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 301–400 of 476
Xiang, Y., Xian, Y., Cairang, X., Dorji, G., Danzeng, P., & Zhao, Q. R. (2026) · International Journal on Document Analysis and Recognition (IJDAR) · doi:10.1007/s10032-025-00567-1 · Cited by 2
“Yet many systems still miss thin strokes and subtle textures, and they struggle to balance global context with local detail on pages with complex layouts [8–10].”
Wang, X., Li, Y., Gong, Y., Liu, M., & Luo, K. (2026) · Journal of Intelligent Manufacturing · doi:10.1007/s10845-025-02774-3 · Cited by 2
DNA: Dual-branch Network with Adaptation for Open-Set Online Handwriting Generation
Huang, T.-L., Do-Tran, N.-T., Le, N.-H.-L., Shuai, H.-H., & Huang, C. (2026) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv61042.2026.00406 · Cited by 2
“A common solution is data augmentation [27–29], which increases dataset diversity us-ing techniques such as flipping, cropping, and more advanced transformations [5, 8, 37].”
A Deep Learning-Based Framework for the Classification of Non-functional Requirements
Sabir, M., Banissi, E. K., & Child, M. (2021) · Advances in intelligent systems and computing · doi:10.1007/978-3-030-72651-5_56 · Cited by 3
Dynamic Channel Selection in Self-Supervised Learning
Krishna, T., Rai, A., Djilali, Y. A. D., Smeaton, A. F., McGuinness, K., & O'Connor, N. E. (2022) · International Machine Vision and Image Processing Conference (IMVIP) · doi:10.56541/lkli8696 · Cited by 2
MEKANSAL PİRAMİT HAVUZLAMA TABANLI EVRİŞİMLİ SİNİR AĞI İLE OTOMATİK DRONE SINIFLANDIRMA
Korkmaz, D., & Açıkgöz, H. (2022) · Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi · doi:10.17780/ksujes.1113669 · Cited by 2
Rui-Jia, L., & Su, Z.-K. (2023) · Journal of Optics · doi:10.1088/2040-8986/acb36d · Cited by 2
“In addition to the above methods, data augmentation [25, 26], batch normalization [29] and early stop method [30] are used to avoid over fitting.”
Saini, D., Garg, R., Malik, R., Prashar, D., & Faheem, M. (2023) · Signal Image and Video Processing · doi:10.1007/s11760-023-02859-7 · Cited by 2
Research on Small Target Detection Method for Industrial Safety Helmets Based on Improved YOLOv8
Lan, Y., Mo, C., Li, C., Wang, Q., & Min, L. (2024) · Journal of Computing and Information Technology · doi:10.20532/cit.2023.1005754 · Cited by 4
Development of the Neural Network Based Recognition Methods at V.L. Arlazarov’s Scientific School
Sheshkus, A. V., Kondrashova, A. N., & Nikolaev, D. (2023) · Pattern Recognition and Image Analysis · doi:10.1134/s1054661823040417 · Cited by 4
“Examples of such techniques include erasing part of an image [6, 38], creating a new image from parts of existing ones [32], or mixing information from two other images [37].”
Fusion of Spiral Convolution-LSTM for Intrusion Detection Modeling
Wang, F., & Dong, Z. X. (2024) · Computers, materials & continua/Computers, materials & continua (Print) · doi:10.32604/cmc.2024.048443 · Cited by 2
Novel Image Data Augmentation Technique for Deep Learning Using Least Significant Bit Encryption
Ghnemat, R., & Almashaqbeh, S. (2024) · International Conference on Machine Learning Technologies (ICMLT) · doi:10.1145/3674029.3674053 · Cited by 2
“The author in [16] suggest a novel method for enhancing data termed random image cropping and patching (RICAP), which randomly patches together four training photos after croping them.”
Image Manipulation Quality Assessment
Wu, X., Lou, J., Wu, Y., Liu, W., Rosin, P. L., Colombo, G. B., Allen, S. M., Whitaker, R. M., & Liu, H. (2024) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2024.3504854 · Cited by 3
“To mitigate the risk of excessive cropping that may lead to the loss of semantic information and alterations of the aspect ratio, each image is randomly cropped to two-thirds and three-quarters of the original size while maintaining the original proportions [58].”
Development of Deep Intelligence for Automatic River Detection (RivDet)
Lee, S., Kong, Y., & Lee, T. (2025) · Remote Sensing · doi:10.3390/rs17020346 · Cited by 2
“…the model on different parts of the object and allowing it to detect objects that are partially visible or obscured, as shown in Figure 5c. Additionally, the cutout operation randomly removes some parts of the river image, enabling the model to recognize objects despite missing information [33].”
Detection and severity assessment of tea leaf blight from UAV remote sensing images
Jiang, Y., Wang, B., & Hu, G. (2025) · Australian Journal of Agricultural Research · doi:10.1071/cp23351 · Cited by 2
Jiang, Y., Yu, G., Wang, Q., Wang, X., Ma, B., Sun, C., Ni, W., & Liu, R. P. (2025) · ACM SIGSAC Conference on Computer and Communications Security (CCS) · doi:10.1145/3719027.3744787 · Cited by 13
“The time overhead for the label expansion strategy discussion between clients is negligible, measured in milliseconds, and does not significantly impact our analysis of training and transmission times.”
