Data Augmentation Using Random Image Cropping and Patching for Deep CNNs
Open accessThe DOI matches its Crossref record; 54 of the 54 references it cites were found in a registry or an index.
Abstract
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
Items 201–300 of 476
Box Supervised Video Segmentation Proposal Network
Hannan, T., Koner, R., Kobold, J., & Schubert, M. (2022) · International Machine Vision and Image Processing Conference (IMVIP) · doi:10.56541/azwk8552 · Cited by 11
Zhang, H., Cao, J., Zheng, D., Yao, X., & Ling, B. W. (2022) · Sensors · doi:10.3390/s22218127 · Cited by 7
“, geometric and color transformation [18], mixing image [19,20], random erasing, and so on.”
BackgroundPham, L. D., Tran, K., Ngo, D. T., Tang, H., Phan, S., & Schindler, A. (2022) · ACM International Conference on Multimedia in Asia (MMAsia) · doi:10.1145/3551626.3564962 · Cited by 14
Using Deep Neural Networks to Evaluate Leafminer Fly Attacks on Tomato Plants
Crispi, G. M., Valente, D. S. M., de Queiroz, D. M., Momin, A., Filho, E. I. F., & Picanço, M. C. (2023) · AgriEngineering · doi:10.3390/agriengineering5010018 · Cited by 7
An augmented mammogram image dataset and its performance analysis for various classification models
Sharma, B. P., & Purwar, R. K. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-023-14566-z · Cited by 7
Self-Supervised Learning of Audio Representations using Angular Contrastive Loss
Wang, S., Tripathy, S., & Mesaros, A. (2023) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49357.2023.10094706 · Cited by 10
“Then we apply data augmentation techniques such as random sized cropping [15], specAugment [16], and Gaussian blurring to get x̃i and x̃j .”
MethodJabbari, H., & Bigdeli, N. (2023) · Neural Computing and Applications · doi:10.1007/s00521-023-08742-3 · Cited by 8
“In terms of data augmentation, image augmentation is the most well-known type, involving making modified versions of images that still belong to the same class and statistical distribution as the original images [49].”
BackgroundGeneralization Gap in Data Augmentation: Insights from Illumination
Xiao, J., Guo, W., Liu, J., & Li, M. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-78444-6_13 · Cited by 14
24th Irish Machine Vision and Image Processing Conference
(2022) · doi:10.56541/iprcs_imvip2022 · Cited by 7
A comparative analysis of different augmentations for brain images
Bajaj, S., Bala, M., & Angurala, M. (2024) · Medical & Biological Engineering & Computing · doi:10.1007/s11517-024-03127-7 · Cited by 6
“Ryo T. and Takashi M. [29] proposed a data augmentation method that makes use of patches and random image cropping to improve deep CNN training.”
Enhancing body fat prediction with WGAN-GP data augmentation and XGBoost algorithm
Wang, X., & Chang, S. (2025) · Science Progress · doi:10.1177/00368504251366850 · Cited by 6
“45,46 By enriching the training data with high-fi delity synthetic samples, the model is compelled to learn more robust and generalizable decision boundaries rather than memorizing the noise inherent in a small sample.”
Liu, D., Kababji, S. E., Mitsakakis, N., Pilgram, L., Walters, T. D., Clemons, M. J., Pond, G. R., El-Hussuna, A., & Emam, K. E. (2025) · BMC Medical Informatics and Decision Making · doi:10.1186/s12911-025-03266-3 · Cited by 20
“For example, flipping and cropping is a commonly used imaging augmentation technique to increase the diversity and expand the available data [7, 29, 30].”
Using Genetic Algorithm to Augment Test Data for Penalty Prediction
Chun-yan, X., Wang, X., Yan, Z., & Hao, Y. (2020) · International Journal of Performability Engineering · doi:10.23940/ijpe.20.07.p10.10781086 · Cited by 6
“In the image field, Ryo et al. [15] proposed an image preprocessing method based on random clipping and patching, which randomly clipped four images and patched them to create a new training image.”
