The DOI matches its Crossref record; 54 of the 54 references it cites were found in a registry or an index.
Data Augmentation Using Random Image Cropping and Patching for Deep CNNs
Open accessAbstract
Deep convolutional neural networks (CNNs) have achieved remarkable results in image processing tasks. However, their high expression ability risks overfitting. Consequently, data augmentation techniques have been proposed to prevent overfitting while enriching datasets. Recent CNN architectures with more parameters are rendering traditional data augmentation techniques insufficient. In this study, we propose a new data augmentation technique called random image cropping and patching (RICAP) which randomly crops four images and patches them to create a new training image. Moreover, RICAP mixes the class labels of the four images, resulting in an advantage of the soft labels. We evaluated RICAP with current state-of-the-art CNNs (e.g., the shake-shake regularization model) by comparison with competitive data augmentation techniques such as cutout and mixup. RICAP achieves a new state-of-the-art test error of 2.19% on CIFAR-10. We also confirmed that deep CNNs with RICAP achieve better results on classification tasks using CIFAR-100 and ImageNet, an image-caption retrieval task using Microsoft COCO, and other computer vision tasks.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“However, inserting a dropout between the convolutional layers (as opposed to within the residual block) made the learning more effective in WideResNet [121, 122].”
From Review of deep learning: concepts, CNN architectures, challenges, applications, future directions (2021) · cites this paper as [122]
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474 · cited 5×
“The backbone model was the 50 layer ResNet [11] pre-trained on ImageNet.”
BackgroundMethodInfluentialVery Deep Convolutional Networks for Large-Scale Image Recognition
Simonyan, K., & Zisserman, A. (2014) · arXiv · doi:10.48550/arxiv.1409.1556 · Cited by 114,148
ImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155 · cited 4×
“AlexNet also performed principal component analysis (PCA) on a set of RGB values to alter the intensities of the RGB channels for evaluation on the ImageNet dataset [24].”
MethodInfluentialAdam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946 · cited 2×
“The weight parameters were updated using the Adam optimizer [42] over 100 epochs with a mini-batch size of 16.”
MethodYou Only Look Once: Unified, Real-Time Object Detection
Redmon, J., Divvala, S., Girshick, R., & Farhadi, A. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.91 · Cited by 53,803 · cited 1×
“YOLO [46] is an end-to-end model achieving faster inference than previous methods.”
MethodDensely Connected Convolutional Networks
Huang, G., Liu, Z., Van Der Maaten, L., & Weinberger, K. Q. (2017) · 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2017.243 · Cited by 48,566
Going deeper with convolutions
Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., & Rabinovich, A. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298594 · Cited by 48,562
Microsoft COCO: Common Objects in Context
Lin, T.-Y., Maire, M., Belongie, S., Hays, J., Perona, P., Ramanan, D., Dollár, P., & Zitnick, C. L. (2014) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-10602-1_48 · Cited by 45,460
ImageNet Large Scale Visual Recognition Challenge
Russakovsky, O., Deng, J., Su, H., Krause, J., Satheesh, S., Ma, S., Huang, Z., Karpathy, A., Khosla, A., Bernstein, M., et al. (2015) · International Journal of Computer Vision · doi:10.1007/s11263-015-0816-y · Cited by 39,380 · cited 16×
“We apply RICAP to existing deep CNNs and evaluate them on the classification tasks using the CIFAR-10, CIFAR-100 [8], and ImageNet [24] datasets in Sections IV-A, IV-B and IV-C.”
MethodInfluentialRethinking the Inception Architecture for Computer Vision
Szegedy, C., Vanhoucke, V., Ioffe, S., Shlens, J., & Wojna, Z. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.308 · Cited by 32,138
Cited by
A survey on Image Data Augmentation for Deep Learning
Shorten, C., & Khoshgoftaar, T. M. (2019) · Journal of Big Data · doi:10.1186/s40537-019-0197-0 · Cited by 13,203
Review of deep learning: concepts, CNN architectures, challenges, applications, future directions
Alzubaidi, L., Zhang, J., Humaidi, A. J., Al-Dujaili, A., Duan, Y., Al-Shamma, O., Santamaría, J., Fadhel, M. A., Al-Amidie, M., & Farhan, L. (2021) · Journal of Big Data · doi:10.1186/s40537-021-00444-8 · Cited by 7,906
A review of medical image data augmentation techniques for deep learning applications
Chlap, P., Min, H., Vandenberg, N., Dowling, J., Holloway, L., & Haworth, A. (2021) · Journal of Medical Imaging and Radiation Oncology · doi:10.1111/1754-9485.13261 · Cited by 956
“also utilised Random Image Cropping And Patching (RICAP), where, rather than blending the images, four images are randomly selected, cropped and then patched together.(89,110) Unlike other augmentation techniques, basic augmentation methods do not aim to produce ‘realistic’ images.”
