基于梯度引导加权‒延迟负梯度衰减损失的长尾图像缺陷检测

doi:10.11772/j.issn.1001-9081.2022091413

《计算机应用》唯一官方网站 ›› 2023, Vol. 43 ›› Issue (10): 3267-3274.DOI: 10.11772/j.issn.1001-9081.2022091413

所属专题：多媒体计算与计算机仿真

• 多媒体计算与计算机仿真 • 上一篇下一篇

基于梯度引导加权‒延迟负梯度衰减损失的长尾图像缺陷检测

李巍(), 梁斯昕, 张建州

四川大学计算机学院，成都 610065

收稿日期:2022-09-26 修回日期:2022-12-04 接受日期:2022-12-12 发布日期:2023-03-07 出版日期:2023-10-10
通讯作者: 李巍
作者简介:李巍（1998—），男，江苏泰州人，硕士研究生，主要研究方向：缺陷检测、目标检测. wana@stu.scu.edu.cn
梁斯昕（1995—），男，云南曲靖人，博士研究生，主要研究方向：机器视觉、摄像机标定、医学图像处理
张建州（1962—），男，河北邯郸人，教授，博士，主要研究方向：机器视觉、计算机视觉。
基金资助:
中国博士后科学基金资助项目(2022M712235);四川大学专职博士后研发基金资助项目(2022SCU12074)

Long-tailed image defect detection based on gradient-guide weighted-deferred negative gradient decay loss

Wei LI(), Sixin LIANG, Jianzhou ZHANG

College of Computer Science，Sichuan University，Chengdu Sichuan 610065，China

Received:2022-09-26 Revised:2022-12-04 Accepted:2022-12-12 Online:2023-03-07 Published:2023-10-10
Contact: Wei LI
About author:LI Wei， born in 1998， M. S. candidate. His research interests include defect detection， object detection.
LIANG Sixin， born in 1995， Ph. D. candidate. His researchinterests include machine vision， camera calibration， medical image processing.
ZHANG Jianzhou， born in 1962， Ph. D.， professor. His research interests include machine vision， computer vision.
Supported by:
China Postdoctoral Science Foundation(2022M712235);Postdoctoral Research and Development Foundation of Sichuan University(2022SCU12074)

摘要/Abstract

摘要：

针对目前图像缺陷检测模型对长尾缺陷数据集中尾部类检测效果较差的问题，提出一个基于梯度引导加权?延迟负梯度衰减损失（GGW-DND Loss）。首先，根据检测器分类节点的累积梯度比值分别对正负梯度重新加权，减轻尾部类分类器的受抑制状态；其次，当模型优化到一定阶段时，直接降低每个节点产生的负梯度，以增强尾部类分类器的泛化能力。实验结果表明，在自制图像缺陷数据集和NEU-DET（NEU surface defect database for Defect Detection Task）上，所提损失的尾部类平均精度均值（mAP）优于二分类交叉熵损失（BCE Loss），分别提高了32.02和7.40个百分点；与EQL v2（EQualization Loss v2）相比，分别提高了2.20和0.82个百分点，验证了所提损失能有效提升网络对尾部类的检测性能。

关键词: 长尾数据集, 累计梯度比值, 加权损失, 图像缺陷检测, 卷积神经网络

Abstract:

Aiming at the problem that the current image defect detection models have poor detection effect on tail categories in long-tail defect datasets， a GGW-DND Loss （Gradient-Guide Weighted-Deferred Negative Gradient decay Loss） was proposed. First， the positive and negative gradients were re-weighted according to the cumulative gradient ratio of the classification nodes in the detector in order to reduce the suppressed state of tail classifier. Then， once the model was optimized to a certain stage， the negative gradient generated by each node was sharply reduced to enhance the generalization ability of the tail classifier. Experimental results on the self-made image defect dataset and NEU-DET （NEU surface defect database for Defect dEtection Task） show that the mean Average Precision （mAP） for tail categories of the proposed loss is better than that of Binary Cross Entropy Loss （BCE Loss）， the former is increased by 32.02 and 7.40 percentage points respectively， and compared with EQL v2 （EQualization Loss v2）， the proposed loss has the mAP increased by 2.20 and 0.82 percentage points respectively， verifying that the proposed loss can effectively improve the detection performance of the network for tail categories.

Key words: long-tail dataset, cumulative gradient ratio, weighted loss, image defect detection, Convolutional Neural Network (CNN)

中图分类号:

TP391

李巍, 梁斯昕, 张建州. 基于梯度引导加权‒延迟负梯度衰减损失的长尾图像缺陷检测[J]. 计算机应用, 2023, 43(10): 3267-3274.

Wei LI, Sixin LIANG, Jianzhou ZHANG. Long-tailed image defect detection based on gradient-guide weighted-deferred negative gradient decay loss[J]. Journal of Computer Applications, 2023, 43(10): 3267-3274.

图/表 14

图1 Faster-RCNN的基本网络结构

Fig. 1 Basic network structure of Faster-RCNN

图2 基于GGW-DND Loss的类别预测网络

Fig. 2 Category prediction network based on GGW-DND Loss

图3 图像缺陷检测数据集样例

Fig. 3 Samples of image defect detection dataset

表1 不同缺陷检测网络的检测结果对比 (%)

Tab. 1 Comparison of detection results of different defect detection networks

模型	骨干网络	损失函数	mAP	mAP_f	mAP_r
Faster R-CNN	ResNet-50	CE Loss	70.05	92.04	48.06
Faster R-CNN	ResNet-50	BCE Loss	68.31	92.60	44.02
SSD	VGG16	CE Loss	64.61	86.68	42.54
YOLOv5	CSP-DarkNet	BCE Loss	49.96	80.67	19.24

表2 基于不同损失函数的检测结果对比 (%)

Tab. 2 Comparison of detection results based on different loss functions

损失函数	mAP	mAP_f	mAP_r
CE Loss	70.05	92.04	48.06
BCE Loss	68.31	92.60	44.02
Focal Loss	54.05	87.28	20.82
Logit Adjustment	72.12	86.72	57.52
EQL	73.92	91.97	55.87
BAGS	82.49	91.62	73.35
EQL v2	83.06	92.28	73.84
GGW-DND Loss	84.09	92.13	76.04

图4 不同β和γ组合下的映射函数

Fig. 4 Mapping functions under different combinations of β and γ

表3 不同组合参数β和γ下的检测结果

Tab. 3 Detection results under different combinations of β and γ

$β$	$γ$	mAP/%	mAP_f/%	mAP_r/%
6	0.6	82.96	92.20	73.71
	0.7	82.60	90.83	74.37
	0.8	83.73	91.72	75.73
8	0.6	82.09	91.79	72.39
	0.7	83.69	92.12	75.26
	0.8	80.70	90.73	70.67
10	0.6	83.07	91.90	74.25
	0.7	81.92	91.41	72.43
	0.8	76.62	89.68	63.56
12	0.6	80.48	91.83	69.12
	0.7	80.46	91.19	69.73
	0.8	75.10	89.14	61.07

表3 不同组合参数β和γ下的检测结果

Tab. 3 Detection results under different combinations of β and γ

$β$	$γ$	mAP/%	mAP_f/%	mAP_r/%
6	0.6	82.96	92.20	73.71
	0.7	82.60	90.83	74.37
	0.8	83.73	91.72	75.73
8	0.6	82.09	91.79	72.39
	0.7	83.69	92.12	75.26
	0.8	80.70	90.73	70.67
10	0.6	83.07	91.90	74.25
	0.7	81.92	91.41	72.43
	0.8	76.62	89.68	63.56
12	0.6	80.48	91.83	69.12
	0.7	80.46	91.19	69.73
	0.8	75.10	89.14	61.07