MethodGildea, K., Mercadal-Baudart, C., Blythman, R., Smolić, A., & Simms, C. K. (2022) · International Machine Vision and Image Processing Conference (IMVIP) · doi:10.56541/qtuv2945 · Cited by 8
Rainwater-Removal Image Conversion Learning with Training Pair Augmentation
Han, Y.-K., Jung, S., Kwon, H., & Lee, S.-H. (2023) · Entropy · doi:10.3390/e25010118 · Cited by 6
“In addition, Takahashi obtained a training dataset using alpha mixing [25].”
MethodSelf-Supervised Representation Learning with Meta Comprehensive Regularization
Guo, H., Ba, Y., Hu, J., Si, L., Qiang, W., & Shi, L. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i3.27966 · Cited by 11
“In practice, data augmentation generates two augmented views of the same image by applying random strategies, such as Cutout (DeVries and Taylor 2017), Coloring (Zhang, Isola, and Efros 2016), Random Cropping (Takahashi, Matsubara, and Uehara 2019), etc.”
MethodBlend AutoAugment: Automatic Data Augmentation for Image Classification Using Linear Blending
Im, J., Kasahara, J. Y. L., Maruyama, H., Asama, H., & Yamashita, A. (2024) · IEEE Access · doi:10.1109/access.2024.3401167 · Cited by 8
“Additionally, there is a method [7] that improves performance by providing eight different ways to mix two images and a method [8] that synthesizes a randomly cropped patch from four images into a single image.”
MethodHe, H., Huan, X., Wang, J., Luo, Y., Hu, H., & An, J. (2024) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2024.3459944 · Cited by 8
Chen, H., Alfred, M. C., Brown, A. D., Atinga, A., & Cohen, E. (2024) · JMIR Formative Research · doi:10.2196/59045 · Cited by 8
“Additionally, we used data augmentation techniques, including random horizontal flip, random affine, and random crop [46-48].”
Kang, W., Jung, I.-T., Lee, D., & Hong, J.-H. (2021) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3411764.3445457 · Cited by 6
Combining Magnification and Measurement for Non-Contact Cardiac Monitoring
Nowara, E. M., McDuff, D., & Veeraraghavan, A. (2021) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw53098.2021.00422 · Cited by 9
“Finally, multiple images can be combined together by cropping and rearranging patches together [44] or by adding and averaging pixel intensities from several images [45].”
BackgroundIsotonic Data Augmentation for Knowledge Distillation
Cui, W., & Yan, S. (2021) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2021/319 · Cited by 9
Su, T., Chen, J.-J., Chen, W.-S., Chang, Y., & Lu, H. H. (2022) · Journal of the Chinese Medical Association · doi:10.1097/jcma.0000000000000833 · Cited by 6
“14,15 We employed object detection to determine the myocardial area.”
Dynamic Data Augmentation Based on Imitating Real Scene for Lane Line Detection
Wang, Q., Wang, L., Chi, Y., Shen, T., Song, J., Gao, J., & Shen, S. (2023) · Remote Sensing · doi:10.3390/rs15051212 · Cited by 9
A Robust and Low Complexity Deep Learning Model for Remote Sensing Image Classification
Le, C., Pham, L. D., Nvn, N., Nguyen, T., & Trang, L. H. (2023) · International Conference on Intelligent Information Technology (ICIIT) · doi:10.1145/3591569.3591601 · Cited by 6
Improving Object Detectors by Exploiting Bounding Boxes for Augmentation Design
Devi, S., Thopalli, K., Dayana, R., Malarvezhi, P., & Thiagarajan, J. J. (2023) · IEEE Access · doi:10.1109/access.2023.3320638 · Cited by 6
“State-of-the-art techniques go beyond conventional image manipulations, and adopt interpolation techniques such as Mixup [2] and CutMix [3], or compositional strategies such as AugMix [4], TrivialAug [5], AugMax, ALT [20] etc.”