MethodData augmentation: A comprehensive survey of modern approaches
Mumuni, A., & Mumuni, F. (2022) · Array · doi:10.1016/j.array.2022.100258 · Cited by 891
AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty
Hendrycks, D., Mu, N., Cubuk, E. D., Zoph, B., Gilmer, J., & Lakshminarayanan, B. (2019) · arXiv · doi:10.48550/arxiv.1912.02781
“Rather than occluding a portion of an image, CutMix replaces a portion of an image with a portion of a different image (Yun et al., 2019; Takahashi et al., 2019).”
BackgroundA Systematic Review on Data Scarcity Problem in Deep Learning: Solution and Applications
Bansal, M. A., Sharma, D. R., & Kathuria, D. M. (2022) · ACM Computing Surveys · doi:10.1145/3502287 · Cited by 306
“[Takahashi et al. 2019] proposed another approach for mixing images by randomly cropping images and concatenating the cropped images to form new images.”
BackgroundMethodInfluentialSingle Model Deep Learning on Imbalanced Small Datasets for Skin Lesion Classification
Yao, P., Shen, S., Xu, M., Liu, P., Zhang, F., Xing, J., Shao, P., Kaffenberger, B., & Xu, R. X. (2022) · IEEE Transactions on Medical Imaging · doi:10.1109/tmi.2021.3136682 · Cited by 227
“On the other hand, the random cropping expanding the scale of the dataset is widely used in the training of the DCNNs, and it can also greatly enhance the spatial robustness of the model [47].”
BackgroundEnhancing Performance of Deep Learning Models with different Data Augmentation Techniques: A Survey
Khosla, C., & Saini, B. S. (2020) · 2020 International Conference on Intelligent Engineering and Management (ICIEM) · doi:10.1109/iciem48762.2020.9160048 · Cited by 189
“The random cropping and translation are different in the sense that cropping will decrease the size of the image while translations preserve its spacial dimensions [12] [18].”
BackgroundManno, D., Cipriani, G., Ciulla, G., Dio, V. D., Guarino, S., & Brano, V. L. (2021) · Energy Conversion and Management · doi:10.1016/j.enconman.2021.114315 · Cited by 171
Su, D., Kong, H., Qiao, Y., & Sukkarieh, S. (2021) · Computers and Electronics in Agriculture · doi:10.1016/j.compag.2021.106418 · Cited by 158
A Review of Data Augmentation Methods of Remote Sensing Image Target Recognition
Hao, X., Liu, L., Yang, R., Yin, L., Zhang, L., & Li, X. (2023) · Remote Sensing · doi:10.3390/rs15030827 · Cited by 178
Nishio, M., 西尾, 瑞., ニシオ, ミ., Noguchi, S., Matsuo, H., Murakami, T., 村上, 卓., & ムラカミ, タ. (2020) · Institutional Repositories DataBase (IRDB) · Cited by 139
Weakly Labeled Data Augmentation for Deep Learning: A Study on COVID-19 Detection in Chest X-Rays
Rajaraman, S., & Antani, S. (2020) · Diagnostics · doi:10.3390/diagnostics10060358 · Cited by 153
Nishio, M., Noguchi, S., Matsuo, H., & Murakami, T. (2020) · Scientific Reports · doi:10.1038/s41598-020-74539-2 · Cited by 160
A Survey of Deep Learning for Scientific DiscoveryPreprint
Raghu, M., & Schmidt, E. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2003.11755 · Cited by 274
“Inspired by these simple methods are two very successful image augmentation strategies, cutout [44], which removes a patch from the input image, and RICAP [222], which combines patches from four different input image to create a new image (with new label a combination of the original labels.)”
MethodData Augmentation and Transfer Learning for Brain Tumor Detection in Magnetic Resonance Imaging
Anaya-Isaza, A., & Mera-Jiménez, L. (2022) · IEEE Access · doi:10.1109/access.2022.3154061 · Cited by 152
“Cropping [61].”
BackgroundGraph-Embedded Convolutional Neural Network for Image-Based EEG Emotion Recognition
Song, T., Zheng, W., Liu, S., Zong, Y., Cui, Z., & Li, Y. (2021) · IEEE Transactions on Emerging Topics in Computing · doi:10.1109/tetc.2021.3087174 · Cited by 145
“For the biased electrode coordinates, the inspiration is random cropping [18] and random erasing data augmentation [19] in image classi fi cation, which can randomly change spatial positions or erase some regions of the input image.”