Shih, P.-C., Li, H., Lin, X., & Zhang, Y. (2019) · International Conference on Computer Science and Application Engineering (CSAE) · doi:10.1145/3331453.3361326 · Cited by 2
“For example, Wang et al. [6] proposed a neural network that can learn augmentations to improve the classifier performance, and Ryo Takahashi et al. [7] proposed random image cropping and patching (RICAP) to create a new training image by randomly cropping four images and patching them into one.”
BackgroundA Crowd Alert System for Pandemic Control using Yolov4
Abhinand, H., Vinod, R., & Anjali, T. (2022) · International Conference on Trends in Electronics and Informatics (ICOEI) · doi:10.1109/icoei53556.2022.9776671 · Cited by 2
Texture improvement for human shape estimation from a single image
Escribano, J. G., Rauno, S., Swaminathan, A., Smyth, D., & Smolić, A. (2022) · International Machine Vision and Image Processing Conference (IMVIP) · doi:10.56541/soww6683 · Cited by 2
Data augmentation of random grid-hiding for video object segmentation
Wang, S., Yao, R., Zhang, Y., Jiang, Q., & Zhang, C. (2019) · Multimedia Tools and Applications · doi:10.1007/s11042-019-7569-5 · Cited by 2
“took a different cropping approach in RICAP [36].”
MethodInfluentialKaradağ, Ö. Ö., & Erdaş, Ö. (2020) · International Journal of Intelligent Systems and Applications in Engineering · doi:10.18201/ijisae.2020261593 · Cited by 2
“We use the “imaug” library [21] for transformation based methods.”
MethodSurvey on Deep Learning-Based Kuzushiji Recognition
Ueki, K., & Kojima, T. (2021) · Lecture notes in computer science · doi:10.1007/978-3-030-68787-8_7 · Cited by 2
Evolving Image Compositions for Feature Representation Learning
Cascante-Bonilla, P., Sekhon, A., Qi, Y., & Ordóñez, V. (2021) · British Machine Vision Conference (BMVC) · doi:10.5244/c.35.325 · Cited by 5
Data Augmentation Guidelines for Cross-Dataset Transfer Learning and Pseudo Labeling
dos Santos, F. P., Thumé, G. S., & Ponti, M. A. (2021) · SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI) · doi:10.1109/sibgrapi54419.2021.00036 · Cited by 4
Unsupervised Scale-Invariant Multispectral Shape Matching
Pazi, I., Ginzburg, D., & Raviv, D. (2022) · International Machine Vision and Image Processing Conference (IMVIP) · doi:10.56541/vhmq4826 · Cited by 5
Hizukuri, A., Kunieda, S., & Nakayama, R. (2022) · Electronics and Communications in Japan · doi:10.1002/ecj.12368 · Cited by 2
Training Strategy for Limited Labeled Data by Learning from Confusion
Idris, A., Khaleel, M., Tavanapong, W., Oh, J., & de Groen, P. (2022) · Proceedings - International Conference on Image Processing · doi:10.1109/icip46576.2022.9897357 · Cited by 2
“The new images are assigned an entirely different label, the not-sure label, rather than using the label of any of the two images as done in [13] and [14].”
BackgroundObject-Based Perspective Transformation Data Augmentation for Object Detection
Nie, Z., Cao, J., Weng, N., Yu, X., & Wang, M. (2022) · International Conference on Frontiers of Artificial Intelligence and Machine Learning (FAIML) · doi:10.1109/faiml57028.2022.00043 · Cited by 2
“Image Mixing can be achieved through multiple approaches, one of which is combining randomly cropped images together to form new images [13].”
BackgroundCNN2Graph: Building Graphs for Image Classification
Trivedy, V., & Latecki, L. J. (2023) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv56688.2023.00009 · Cited by 5
“For data augmentation, we use Random Crop [51] and color jitter on both the baseline and experimental models.”
MethodRegional Patch-Based Feature Interpolation Method for Effective Regularization
Jang, S., Jin, K., An, J., & Kim, Y. (2020) · IEEE Access · doi:10.1109/access.2020.2974328 · Cited by 2
“In addition to mixing pixels, there has been a study to connect images to each region [27].”
BackgroundDomain-Agnostic Clustering with Self-DistillationPreprint
Adnan, M., Ioannou, Y., Tsai, C.-Y., & Taylor, G. W. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2111.12170 · Cited by 3
“For example, color jittering commonly used in SSL cannot be used on black and white x-ray images [22], and random cropping [23] is not relevant to histopathology images where the region of interest is significantly smaller than the image itself.”