MethodLeveraging Support Vector Machine for Sports Injury Classification
Surasak, T., Praking, P., & Kitchat, K. (2023) · doi:10.1109/icitee59582.2023.10317743 · Cited by 6
“Data augmentation solves two concerns for researchers: it increases the variety of training samples and minimizes overfitting [22], [23].”
BackgroundAn Efficient Iris Recognition Technique using CNN and Vision Transformer
Latif, S. A., Sidek, K. A., & Hashim, A. H. A. (2023) · Journal of Advanced Research in Applied Sciences and Engineering Technology · doi:10.37934/araset.34.2.235245 · Cited by 6
CNN-based data augmentation for handwritten gurumukhi text recognition
Sareen, B., Ahuja, R., & Singh, A. (2024) · Multimedia Tools and Applications · doi:10.1007/s11042-024-18278-w · Cited by 10
“Since then various techniques have been proposed in the last decade [33, 35].”
Exploring Data Augmentation Effects on A Singular Illumination Distribution Dataset with ColorJitter
Xiao, J., Guo, W., & Liu, J. (2024) · International Conference on Image Processing and Media Computing (ICIPMC) · doi:10.1109/icipmc62364.2024.10586593 · Cited by 7
Speech Emotion Recognition by Combining a Unified First-Order Attention Network With Data Balance
Chen, G., Zhang, S., Tao, X., & Zhao, X. (2020) · IEEE Access · doi:10.1109/access.2020.3038493 · Cited by 7
“However, most speech data augmentation methods [40]–[43]”
Self-Supervised Domain Adaptation for Visual Navigation with Global Map Consistency
Lee, E. S., Kim, J., & Kim, Y. M. (2022) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv51458.2022.00193 · Cited by 8
Weber, I., Bongartz, J., & Roscher, R. (2021) · IEEE International Geoscience and Remote Sensing Symposium (IGARSS) · doi:10.1109/igarss47720.2021.9553162 · Cited by 8
Wang, Y., Lin, M., Hu, Q., Bao, L., Bai, S., & Li, Y. (2025) · Journal of King Saud University - Computer and Information Sciences · doi:10.1007/s44443-025-00055-w · Cited by 10
“The traditional DA methods are typically rule-based approaches (Cubuk et al. 2019; Krell and Kim 2017; Liu et al. 2020; Takahashi et al. 2019; Zhang et al. 2024) that generate new samples through operations such as insertion, substitution, and deletion.”
Generalize Deep Neural Networks With Adaptive Regularization for Classifying
Guo, K., Tao, Z., Zhang, L., Hu, B., & Kui, X. (2023) · IEEE Transactions on Computational Social Systems · doi:10.1109/tcss.2023.3264804 · Cited by 7
“Some data regularization methods [14], [15], [16] improve feature generalization by preserving the original labels of the data and generating different distribution training datasets.”
MethodTao, W., Wang, X., Yan, T., Bi, H., & Yan, J. (2024) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2024.3510948 · Cited by 10
Copy‐paste with self‐adaptation: A self‐adaptive adjustment method based on copy‐paste augmentation
Yu, X., Li, F., Bai, P., Liu, Y., & Chen, Y. (2023) · IET Computer Vision · doi:10.1049/cvi2.12207 · Cited by 6
What Are Effective Labels for Augmented Data? Improving Calibration and Robustness with AutoLabel
Qin, Y., Wang, X., Lakshminarayanan, B., Ed, H., & Beutel, A. (2023) · IEEE Conference on Secure and Trustworthy Machine Learning (SaTML) · doi:10.1109/satml54575.2023.00032 · Cited by 12
Detecting the corneal neovascularisation area using artificial intelligence
Mergen, B., Safi, T., Nadig, M., Bhattrai, G., Daas, L., Alexandersson, J., & Seitz, B. (2023) · British Journal of Ophthalmology · doi:10.1136/bjo-2023-323308 · Cited by 7
“22 We intend to investigate the effect of cropping on our dataset more in-depth in the future.”