BackgroundLi, F., Zhou, H., Wang, Z., & Wu, X. (2020) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2020.2999365 · Cited by 135
Shen, Y., Zhu, S., Yang, T., Chen, C., Pan, D., Chen, J., Xiao, L., & Du, Q. (2021) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2021.3080580 · Cited by 170
“technique to artificially increase the size of dataset in computer vision tasks [49], [21], [50].”
MethodMahmood, T., Li, J., Pei, Y., Akhtar, F., Rehman, M. U., & Wasti, S. H. (2022) · PLoS ONE · doi:10.1371/journal.pone.0263126 · Cited by 118
“Therefore, instead of screening the whole mammogram, the suspicious region containing the intensity changes is processed Cropping inflates datasets by integrating the spatial dimensions of mammography images without affecting their labels [44].”
MethodSnapMix: Semantically Proportional Mixing for Augmenting Fine-grained Data
Huang, S., Wang, X., & Tao, D. (2021) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v35i2.16255 · Cited by 164
PixMix: Dreamlike Pictures Comprehensively Improve Safety Measures
Hendrycks, D., Zou, A., Mazeika, M., Tang, L., Li, B., Song, D. X., & Steinhardt, J. (2022) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52688.2022.01628 · Cited by 214
“Various augmentation techniques for image data have been proposed, including Cutout [8, 55], Mixup [47, 54], CutMix [45, 52], and AutoAug-ment [6, 50].”
FMix: Enhancing Mixed Sample Data AugmentationPreprint
Harris, E., Marcu, A., Painter, M., Niranjan, M., Prügel‐Bennett, A., & Hare, J. S. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2002.12047 · Cited by 242
“Recently, a plethora of approaches to Mixed Sample Data Augmentation (MSDA) have been proposed which obtain state-of-the-art results, particularly in classification tasks (Chawla et al., 2002; Zhang et al., 2017; Tokozume et al., 2017; 2018; Inoue, 2018; Yun et al., 2019; Takahashi et al., 2019; Summers & Dinneen, 201…”
BackgroundStudent Network Learning via Evolutionary Knowledge Distillation
Zhang, K., Zhang, C., Li, S., Zeng, D., & Ge, S. (2021) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2021.3090902 · Cited by 123
A Comprehensive Survey on Data Augmentation
Wang, Z., Wang, P., Liu, K., Wang, P., Fu, Y., Lu, C.-T., Aggarwal, C. C., Pei, J., & Zhou, Y. (2025) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2025.3622600 · Cited by 275
“RICAP [147] randomly crops four images and patches them to construct a new image.”
Petrographic microfacies classification with deep convolutional neural networks
de Lima, R. P., Duarte, D., Nicholson, C., Slatt, R. M., & Marfurt, K. J. (2020) · Computers & Geosciences · doi:10.1016/j.cageo.2020.104481 · Cited by 89
“Data augmentation increases the diversity of training samples thereby reducing overfitting (Cires¸an et al., 2011; Takahashi et al., 2018).”
BackgroundM-FFN: multi-scale feature fusion network for image captioning
Prudviraj, J., Vishnu, C., & Mohan, C. K. (2022) · Applied Intelligence · doi:10.1007/s10489-022-03463-x · Cited by 79
“Further, a random image cropping and patching technique is introduced in [38] to augment new images to the dataset for image captioning task.”
MethodAn intelligent system for complex violence pattern analysis and detection
Ullah, F. U. M., Obaidat, M. S., Muhammad, K., Ullah, A., Baik, S. W., Cuzzolin, F., Rodrigues, J. J. P. C., & de Albuquerque, V. H. C. (2021) · International Journal of Intelligent Systems · doi:10.1002/int.22537 · Cited by 87
“36 Specifically, we apply the image augmentation (IMGAUG) technique 1 to the video frames before training.”
MethodMohammed, F. A., Tune, K. K., Assefa, B. G., Jett, M., & Muhie, S. Y. (2024) · Machine Learning and Knowledge Extraction · doi:10.3390/make6010033 · Cited by 108
“…change the hue, saturation and brightness of the images to introduce variations in color), random cropping (a portion of the image, forcing the model to focus on different regions) and shearing (apply shearing transformations to the images, simulating changes in the viewing angle) [116–118].”
BackgroundShang, J., Xu, D., Qiu, H., Gao, L., Jiang, C., & Yi, P. (2024) · Journal of Manufacturing Systems · doi:10.1016/j.jmsy.2024.02.011 · Cited by 75
Food Image Classification with Improved MobileNet Architecture and Data Augmentation
Phiphitphatphaisit, S., & Surinta, O. (2020) · International Conference on Information Science and Systems (ICISS) · doi:10.1145/3388176.3388179 · Cited by 72
“Furthermore, random cropping is applied [20].”
Benchmarking the Robustness of Semantic Segmentation Models with Respect to Common Corruptions
Kamann, C., & Rother, C. (2020) · International Journal of Computer Vision · doi:10.1007/s11263-020-01383-2 · Cited by 92
Small Object Detection Method Based on Global Multi-Level Perception and Dynamic Region Aggregation
Zhu, Z., Zheng, R., Qi, G., Li, S., Li, Y., & Gao, X. (2024) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2024.3402097 · Cited by 80
“According to the strategies adopted, main-stream small object detection methods can be roughly divided into three categories: data augmentation-based methods [7], [8] multi-scale learning-based methods [9], [10] and context learning-based methods [11], [12].”
MethodFarda, N. A., Lai, J., Wang, J., Lee, P., Liu, J.-W., & Hsieh, I.-H. (2020) · Injury · doi:10.1016/j.injury.2020.09.010 · Cited by 73
TokenMix: Rethinking Image Mixing for Data Augmentation in Vision Transformers
Liu, J., Liu, B., Zhou, H., Li, H., & Liu, Y. (2022) · Lecture notes in computer science · doi:10.1007/978-3-031-19809-0_26 · Cited by 103
Chen, Y., Li, D., Zhang, X., Jin, J., & Shen, Y. (2020) · Medical Image Analysis · doi:10.1016/j.media.2020.101819 · Cited by 67
Li, L., Wu, F., Wang, S., Luo, X., Martín-Isla, C., Zhai, S., Zhang, J., Liu, Y., Zhang, Z., Ankenbrand, M. J., et al. (2023) · Medical Image Analysis · doi:10.1016/j.media.2023.102808 · Cited by 98
Fine-Grained Recognition With Learnable Semantic Data Augmentation
Pu, Y., Han, Y., Wang, Y., Feng, J., Deng, C., & Huang, G. M. (2024) · IEEE Transactions on Image Processing · doi:10.1109/tip.2024.3364500 · Cited by 83
“[69], [70], [71], [72], [73], [74], [75], [76], [77], [78], [79], [80], [81], [82].”
Chen, T., Sun, L., Yu, H., Wang, W., Qi, L., Zhang, P., & Zeng, P. (2021) · Applied Geochemistry · doi:10.1016/j.apgeochem.2021.105135 · Cited by 62
A comparative study for glioma classification using deep convolutional neural networks
Özcan, H., Emiroğlu, B. G., Sabuncuoğlu, H., Özdoğan, S., Soyer, A., & Saygı, T. (2021) · Mathematical Biosciences & Engineering · doi:10.3934/mbe.2021080 · Cited by 60
“This helped expand the data, without actual introducing new data [35,36].”
BackgroundDetection of Breast Cancer Using Infrared Thermography and Deep Neural Networks
Fernández-Ovies, F. J., Alférez, S., deAndrés‐Galiana, E. J., Cernea, A., Fernández‐Muñiz, Z., & Fernández‐Martínez, J. L. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-17935-9_46 · Cited by 54
YOLOv10-pose and YOLOv9-pose: Real-time strawberry stalk pose detection models
Meng, Z., Du, X., Sapkota, R., Ma, Z., & Cheng, H. (2024) · Computers in Industry · doi:10.1016/j.compind.2024.104231 · Cited by 67
Data Augmentation for Object Detection via Progressive and Selective Instance-SwitchingPreprint
Wang, H., Wang, Q., Yang, F., Zhang, W., & Zuo, W. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1906.00358 · Cited by 72
“Albeit many data augmentation methods are proposed for deep CNNs in image classification task [9, 10, 28, 45, 49], it is not clear whether they are suitable for object detection or not.”
BackgroundFeature pyramid network with self-guided attention refinement module for crack segmentation
Ong, J. C., Lau, S. L. H., Ismadi, M., & Wang, X. (2022) · Structural Health Monitoring · doi:10.1177/14759217221089571 · Cited by 47
“52 During training and testing of both datasets, the images are normalized to have a mean of 0 and a standard”
MethodPhysical Knowledge-Enhanced Deep Neural Network for Sea Surface Temperature Prediction
Meng, Y., Gao, F., Rigall, E., Dong, R., Dong, J.-Y., & Du, Q. (2023) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2023.3257039 · Cited by 52
“The resampling technique lays particular emphasis on new instance composition, such as image mixup [46]–[48], feature space enhancement [49], [50] and generative adversarial network (GAN) [16].”