表4 不同α下的检测结果

Tab. 4 Detection results at different α

$α$	mAP/%	mAP_f/%	mAP_r/%
1	77.75	91.29	64.22
2	79.37	91.67	67.06
4	81.54	91.80	71.29
8	83.69	92.12	75.26
12	82.38	91.49	73.27

表4 不同α下的检测结果

Tab. 4 Detection results at different α

$α$	mAP/%	mAP_f/%	mAP_r/%
1	77.75	91.29	64.22
2	79.37	91.67	67.06
4	81.54	91.80	71.29
8	83.69	92.12	75.26
12	82.38	91.49	73.27

图5 不同λ下缩放后的对应负梯度

Fig. 5 Corresponding negative gradient after scaling at different λ

表5 不同λ下的检测结果

Tab. 5 Detection results at different λ

$λ$	mAP/%	mAP_f/%	mAP_r/%
1	83.06	92.28	73.84
2	83.68	92.38	74.98
4	83.78	92.16	75.39
6	84.09	92.13	76.04
8	83.69	92.12	75.26
12	83.99	92.57	75.42

表5 不同λ下的检测结果

Tab. 5 Detection results at different λ

$λ$	mAP/%	mAP_f/%	mAP_r/%
1	83.06	92.28	73.84
2	83.68	92.38	74.98
4	83.78	92.16	75.39
6	84.09	92.13	76.04
8	83.69	92.12	75.26
12	83.99	92.57	75.42

表6 基于BCE Loss和GGW-DND Loss的分类器节点的累计梯度比值

Tab. 6 Cumulative gradient ratios of classifier nodes based on BCE Loss and GGW-DND Loss

损失函数	马赛克	偏色	亮度异常	高斯噪声	椒盐噪声	坏块
BCE Loss	0.73	0.65	0.67	0.13	0.10	0.18
GGW-DND Loss	1.22	1.16	1.16	0.84	0.84	0.98

表7 消融实验结果 (%)

Tab. 7 Results of ablation experiment

GGW Loss	DND Loss	mAP	mAP_f	mAP_r
		68.31	92.60	44.02
√		83.06	92.28	73.84
√	√	84.09	92.13	76.04

表8 NEU-DET数据集

Tab. 8 NEU-DET dataset

缺陷类型	训练集样本数	测试集样本数
银纹	482	205
夹杂物	630	287
斑块	570	270
麻点表面	4	128
轧入氧化皮	3	185
划痕	3	155

表9 NEU-DET数据集上基于不同损失函数的检测结果比较 (%)

Tab. 9 Comparison of detection results based on different loss functions on NEU-DET dataset

损失函数	mAP	mAP_f	mAP_r
BCE Loss^［33］	32.38	64.76	0.01
Focal Loss^［38］	33.45	66.90	0.00
Logit Adjustment^［54］	34.08	66.24	1.93
EQL^［55］	36.48	69.76	3.20
BAGS^［40］	34.89	66.03	3.75
EQL v2^［47］	36.71	66.83	6.59
GGW-DND Loss	36.85	66.29	7.41