BackgroundMAGICROP: A Cropping System for Novice Users to Make Artful Portrait Images
Wang, B., QvZhen, B., Zhang, Z., Li, Y., Yao, W., Wei, S., & Xing, S. (2023) · International Journal of Human-Computer Interaction · doi:10.1080/10447318.2023.2247570 · Cited by 5
“With the development of deep learning, data-driven methods (Takahashi et al., 2020; Zhang et al., 2022) are gradually becoming the main solution.”
Yan, J., Zhang, M., & Xu, Y. (2023) · Sustainability · doi:10.3390/su152015051 · Cited by 6
Archaeological Site Detection: Latest Results from a Deep Learning Based Europe Wide Hillfort Search
Landauer, J., Maddison, S., Fontana, G., & Posluschny, A. (2025) · Journal of Computer Applications in Archaeology · doi:10.5334/jcaa.178 · Cited by 5
CNN-based Approaches for Weed Detection
Tlebaldinova, A., Карменова, М. А., Владимировна, П. Е., & BONDAROVICH, A. (2022) · International Scientific Conference on Computer Science (COMSCI) · doi:10.1109/comsci55378.2022.9912593 · Cited by 5
Automatic Cropping of Handwritten Scanned Documents with Object Detection Algorithm
Aanchal, Nidhi, N., Preeti, P., & Gurpratap (2023) · Procedia Computer Science · doi:10.1016/j.procs.2023.01.151 · Cited by 5
Data augmentation by morphological mixup for solving Raven’s progressive matrices
He, W., Ren, J., & Bai, R. (2023) · The Visual Computer · doi:10.1007/s00371-023-02930-x · Cited by 5
“Specifically, the idea of mixing images can be classified into nonlinearly mixup, i.e. , assembling of sample instances [14, 29] or sub-images [30], and a linear interpolation of image pairs such as alpha-blending [37].”
Multi-scale defect detection method for involute gear tooth flank in the context of remanufacturing
Liu, Y., Yan, C., Huo, D., Wen, J., Lei, H., Chen, X., Lin, C., & Chen, Z. (2025) · Measurement · doi:10.1016/j.measurement.2025.118680 · Cited by 4
Gharib, M. H., Kavehvash, Z., Shabany, M., & Rahemi, Z. (2025) · IEEE Sensors Journal · doi:10.1109/jsen.2025.3601350 · Cited by 4
“Due to the characteristics of MMW images, conventional data augmentation methods, such as affine transformations (rotation, scaling and translation) [21], perspective transformation [21], or data augmentation method of [22], usually do not produce similar results to the real personal images.”
Solving jigsaw puzzles by predicting fragment’s coordinate based on vision transformer
Kim, G., Cho, H., & Nam, H. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.126776 · Cited by 7
NeuBridge: bridging quantized activations and spiking neurons for ANN-SNN conversion
Yang, Y., Liu, J., Yu, C., Yang, C., Wang, G., & Wang, A. (2025) · Neuromorph. Comput. Eng. · doi:10.1088/2634-4386/ade183 · Cited by 6
“On ImageNet, we use RandomCrop [57] and Normalize [58] during training.”