MethodShokouhmand, A., & Tavassolian, N. (2022) · IEEE Transactions on Biomedical Engineering · doi:10.1109/tbme.2022.3189617 · Cited by 54
“Hence, the Beta distribution whose superiority for data augmentation over the uniform distribution has been demonstrated [77], was utilized to specify the cropping width,”
MethodA Comprehensive Review of Advanced Deep Learning Approaches for Food Freshness Detection
Singh, R., Nickhil, C., R.Nisha, Upendar, K., Bhukya, J., & Deka, S. C. (2024) · Food Engineering Reviews · doi:10.1007/s12393-024-09385-3 · Cited by 59
MixMo: Mixing Multiple Inputs for Multiple Outputs via Deep Subnetworks
Ramé, A., Sun, R., & Cord, M. (2021) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv48922.2021.00086 · Cited by 76
“Such non-linear binary masking improves generalization [68, 70] by increasing dataset: it creates new images with usually disjoint patches [29].”
Sun, X., Jia, X., Liang, Y., Wang, M., Wang, M., & Chi, X. (2022) · IEEE Access · doi:10.1109/access.2022.3204683 · Cited by 50
Data Augmentation in Defect Detection of Sanitary Ceramics in Small and Non-i.i.d Datasets
Ren, X., Lin, W., Yang, X., Yu, X., & Gao, H. (2022) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2022.3152245 · Cited by 49
“A weighted sum of two images is not commonsensical, but it worked in [13].”
BackgroundBadawi, D., & Gildin, E. (2024) · Computers & Geosciences · doi:10.1016/j.cageo.2024.105826 · Cited by 62
Diagnosis of Autism Spectrum Disorder Using Convolutional Neural Networks
Hendr, A. M., Ozgunalp, U., & Kaya, M. E. (2023) · Electronics · doi:10.3390/electronics12030612 · Cited by 55
A corn canopy organs detection method based on improved DBi-YOLOv8 network
Guan, H., Deng, H., Ma, X., Zhang, T., Zhang, Y., Zhu, T., Zhou, H., Gu, Z., & Lu, Y. (2024) · European Journal of Agronomy · doi:10.1016/j.eja.2023.127076 · Cited by 43
Nishio, M., Noguchi, S., & Fujimoto, K. (2020) · Applied Sciences · doi:10.3390/app10103360 · Cited by 45
“Recently proposed methods of data augmentation, such as mixup [16] and random image cropping and patching (RICAP) [17], were not evaluated.”
Crossover based technique for data augmentation
Raj, R., Mathew, J., Kannath, S. K., & Rajan, J. (2022) · Computer Methods and Programs in Biomedicine · doi:10.1016/j.cmpb.2022.106716 · Cited by 40
Kurata, Y., Nishio, M., Moribata, Y., Kido, A., Himoto, Y., Otani, S., Fujimoto, K., Yakami, M., Minamiguchi, S., Mandai, M., et al. (2021) · Scientific Reports · doi:10.1038/s41598-021-93792-7 · Cited by 40
YOLO-MSFG: Toward Real-Time Detection of Concealed Objects in Passive Terahertz Images
Xu, F., Huang, X., Wu, Q., Zhang, X., Shang, Z., & Zhang, Y.-J. (2021) · IEEE Sensors Journal · doi:10.1109/jsen.2021.3127686 · Cited by 46
Mahmud, S., Hossain, M. S., Chowdhury, M. E. H., & Reaz, M. B. I. (2022) · Neural Computing and Applications · doi:10.1007/s00521-022-08111-6 · Cited by 47
“This approach is similar to patching [80] for images.”
MethodKonovalenko, І. V., Maruschak, P. O., Brevus, V., & Prentkovskis, O. (2021) · Metals · doi:10.3390/met11040549 · Cited by 40
“Therefore, we used the random crop technique to train the neural network classifier [30].”
MethodWeed Identification in Maize Fields Based on Improved Swin-Unet
Zhang, J., Gong, J., Zhang, Y., Mostafa, K., & Yuan, G. (2023) · Agronomy · doi:10.3390/agronomy13071846 · Cited by 37
Enhancing Mixture-of-Experts by Leveraging Attention for Fine-Grained Recognition
Zhang, L., Huang, S., & Liu, W. (2021) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2021.3117064 · Cited by 49
“by concatenating image pieces after randomly cropped from different images [54], which also shows effectiveness on different tasks.”