参考文献 55

1	王志明. 无参考图像质量评价综述［J］. 自动化学报， 2015， 41（6）：1062-1079. 10.16383/j.aas.2015.c140404
	WANG Z M. Review of no-reference image quality assessment［J］. Acta Automatica Sinica， 2015， 41（6）： 1062-1079. 10.16383/j.aas.2015.c140404
2	房亚男. 数字视频图像质量检测关键技术研究［J］. 科技创新导报， 2013（16）： 22-26. 10.3969/j.issn.1674-098X.2013.16.016
	FANG Y N. Research on key technologies of digital video image quality detection［J］. Science and Technology Innovation Herald， 2013（16）： 22-26. 10.3969/j.issn.1674-098X.2013.16.016
3	WANG F L， ZUO B. Detection of surface cutting defect on MagNet using Fourier image reconstruction［J］. Journal of Central South University， 2016， 23（5）： 1123-1131. 10.1007/s11771-016-0362-y
4	WEE C Y， PARAMESRAN R. Image sharpness measure using eigenvalues［C］// Proceedings of the 9th International Conference on Signal Processing. Piscataway： IEEE， 2008： 840-843. 10.1109/icosp.2008.4697259
5	HOU Z， PARKER J M. Texture defect detection using support vector machines with adaptive Gabor wavelet features［C］// Proceedings of the 7th IEEE Workshops on Applications of Computer Vision. Piscataway： IEEE， 2005： 275-280. 10.1109/acvmot.2005.115
6	CHA Y J， CHOI W， SUH G， et al. Autonomous structural visual inspection using region-based deep learning for detecting multiple damage types［J］. Computer-Aided Civil and Infrastructure Engineering， 2018， 33（9）： 731-747. 10.1111/mice.12334
7	TAO X， ZHANG D P， WANG Z H， et al. Detection of power line insulator defects using aerial images analyzed with convolutional neural networks［J］. IEEE Transactions on Systems， Man， and Cybernetics： Systems， 2020， 50（4）： 1486-1498. 10.1109/tsmc.2018.2871750
8	HE Y， SONG K C， MENG Q G， et al. An end-to-end steel surface defect detection approach via fusing multiple hierarchical features［J］. IEEE Transactions on Instrumentation and Measurement， 2020， 69（4）： 1493-1504. 10.1109/tim.2019.2915404
9	陶显，侯伟，徐德. 基于深度学习的表面缺陷检测方法综述［J］. 自动化学报， 2021， 47（5）：1017-1034.
	TAO X， HOU W， XU D. A survey of surface defect detection methods based on deep learning［J］. Acta Automatica Sinica， 2021， 47（5）： 1017-1034.
10	LI J Y， SU Z F， GENG J H， et al. Real-time detection of steel strip surface defects based on improved YOLO detection network［J］. IFAC-PapersOnLine， 2018， 51（21）： 76-81. 10.1016/j.ifacol.2018.09.412
11	ZHANG C B， CHANG C C， JAMSHIDI M. Concrete bridge surface damage detection using a single-stage detector［J］. Computer-Aided Civil and Infrastructure Engineering， 2020， 35（4）： 389-409. 10.1111/mice.12500
12	CHEN S H， TSAI C C. SMD LED chips defect detection using a YOLOv3-dense model［J］. Advanced Engineering Informatics， 2021， 47： No.101255. 10.1016/j.aei.2021.101255
13	RUFF L， VANDERMEULEN R A， GÖRNITZ N， et al. Deep one-class classification［C］// Proceedings of the 35th International Conference on Machine Learning. New York： JMLR.org， 2018： 4393-4402.
14	BERGMANN P， FAUSER M， SATTLEGGER D， et al. Uninformed students： student-teacher anomaly detection with discriminative latent embeddings［C］// Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2020： 4182-4191. 10.1109/cvpr42600.2020.00424
15	LIZNERSKI P， RUFF L， VANDERMEULEN R A， et al. Explainable deep one-class classification［EB/OL］. （2021-03-18）［2022-09-09］..
16	GOLAN I， EL-YANIV R. Deep anomaly detection using geometric transformations［C］// Proceedings of the 32nd International Conference on Neural Information Processing Systems. Red Hook， NY： Curran Associates Inc.， 2018： 9781-9791.
17	HASELMANN M， GRUBER D P， TABATABAI P. Anomaly detection using deep learning based image completion［C］// Proceedings of the 17th IEEE International Conference on Machine Learning and Applications. Piscataway： IEEE， 2018： 1237-1242. 10.1109/icmla.2018.00201
18	SAKURADA M， YAIRI T. Anomaly detection using autoencoders with nonlinear dimensionality reduction［C］// Proceedings of the MLSDA 2nd Workshop on Machine Learning for Sensory Data Analysis. New York： ACM， 2014： 4-11. 10.1145/2689746.2689747
19	ZHOU C， PAFFENROTH R C. Anomaly detection with robust deep autoencoders［C］// Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. New York： ACM， 2017： 665-674. 10.1145/3097983.3098052
20	GUPTA K， BHAVSAR A， SAO A K. Detecting mitotic cells in HEp-2 images as anomalies via one class classifier［J］. Computers in Biology and Medicine， 2019， 111： No.103328. 10.1016/j.compbiomed.2019.103328
21	NAPOLETANO P， PICCOLI F， SCHETTINI R. Anomaly detection in nanofibrous materials by CNN-based self-similarity［J］. Sensors， 2018， 18（1）： No.209. 10.3390/s18010209
22	CANDÈS E J， LI X D， MA Y， et al. Robust principal component analysis？［J］. Journal of the ACM， 2011， 58（3）： No.11. 10.1145/1970392.1970395
23	GONG D， LIU L Q， LE V， et al. Memorizing normality to detect anomaly： memory-augmented deep autoencoder for unsupervised anomaly detection［C］// Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision. Piscataway： IEEE， 2019： 1705-1714. 10.1109/iccv.2019.00179
24	SCHLEGL T， SEEBÖCK P， WALDSTEIN S M， et al. Unsupervised anomaly detection with generative adversarial networks to guide marker discovery［C］// Proceedings of the 2017 International Conference on Information Processing in Medical Imaging， LNCS 10265. Cham： Springer， 2017： 146-157.
25	LIN T Y， MAIRE M， BELONGIE S， et al. Microsoft COCO： common objects in context［C］// Proceedings of the 2014 European Conference on Computer Vision， LNCS 8693. Cham： Springer， 2014： 740-755.
26	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. 10.1007/s11263-009-0275-4
27	BUDA M， MAKI A， MAZUROWSKI M A. A systematic study of the class imbalance problem in convolutional neural networks［J］. Neural Networks， 2018， 106： 249-259. 10.1016/j.neunet.2018.07.011
28	HE H B， GARCIA E A. Learning from imbalanced data［J］. IEEE Transactions on Knowledge and Data Engineering， 2009， 21（9）： 1263-1284. 10.1109/tkde.2008.239
29	OUYANG W L， WANG X G， ZHANG C， et al. Factors in finetuning deep model for object detection with long-tail distribution［C］// Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2016： 864-873. 10.1109/cvpr.2016.100
30	BYRD J， LIPTON Z C. What is the effect of importance weighting in deep learning？［C］// Proceedings of the 36th International Conference on Machine Learning. New York： JMLR.org， 2019： 872-881.
31	SHEN L， LIN Z C， HUANG Q M. Relay backpropagation for effective learning of deep convolutional neural networks［C］// Proceedings of the 2016 European Conference on Computer Vision， LNCS 9911. Cham： Springer， 2016： 467-482.
32	CHAWLA N V， BOWYER K W， HALL L O， et al. SMOTE： synthetic minority over-sampling technique［J］. Journal of Artificial Intelligence Research， 2002， 16： 321-357. 10.1613/jair.953
33	CUI Y， JIA M L， LIN T Y， et al. Class-balanced loss based on effective number of samples［C］// Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2019： 9260-9269. 10.1109/cvpr.2019.00949
34	WANG Y X， RAMANAN D， HEBERT M. Learning to model the tail［C］// Proceedings of the 31st International Conference on Neural Information Processing Systems. Red Hook， NY： Curran Associates Inc.， 2017： 7032-7042.
35	HUANG C， LI Y N， LOY C C， et al. Learning deep representation for imbalanced classification［C］// Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2016： 5375-5384. 10.1109/cvpr.2016.580
36	ZHANG X， FANG Z Y， WEN Y D， et al. Range loss for deep face recognition with long-tailed training data［C］// Proceedings of the 2017 IEEE International Conference on Computer Vision. Piscataway： IEEE， 2017： 5419-5428. 10.1109/iccv.2017.578