BackgroundYang, S. W., Park, H. J., Kim, J. S., Choi, W., Park, J., & Han, S. W. (2022) · SSRN Electronic Journal · doi:10.2139/ssrn.4181525 · Cited by 2
Görüntü artırma tekniklerinin karşılaştırılması
Özcan, Ö., & KARAALTUN, M. (2023) · Ömer Halisdemir Üniversitesi Mühendislik Bilimleri Dergisi · doi:10.28948/ngumuh.1270466 · Cited by 1
ScrambleMix: A Privacy-Preserving Image Processing for Edge-Cloud Machine Learning
Madono, K., Tanaka, M., & Onishi, M. (2024) · Lecture notes in computer science · doi:10.1007/978-981-97-0376-0_25 · Cited by 2
Technology and Information: Data Augmentation Applied to Industry in Modern Times
Vaz, G. C., de Oliveira, G. G., Arthur, R., & Iano, Y. (2024) · Smart innovation, systems and technologies · doi:10.1007/978-3-031-66961-3_26 · Cited by 2
SegTest: Metamorphic Testing of Image Segmentation via Guided Instance‐Level Test Data Augmentation
Hou, Z., Wang, X., Zhang, S., & Chen, Z. (2024) · Software Testing Verification and Reliability · doi:10.1002/stvr.1910 · Cited by 1
“These methods, known as image-level data augmentation methods, can lead to a coupling between the background and the instances within the image [17, 18].”
Data-Driven Position Estimation and Sensorless Control of PMSM Drive in Low-Speed Region
Shim, J.-H., Park, B. R., Kim, M., & Ha, J.-I. (2025) · IEEE Transactions on Industry Applications · doi:10.1109/tia.2025.3595132 · Cited by 2
“Notably, while data augmentation techniques are widely used in the machine learning field [26], [27], [28], conventional studies on data-driven sensorless methods have rarely explored such approaches [22], [23], [24], [25].”
Koch, D., Despotović, M., Thaler, S., & Zhang, Z. (2025) · Journal of European real estate research · doi:10.1108/jerer-03-2024-0013 · Cited by 1
Advanced Data Augmentation Techniques for Biosignal Processing in Online Systems
Chen, P., Tang, Y., Li, D., Jiang, H. Y., Togo, S., Yokoi, H., & Jiang, Y. (2025) · IEEE/ASME International Conference on Advanced Intelligent Mechatronics · doi:10.1109/aim64088.2025.11175708 · Cited by 1
“In the field of computer vision, real-world image variations are often simulated through operations such as jittering [23], cropping [24], and rotation.”
Enhancing Distilled Datasets Via Natural Data Mixing
Pons, I., Stern, G., Costa, A. H. R., & Jordao, A. (2025) · SIBGRAPI Conference on Graphics, Patterns and Images (SIBGRAPI) · doi:10.1109/sibgrapi67909.2025.11223402 · Cited by 1
Asare, J. W., Kyei, E. A., Ujakpa, M. M., Alornyo, S., Mutasa, L. S., Brown‐Acquaye, W. L., Лемпого, Ф., Freeman, E., & Coleman, A. (2026) · Communications in computer and information science · doi:10.1007/978-3-032-13056-3_11 · Cited by 1
Graph Neural Networks With Multi-Level Augmentation for Low-Degree Graph
Heng, G., Chen, Y., Huang, Q., Hu, L. Y., Li, J., Wang, S., & Xiao, S. (2026) · IEEE Transactions on Network Science and Engineering · doi:10.1109/tnse.2026.3656815 · Cited by 1
“Data augmentation methods have found extensive applications across various fields [26], [27], [28], such as image rotation [29], cropping [30], and flipping [31] in computer vision, and synonym replacement [32] and text perturbation [33] in nat-ural language processing.”
Grid-Shift: An Image Preprocessing Approach to Reduce Overfitting in AI Training
Kocon, K., Krämer, M., & Budde, L. E. (2025) · International Conference on Frontiers of Signal Processing (ICFSP) · doi:10.1109/icfsp67350.2025.11353536 · Cited by 1
“Takahashi et al. have already shown that non-linear blending methods are superior to linear stacking methods [29].”
Skin Cancer Classification Via Shifted Window Attention: A Swin Transformer Approach
Alrahhal, M., Yateem, N. A., Bonny, T., Hamad, M. S., Al-Shabi, M. A. Q., & Hussain, A. J. (2025) · Advances in Science and Engineering Technology International Conferences (ASET) · doi:10.1109/aset66891.2025.11428002 · Cited by 1
“These include random cropping to a fixed dimension 𝐷 × 𝐷 , followed by random horizontal flipping [30].”
AM-Adapter: Appearance Matching Adapter for Exemplar-Based Semantic Image Synthesis In-the-Wild
Jin, S., Nam, J., Kim, J., Chung, D., Kim, Y.-S., Park, J., Chu, H., & Kim, S. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.01586 · Cited by 1
“Specifically, we apply different augmentations, such as random cropping [30] and flipping [39], to an anchor image to generate exemplar-target pairs ( I X , I Y ) .”
Automated data synthesis and retrieval-augmented generation for legal large language models
Ren, W., Shen, L., Zhu, J., Li, S., Hong, Y., Wang, J., & