification[J].,2022,32(01):91.[doi:10. 3969 / j. issn. 1673-629X. 2022. 01. 016]
[14]陈晓艺,陆一鸣,沈加炜,等.基于深度学习的灾后建筑物损坏程度检测综述[J].计算机技术与发展,2023,33(09):1.[doi:10. 3969 / j. issn. 1673-629X. 2023. 09. 001]
　CHEN Xiao-yi,LU Yi-ming,SHEN Jia-wei,et al.Review of Post-disaster Building Damage Detection Based on Deep Learning[J].,2023,33(01):1.[doi:10. 3969 / j. issn. 1673-629X. 2023. 09. 001]

更新日期/Last Update: 2024-01-10