Multimodal Retrieval with Contrastive Pretraining
Alsan, H. F., Yildiz, E., Safdil, E. B., Arslan, F., & Arsan, T. (2021) · International Conference on Innovations in Intelligent Systems and Applications (INISTA) · doi:10.1109/inista52262.2021.9548414 · Cited by 4
Investigating the Generalization of Image Classifiers with Augmented Test Sets
Shorten, C., & Khoshgoftaar, T. M. (2021) · IEEE International Conference on Tools with Artificial Intelligence (ICTAI) · doi:10.1109/ictai52525.2021.00010 · Cited by 4
The Effect of Data Augmentation in Deep Learning Approach for Peripheral Blood Leukocyte Recognition
Nozaka, H., Oda, M., Sasaki, A., Harako, H., Miyazaki, M., Kaga, S., Sakaiya, N., Kudo, K., Kimura, S., Nakano, M., et al. (2022) · Studies in health technology and informatics · doi:10.3233/shti220077 · Cited by 4
A Novel Optimizer in Deep Neural Network for Diabetic Retinopathy Classification
Nanda, P., & Duraipandian, N. (2022) · Computer Systems Science and Engineering · doi:10.32604/csse.2022.024695 · Cited by 5
MSNet: Multi-Resolution Synergistic Networks for Adaptive Inference
Hang, R., Qian, X., & Liu, Q. (2022) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2022.3218891 · Cited by 6
“R ECENT years have witnessed remarkable success in the field of image classification using convolutional neural networks (CNNs) [3], [4], [5].”
BackgroundInteractive Generation of Image Variations for Copy-Paste Data Augmentation
Higuchi, K., Mizuhashi, T., Matulic, F., & Igarashi, T. (2023) · CHI Conference Extended Abstracts on Human Factors in Computing Systems (CHI EA) · doi:10.1145/3544549.3585856 · Cited by 4
“Recent work proposed advanced data augmentation techniques based on image composition, which generate new composite images from multiple source samples [4, 9, 12, 16, 26, 27, 29].”
BackgroundDetecting Objects from No-Object Regions: A Context-Based Data Augmentation for Object Detection
Zhang, J., Han, F., Chun, Y., Liu, K., & Chen, W. (2021) · International Journal of Computational Intelligence Systems · doi:10.2991/ijcis.d.210622.003 · Cited by 4
“Instead only cropping, the paper [21] randomly crops four images and patches them to create a new training image.”
BackgroundPK-GCN: Prior Knowledge Assisted Image Classification using Graph Convolution NetworksPreprint
Xiao, X., Ji, C., Mudiyanselage, T. B., & Pan, Y. F. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2009.11892 · Cited by 5
StackMix: A complementary Mix algorithmPreprint
Chen, J., Sinha, S., & Kyrillidis, A. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2011.12618 · Cited by 8
Nguyen, K. T. A., Tran, N. H., & Dinh, Q.-V. (2024) · IEEE Access · doi:10.1109/access.2024.3485479 · Cited by 4
“Reference [15] introduced an innovative augmentation method, Random Image Cropping and Patching (RICAP), which further pushed the boundaries of augmentation techniques, achieving state-of-the-art results in specific tasks.”
MethodLang, V., Herrmann, C. T. E., Fuchs, M., & Ihlenfeldt, S. (2025) · Applied Sciences · doi:10.3390/app15041727 · Cited by 5
Semantic Image Captioning using Cosine Similarity Ranking with Semantic Search
Jain, K., Gandhi, S., Singhal, S., & Rajput, S. (2023) · International Conference on Contemporary Computing · doi:10.1145/3607947.3607987 · Cited by 3
“We also perform some data augmentation techniques, such as random cropping and flipping [7], to increase the diversity of the training data.”
MethodSynthetic Dataset Generation Using Photo-Realistic Simulation with Varied Time and Weather Axes
Lee, T., McKeever, S., & Courtney, J. (2024) · Electronics · doi:10.3390/electronics13081516 · Cited by 3
Deep learning-based beat-to-beat arterial blood pressure estimation using distant radar signals
Chowdhury, F. A., Hosain, M. K., Hossain, M. S., Chowdhury, M. E. H., Mahmud, S., Kabir, M. A., Alqahtani, A., & Hasan, A. (2025) · Neural Computing and Applications · doi:10.1007/s00521-025-11327-x · Cited by 3
“This technique is akin to the patching method used in image processing [83], improving model performance and reducing the computational load during training.”