BackgroundArtificial Intelligence–Assisted Classification of Gliomas Using Whole Slide Images
Jose, L., Liu, S., Russo, C., Cong, C., Song, Y., Rodriguez, M. L., & Ieva, A. D. (2022) · Archives of Pathology & Laboratory Medicine · doi:10.5858/arpa.2021-0518-oa · Cited by 37
Brilhador, A., Gutoski, M., Hattori, L. T., Inácio, A. D. S., Lazzaretti, A. E., & Lopes, H. S. (2019) · IEEE Latin American Conference on Computational Intelligence (LA-CCI) · doi:10.1109/la-cci47412.2019.9037044 · Cited by 38
“In addition to these transformations, we apply a patch augmentation strategy, which leaded to interesting results in recent works in other contexts [21], [22].”
ResultAuto-FSL: Searching the Attribute Consistent Network for Few-Shot Learning
Zhang, L., Wang, S., Chang, X., Liu, J., Ge, Z., & Zheng, Q. (2021) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2021.3076523 · Cited by 35
Automated Defect Recognition of Castings Defects Using Neural Networks
García-Pérez, A., Silva, M. J. G., & de la Escalera, A. (2021) · Journal of Nondestructive Evaluation · doi:10.1007/s10921-021-00842-1 · Cited by 41
“In other occasions, random crop of images may lead to remove key features within the label area in such a way that a not safe transformation takes place [33,34], which may be the case for experiment 25 described in Table 3.”
BackgroundAttentive Feature Augmentation for Long-Tailed Visual Recognition
Wang, W., Zhao, Z., Wang, P., Su, F., & Meng, H. (2022) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2022.3161427 · Cited by 37
“transformation strategies [31], [32] and nonlinear generation modules to enhance the diversity of training samples,”
BackgroundA Survey on Deep Learning in COVID-19 Diagnosis
Han, X., Hu, Z., Wang, S., & Zhang, Y. (2022) · Journal of Imaging · doi:10.3390/jimaging9010001 · Cited by 38
“[119] generate a new training image by mixing four images obtained Takahashi et al. [119] generate a new training image by mixing four images obtained through random image cropping and patching (RICAP). through random image cropping and patching (RICAP).”
MethodA Coarse-to-Fine Pseudo-Labeling (C2FPL) Framework for Unsupervised Video Anomaly Detection
Al-lahham, A., Tastan, N., Zaheer, M. Z., & Nandakumar, K. (2024) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv57701.2024.00665 · Cited by 34
“Following [21], we also apply 10-crop augmentation to the I3D features.”
MethodAugmentasi Data Pengenalan Citra Mobil Menggunakan Pendekatan Random Crop, Rotate, dan Mixup
Sanjaya, J., & Ayub, M. (2020) · Jurnal Teknik Informatika dan Sistem Informasi · doi:10.28932/jutisi.v6i2.2688 · Cited by 33
Intra-class Part Swapping for Fine-Grained Image Classification
Zhang, L., Huang, S., & Liu, W. (2021) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv48630.2021.00325 · Cited by 40
“More recent works are RI-CAP [27] and CutMix [38].”
MethodAn Overview of Deep Neural Networks for Few-Shot Learning
Zhao, J., Kong, L., & Lv, J. (2024) · Big Data Mining and Analytics · doi:10.26599/bdma.2024.9020049 · Cited by 36
Deep Learning-Based Plant-Image Classification Using a Small Training Dataset
Batchuluun, G., Nam, S. H., & Park, K. R. (2022) · Mathematics · doi:10.3390/math10173091 · Cited by 36
BD-SKUNet: Selective-Kernel UNets for Building Damage Assessment in High-Resolution Satellite Images
Ahmadi, S. A., Mohammadzadeh, A., Yokoya, N., & Ghorbanian, A. (2023) · Remote Sensing · doi:10.3390/rs16010182 · Cited by 29
“Preprocessing techniques such as data augmentation have been widely utilized to increase the size of data and robustness of the model [58].”
An Improved Aggregated-Mosaic Method for the Sparse Object Detection of Remote Sensing Imagery
Zhao, B., Wu, Y., Guan, X., Gao, L., & Zhang, B. (2021) · Remote Sensing · doi:10.3390/rs13132602 · Cited by 38
“Random Image Cropping and Patching (RICAP) [43] randomly crops four images and patches the images and labels to create a new training sample.”