37	CAO K D， WEI C， GAIDON A， et al. Learning imbalanced datasets with label-distribution-aware margin loss［C］// Proceedings of the 33rd International Conference on Neural Information Processing Systems. Red Hook， NY： Curran Associates Inc.， 2019： 1567-1578.
38	LIN T Y， GOYAL P， GIRSHICK R， et al. Focal loss for dense object detection［C］// Proceedings of the 2017 IEEE International Conference on Computer Vision. Piscataway： IEEE， 2017： 2999-3007. 10.1109/iccv.2017.324
39	LI B Y， LIU Y， WANG X G. Gradient harmonized single-stage detector［C］// Proceedings of the 33rd AAAI Conference on Artificial Intelligence. Palo Alto， CA： AAAI Press， 2019： 8577-8584. 10.1609/aaai.v33i01.33018577
40	LI Y， WANG T， KANG B Y， et al. Overcoming classifier imbalance for long-tail object detection with balanced group softmax［C］// Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2020： 10988-10997. 10.1109/cvpr42600.2020.01100
41	YIN X， YU X， SOHN K， et al. Feature transfer learning for face recognition with under-represented data［C］// Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2019： 5697-5706. 10.1109/cvpr.2019.00585
42	LIU Z W， MIAO Z Q， ZHAN X H， et al. Large-scale long-tailed recognition in an open world［C］// Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2019： 2532-2541. 10.1109/cvpr.2019.00264
43	CHU P， BIAN X， LIU S P， et al. Feature space augmentation for long-tailed data［C］// Proceedings of the 2020 European Conference on Computer Vision， LNCS 12374. Cham： Springer， 2020： 694-710.
44	REN S Q， HE K M， GIRSHICK R， et al. Faster R-CNN： towards real-time object detection with region proposal networks［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence， 2017， 39（6）： 1137-1149. 10.1109/tpami.2016.2577031
45	HE K M， ZHANG X Y， REN S Q， et al. Deep residual learning for image recognition［C］// Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2016： 770-778. 10.1109/cvpr.2016.90
46	HE K M， GKIOXARI G， DOLLÁR P， et al. Mask R-CNN［C］// Proceedings of the 2017 IEEE International Conference on Computer Vision. Piscataway： IEEE， 2017： 2980-2988. 10.1109/iccv.2017.322
47	TAN J R， LU X， ZHANG G， et al. Equalization loss v2： a new gradient balance approach for long-tailed object detection［C］// Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2021： 1685-1694. 10.1109/cvpr46437.2021.00173
48	HE Y， SONG K， MENG Q， et al. An end-to-end steel surface defect detection approach via fusing multiple hierarchical features［J］. IEEE Transactions on Instrumentation and Measurement， 2019， 69（4）： 1493-1504. 10.1109/tim.2019.2915404
49	KINGMA D P， BA J L. Adam： a method for stochastic optimization［EB/OL］. （2017-01-30）［2022-07-13］..
50	LIU W， ANGUELOV D， ERHAN D， et al. SSD： single shot multiBox detector［C］// Proceedings of the 2016 European Conference on Computer Vision， LNCS 9905. Cham： Springer， 2016： 21-37.
51	KUZNETSOVA A， MALEVA T， SOLOVIEV V. YOLOv5 versus YOLOv3 for apple detection［M］// KRAVETS A G， BOLSHAKOV A A， SHCHERBAKOV M. Cyber-Physical Systems： Modelling and Intelligent Control， SSDC 338. Cham： Springer， 2021： 349-358. 10.1007/978-3-030-66077-2_28
52	SIMONYAN K， ZISSERMAN A. Very deep convolutional networks for large-scale image recognition［EB/OL］. （2015-04-10）［2022-07-24］..
53	REDMON J， FARHADI A. YOLOv3： an incremental improvement［EB/OL］. （2018-04-08）［2022-08-09］.. 10.1109/cvpr.2017.690
54	MENON A K， JAYASUMANA S， RAWAT A S， et al. Long-tail learning via logit adjustment［EB/OL］. （2021-07-09）［2022-05-22］..
55	TAN J R， WANG C B， LI B Y， et al. Equalization loss for long-tailed object recognition［C］// Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2020： 11659-11668. 10.1109/cvpr42600.2020.01168