GUNNEL: guided mixup augmentation and multi-model fusion for aquatic animal segmentation
Le, M.-Q., Le, T.-N., Nguyen, T., Echizen, I., & Tran, M. (2025) · Neural Computing and Applications · doi:10.1007/s00521-025-11417-w · Cited by 3
“Data warping augmentation methods, which include geometric and color transformations, are always used in training deep learning models ( e.g .scaling, flipping, rotation, random crop [34], color jittering [35]).”
NERULA: A Dual-Pathway Self-supervised Learning Framework for Electrocardiogram Signal Analysis
Manimaran, G., Puthusserypady, S. K., Domínguez, H., Atienza, A., & Bardram, J. E. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-11402-0_3 · Cited by 3
Improving the performance of Transformer Context Encoders for NER
Chawla, A., Mulay, N., Bishnoi, V., & Dhama, G. (2021) · International Conference on Information Fusion (FUSION) · doi:10.23919/fusion49465.2021.9627061 · Cited by 3
Wei, A.-T., Wang, H., Dickens, T. J., & Chi, H. (2022) · IISE Transactions · doi:10.1080/24725854.2022.2080306 · Cited by 3
“The size of the input data for learning can be increased by applying crop, flip, scale, rotate, translation, and Gaussian noises (Takahashi et al ., 2019) to the limited data.”
BackgroundOkadome, Y., Ata, K., Ishiguro, H., & Nakamura, Y. (2022) · Transactions of the Japanese Society for Artificial Intelligence · doi:10.1527/tjsai.37-6_b-m43 · Cited by 3
Shrimp Surfacing Recognition System in the Pond Using Deep Computer Vision
Varma, G. T., & Adusumalli, S. K. (2022) · Smart innovation, systems and technologies · doi:10.1007/978-981-19-4162-7_21 · Cited by 3
Maeda, K., Takada, S., Haruyama, T., Togo, R., Ogawa, T., & Haseyama, M. (2022) · Sensors · doi:10.3390/s22228932 · Cited by 3
“In this case, the data augmentation methods, such as RICAP, are likely to generate images that contain many regions where no distressed region exists, limiting the improvement of the distress detection performance.”
MethodInfluentialStrong Baseline with Auto-encoder for Scale-Invariant Person Re-identification
Bilakeri, S., & Kotegar, K. (2022) · IEEE International Conference on Distributed Computing, VLSI, Electrical Circuits and Robotics (DISCOVER) · doi:10.1109/discover55800.2022.9974635 · Cited by 3
“By altering the location of visible features in an image, random cropping prevents a neural network from over-fitting particular features [24].”
BackgroundA metal surface defect detection method based on attention mechanism and softpool
Wang, L., Fang, X., Zhang, T., Liang, W., & Liu, C. (2023) · International Conference on Optical and Photonic Engineering (icOPEN) · doi:10.1117/12.2667074 · Cited by 3
Random Pixel Selection through Image Cropping for Data Augmentation and Classification
Boudouh, N., & Mokhtari, B. (2022) · International Symposium on iNnovative Informatics of Biskra (ISNIB) · doi:10.1109/isnib57382.2022.10076253 · Cited by 3
“[12], it randomly patches and crops four images to produce a new training image.”
BackgroundFaceMix: Transferring Local Regions for Data Augmentation in Face Recognition
Garaev, N., Smirnov, E., Galyuk, V., & Lukyanets, E. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-30111-7_56 · Cited by 3
Nozaka, H., Kamata, K., & Yamagata, K. (2023) · IEICE Transactions on Information and Systems · doi:10.1587/transinf.2022dlp0066 · Cited by 3
Data Augmentation Using Corner CutMix and an Auxiliary Self-Supervised Loss
Fang, F., Hoang, N. M., Xu, Q., & Lim, J. (2023) · Proceedings - International Conference on Image Processing · doi:10.1109/icip49359.2023.10222009 · Cited by 3
“Three commonly used augmentation methods are random cropping [1], random color distortion [2] and random flipping [3].”