BackgroundCaicho, J., Chuya-Sumba, C., Jara, N., Salum, G. M., Tirado-Espín, A., Villalba-Meneses, F., Alvarado-Cando, O., Cadena-Morejón, C., & Almeida-Galárraga, D. A. (2022) · Communications in computer and information science · doi:10.1007/978-3-030-99170-8_19 · Cited by 44
Rejusha, T. R., & S, V. K. K. (2021) · International Conference on Communication, Control and Information Sciences (ICCISc) · doi:10.1109/iccisc52257.2021.9484902 · Cited by 35
OccluMix: Towards De-Occlusion Virtual Try-on by Semantically-Guided Mixup
Yang, Z., Chen, J., Shi, Y.-K., Li, H., Chen, T., & Lin, L. (2023) · IEEE transactions on multimedia · doi:10.1109/tmm.2023.3234399 · Cited by 30
“( i.e. , Inherent-and Acquired-Occlusion).”
Key Role Guided Transformer for Group Activity Recognition
Pei, D., Huang, D., Kong, L., & Wang, Y. (2023) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2023.3283282 · Cited by 34
“For example, Mixup variants [73], [74], [75], [76] obtain an augmented image by pixel-wise weighted combination of two global images, and CutMix variants [77], [78], [79], [80], [81] replace removed regions with a patch from another image.”
BackgroundSurvey of Accuracy Prediction on the PlantVillage Dataset using different ML techniques
Pandey, V., Tripathi, U., Singh, V. K., Gaur, Y. S., & Gupta, D. (2023) · EAI Endorsed Transactions on Internet of Things · doi:10.4108/eetiot.4578 · Cited by 30
Data Augmentation for Deep Learning based Cattle Segmentation in Precision Livestock Farming
Qiao, Y., Su, D., Kong, H., Sukkarieh, S., Lomax, S., & Clark, C. (2020) · IEEE International Conference on Automation Science and Engineering (CASE) · doi:10.1109/case48305.2020.9216758 · Cited by 27
“In addition, Takahashi et al. (2019) proposed RICAP, which randomly cropped four images and patched them to create a new training image [25], and validated its applicability in image classification tasks.”
MethodIncreasing the Robustness of Semantic Segmentation Models with Painting-by-Numbers
Kamann, C., & Rother, C. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-58607-2_22 · Cited by 35
“Whereas some work occludes parts of images [68,15], crops, replaces and mixes several images [64,65,60], or applies various (learned) sets of distortions [32,12], other methods augment with artificial noise to increase robustness [24,46,53].”
BackgroundLi, L., Qiao, G., Liu, S., Qing, X., Zhang, H., Mazhar, S., & Niu, F. (2021) · The Journal of the Acoustical Society of America · doi:10.1121/10.0007291 · Cited by 26
Data augmentation with Mobius transformations
Zhou, S., Zhang, J., Jiang, H., Lundh, T., & Ng, A. Y. (2020) · Machine Learning Science and Technology · doi:10.1088/2632-2153/abd615 · Cited by 24
“Comparably, RICAP [26], VH-Mixup and VH-BC+ [27] form composites of several images into one.”
RAHC_GAN: A Data Augmentation Method for Tomato Leaf Disease Recognition
Deng, H., Luo, D., Chang, Z., Li, H., & Yang, X. (2021) · Symmetry · doi:10.3390/sym13091597 · Cited by 29
ECG waveform generation from radar signals: A deep learning perspective
Chowdhury, F. A., Hosain, M. K., Islam, M. S. B., Hossain, M. S., Basak, P., Mahmud, S., Murugappan, M., & Chowdhury, M. E. H. (2024) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2024.108555 · Cited by 24
Constrained Online Cut-Paste for Object Detection
Wang, H., Wang, Q., Zhang, H., Yang, J. J., & Zuo, W. (2020) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2020.3045978 · Cited by 23
Jaihuni, M., Khan, F., Lee, D., Basak, J. K., Bhujel, A., Moon, B. E., Park, J., & Kim, H. T. (2021) · IEEE Access · doi:10.1109/access.2021.3059314 · Cited by 23
“Consequently, these images were subdivided into 55 × 55 pixel small images, after which the data augmentation process was applied by rotating them 90 ; therefore, the total quantity of the images was doubled, which would be influential in training the deep learning model [36].”
MethodModel-Agnostic Augmentation for Accurate Graph Classification
Yoo, J., Shim, S., & Kang, U. (2022) · International World Wide Web Conference (WWW) · doi:10.1145/3485447.3512175 · Cited by 44
“On the other hand, patch-based mixing approaches that replace a random patch of an image to that of another image [14, 32, 50] give us a motivation to propose SubMix, which replaces an induced subgraph of a graph, instead of an image patch.”