[1]	秦璟, 秦志光, 李发礼, 彭悦恒. 基于概率稀疏自注意力神经网络的重性抑郁疾患诊断[J]. 《计算机应用》唯一官方网站, 2024, 44(9): 2970-2974.
[2]	李云, 王富铕, 井佩光, 王粟, 肖澳. 基于不确定度感知的帧关联短视频事件检测方法[J]. 《计算机应用》唯一官方网站, 2024, 44(9): 2903-2910.
[3]	陈虹, 齐兵, 金海波, 武聪, 张立昂. 融合1D-CNN与BiGRU的类不平衡流量异常检测[J]. 《计算机应用》唯一官方网站, 2024, 44(8): 2493-2499.
[4]	张春雪, 仇丽青, 孙承爱, 荆彩霞. 基于两阶段动态兴趣识别的购买行为预测模型[J]. 《计算机应用》唯一官方网站, 2024, 44(8): 2365-2371.
[5]	赵宇博, 张丽萍, 闫盛, 侯敏, 高茂. 基于改进分段卷积神经网络和知识蒸馏的学科知识实体间关系抽取[J]. 《计算机应用》唯一官方网站, 2024, 44(8): 2421-2429.
[6]	王东炜, 刘柏辰, 韩志, 王艳美, 唐延东. 基于低秩分解和向量量化的深度网络压缩方法[J]. 《计算机应用》唯一官方网站, 2024, 44(7): 1987-1994.
[7]	高阳峄, 雷涛, 杜晓刚, 李岁永, 王营博, 闵重丹. 基于像素距离图和四维动态卷积网络的密集人群计数与定位方法[J]. 《计算机应用》唯一官方网站, 2024, 44(7): 2233-2242.
[8]	沈君凤, 周星辰, 汤灿. 基于改进的提示学习方法的双通道情感分析模型[J]. 《计算机应用》唯一官方网站, 2024, 44(6): 1796-1806.
[9]	黄梦源, 常侃, 凌铭阳, 韦新杰, 覃团发. 基于层间引导的低光照图像渐进增强算法[J]. 《计算机应用》唯一官方网站, 2024, 44(6): 1911-1919.
[10]	李健京, 李贯峰, 秦飞舟, 李卫军. 基于不确定知识图谱嵌入的多关系近似推理模型[J]. 《计算机应用》唯一官方网站, 2024, 44(6): 1751-1759.
[11]	姚迅, 秦忠正, 杨捷. 生成式标签对抗的文本分类模型[J]. 《计算机应用》唯一官方网站, 2024, 44(6): 1781-1785.
[12]	孙敏, 成倩, 丁希宁. 基于CBAM-CGRU-SVM的Android恶意软件检测方法[J]. 《计算机应用》唯一官方网站, 2024, 44(5): 1539-1545.
[13]	席治远, 唐超, 童安炀, 王文剑. 基于双路时空网络的驾驶员行为识别[J]. 《计算机应用》唯一官方网站, 2024, 44(5): 1511-1519.
[14]	高文烁, 陈晓云. 基于节点结构的点云分类网络[J]. 《计算机应用》唯一官方网站, 2024, 44(5): 1471-1478.
[15]	王杰, 孟华. 基于点云整体拓扑结构的图像分类算法[J]. 《计算机应用》唯一官方网站, 2024, 44(4): 1107-1113.

基于梯度引导加权‒延迟负梯度衰减损失的长尾图像缺陷检测

Long-tailed image defect detection based on gradient-guide weighted-deferred negative gradient decay loss

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 14

参考文献 55

相关文章 15

编辑推荐

Metrics