MethodUrban scene semantic segmentation using the U-Net model
Ciecholewski, M. (2023) · Annals of Computer Science and Information Systems · doi:10.15439/2023f3686 · Cited by 4
“2) Random cropping: In the next step, random cropping [20] was used to obtain images with the size of 256 × 256 pixels.”
MethodDeep learning based recognition of shape-coded microparticles
Şahin, M. A., van den Eijnden, L., Bhiri, C., & Destgeer, G. (2023) · Frontiers in Lab on a Chip Technologies · doi:10.3389/frlct.2023.1248265 · Cited by 3
Super-Resolution of Color Halftone Images Using Convolutional Neural Networks
Novaes, G. A. S., & Kim, H. Y. (2024) · IEEE Access · doi:10.1109/access.2024.3352431 · Cited by 3
“The random cropping of images during the training works as a form of regularization [24] that helps to prevent overfitting and to improve the generalization capabilities of the model.”
BackgroundMHAiR: A Dataset of Audio-Image Representations for Multimodal Human Actions
Shaikh, M. B., Chai, D., Islam, S. M. S., & Akhtar, N. (2024) · Data · doi:10.3390/data9020021 · Cited by 3
“DA was customarily performed through Rotation [22], Flipping [23], Cropping [24], Scaling [25], Translation [26], Noise Injection [27], Color Modification [28], and other modes.”
MethodEnhancing Road Network Extraction from Remote Sensing Images
Imam, M. E.-M., Meddeber, L., Belaid, B. M., Bendoukha, H., & Zouagui, T. (2024) · IEEE Mediterranean and Middle-East Geoscience and Remote Sensing Symposium (M2GARSS) · doi:10.1109/m2garss57310.2024.10537360 · Cited by 3
“[16] [17].”
BackgroundChen, H., Zhao, B., Yue, G., Liu, W., Lv, C., Wang, R., & Zhou, F. (2024) · Proceedings - International Conference on Image Processing · doi:10.1109/icip51287.2024.10648174 · Cited by 3
Estimating the Sonic Mach Number in the Interstellar Medium with Convolutional Neural Networks
Schmaltz, T., Hu, Y., & Lazarian, A. (2025) · The Astrophysical Journal · doi:10.3847/1538-4357/adb7ce · Cited by 4
A Novel System for Image Text Recognition and Classification using Deep Learning
Manzoor, S. I., & Singla, J. (2021) · International Conference on Computing Sciences (ICCS) · doi:10.1109/iccs54944.2021.00020 · Cited by 3
Zheng, Y., Shen, Q., Wang, M., Yang, M., Huang, J., & Feng, C. (2021) · International Journal of Remote Sensing · doi:10.1080/01431161.2021.1907866 · Cited by 3
Detection of Covid-19 from Joint Time and Frequency Analysis of Speech, Breathing and Cough Audio
Harvill, J. B., Wani, Y. R., Chatterjee, M., Alam, M., Beiser, D. G., Chestek, D. W., Hasegawa‐Johnson, M., & Ahuja, N. (2022) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp43922.2022.9746015 · Cited by 4
A Fast and Efficient Data Augmentation for Sematic Segmentation Based on LQE Head and BAS-DP
Wang, F., & Wang, Z. (2022) · IEEE Access · doi:10.1109/access.2022.3174199 · Cited by 3
Random Shadows and Highlights: A new data augmentation method for extreme lighting conditionsPreprint
Mazhar, O., & Kober, J. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2101.05361 · Cited by 11
“To prevent this, RICAP [14] spatially blend four cropped images to create a new image.”