BackgroundAttention-Based Knowledge Distillation in Scene Recognition: The Impact of a DCT-Driven Loss
López-Cifuentes, A., Escudero-Viñolo, M., Bescós, J., & SanMiguel, J. C. (2023) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2023.3250031 · Cited by 24
Self-paced data augmentation for training neural networks
Takase, T., Karakida, R., & Asoh, H. (2021) · Neurocomputing · doi:10.1016/j.neucom.2021.02.080 · Cited by 23
Stride Random Erasing Augmentation
Kumar, T., Brennan, R., & Bendechache, M. (2022) · International Conference on Artificial Intelligence, Soft Computing and Applications (AIAPP) · doi:10.5121/csit.2022.120201 · Cited by 22
“The two main techniques used for model generalization are: regularization [11, 12, 21, 22, 23] and data augmentation [13, 15, 14, 16, 17, 18, 20].”
MethodLightAnomalyNet: A Lightweight Framework for Efficient Abnormal Behavior Detection
Mehmood, A. (2021) · Sensors · doi:10.3390/s21248501 · Cited by 27
“Further ore, data aug entation is a standard technique to deal ith class i balance and the lack of data for training [63].”
MethodWang, J., Xi, R., Cai, T., Lu, H., Zhu, R., Zheng, B., & Chen, H. (2022) · Advanced Optical Materials · doi:10.1002/adom.202200178 · Cited by 21
Visual Embedding Augmentation in Fourier Domain for Deep Metric Learning
Wang, Z., Gao, Z., Wang, G., Yang, Y., & Shen, H. T. (2023) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2023.3260082 · Cited by 25
“cially random spatial image augmentation [56], [57] including cropping, rotation, and dropping.”
BackgroundPredicting fatigue crack growth metrics from fractographs: Towards fractography by computer vision
Jones, K., Musinski, W. D., Pilchak, A. L., John, R., Shade, P. A., Rollett, A. D., & Holm, E. A. (2023) · International Journal of Fatigue · doi:10.1016/j.ijfatigue.2023.107915 · Cited by 24
RSMDA: Random Slices Mixing Data Augmentation
Kumar, T., Mileo, A., Brennan, R., & Bendechache, M. (2023) · Applied Sciences · doi:10.3390/app13031711 · Cited by 24
Cheng, H., Huang, S., Cai, L., Xu, Y., Wang, R., & Zhang, Y. (2025) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2025.3528625 · Cited by 25
“As an effective solution for enhancing the feature learning of the model, data augmentation has been widely studied in natural scenarios [18], [19], [20], [21].”
Principled Ultrasound Data Augmentation for Classification of Standard Planes
Lee, L. H., Gao, Y., & Noble, J. A. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-78191-0_56 · Cited by 21
Grad-CAM++ is Equivalent to Grad-CAM With Positive Gradients
Lerma, M., & Lucas, M. (2022) · International Machine Vision and Image Processing Conference (IMVIP) · doi:10.56541/awjv6348 · Cited by 28
Dynamic Mosaic algorithm for data augmentation
Li, Y., Cheng, R., Zhang, C., Chen, M., Liang, H., & Wang, Z. (2023) · Mathematical Biosciences & Engineering · doi:10.3934/mbe.2023311 · Cited by 25
Metadata
- Authors
- Takahashi, R., Matsubara, T., & Uehara, K.
- Affiliations
- Kobe University (JP)
- Venue
- IEEE Transactions on Circuits and Systems for Video Technology
- Published
- 2020-09
- Volume
- 30
- Issue
- 9
- Pages
- 2917-2931
- ISSN
- 1051-8215 (ISSN-L)
- 1558-2205
- OpenAlex
- W2900595477
- Cited by
- 477
- References
- 54
- Citations by year
- 2026: 30 · 2025: 68 · 2024: 73 · 2023: 89 · 2022: 100 · 2021: 72 · 2020: 34
- FWCI
- 24.17
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Licence
- Locations
- IEEE Transactions on Circuits and Systems for Video Technology (publishedVersion)
- arXiv (Cornell University) (acceptedVersion)
- Institutional Repositories DataBase (IRDB) (acceptedVersion)
- Kobe University Repository Kernel (Kobe University) (submittedVersion, other-oa)
- arXiv (Cornell University)
- Funders
- Strategic Information and Communications R&D Promotion Programme of Ministry of Internal Affairs and Communications, Japan (172107101)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning and Data Classification
- Keywords
- data augmentation, overfitting, convolutional neural networks, soft labels, cutouts, mixup, image classification
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.