BackgroundSugarcane Bud Detection Method Based onYOLOv3-CSE NetworkPreprint
Xu, H., Liu, J., He, J., Shen, M., & Duan, Y. (2022) · Research Square · doi:10.21203/rs.3.rs-1298165/v1 · Cited by 3
An Improved Yolo-V5 Network for Defect Detection of a Boiler Inner WallPreprint
Sun, X., Jia, X., Liang, Y., Yang, B., Wang, M., & Chi, X. (2022) · SSRN Electronic Journal · doi:10.2139/ssrn.4057058 · Cited by 3
Pan, F., Fan, Q., Xie, H., Bai, C., Zhang, Z., Chen, H., Yang, L., Zhou, X. H., Bao, Q., & Liu, C. (2023) · Bioengineering · doi:10.3390/bioengineering10101192 · Cited by 3
“Before training, we performed random cropping for data augmentation on the original images to obtain 128 × 128-pixel images to prevent overfitting issue in the training process [35].”
MethodAizaz, Z., Khare, K., & Tirmizi, A. (2023) · Lecture notes in networks and systems · doi:10.1007/978-981-99-3963-3_43 · Cited by 3
Monson, P. M. D. C., de Almeida, V. A. D., David, G. A., Oliveira, P., & Dotto, F. R. L. (2024) · International Electronic Conference on Sensors and Applications (ECSA) · doi:10.3390/ecsa-11-20476 · Cited by 2
Li, Z., Wang, Z., Xu, X., Chen, Y., & Schuller, B. W. (2025) · IEEE Transactions on Medical Robotics and Bionics · doi:10.1109/tmrb.2025.3561865 · Cited by 2
“Note that we perform image cropping [71], [72] for all the training samples, with the sizes for ALID → EndoVis 2015 set to 480 × 480, and for ALID → EndoVis 2017 adjusted to 512 × 512.”
Enhancing out-of-distribution learning in computer vision through dominant feature masking
Pilzak, A., & Thivierge, J. (2025) · Pattern Analysis and Applications · doi:10.1007/s10044-025-01474-1 · Cited by 2
GCE-Net: single-image deraining with adaptive normalization and multi-scale feature integration
Jin, T., Zhang, J., Qin, G., & Mingzhou, M. (2025) · The Journal of Supercomputing · doi:10.1007/s11227-025-07603-1 · Cited by 2
“Various data augmentation techniques were applied during model training, such as random rotation, flipping, and cropping [41].”
Asare, J. W., Ujakpa, M. M., Alornyo, S., Kyei, E. A., Modey, P., Brown‐Acquaye, W. L., Freeman, E., & Coleman, A. (2025) · Journal of Electrical Systems and Information Technology · doi:10.1186/s43067-025-00238-5 · Cited by 2
Li, J., Liu, M., Peng, H., Liu, L., Zheng, X., Chen, Y., Liu, G., Zhou, J., & Lyu, F. (2025) · Peer-to-Peer Networking and Applications · doi:10.1007/s12083-025-02079-5 · Cited by 2
“Subsequently, in the data augmentation phase, we apply a series of random transformations techniques to the original images to extend the quantity and diversity of the collected image dataset [38].”
Intelligent recognition of weak discontinuities on outcrops of hard rock masses
Zhang, W., Xu, G., Li, T., Wu, D., Zhou, H., Chen, L., & Chen, X. (2025) · Journal of Rock Mechanics and Geotechnical Engineering · doi:10.1016/j.jrmge.2025.07.020 · Cited by 2
Priananda, C. W., Illias, H. A., Raymond, W. J. K., Hechifa, A., & Negara, I. M. Y. (2025) · IEEE Transactions on Dielectrics and Electrical Insulation · doi:10.1109/tdei.2025.3648258 · Cited by 2
“Toaddress class imbalance in the original data, an initial augmentation step was applied to expand each class to approximately 100 samples [11], [16], [17].”
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.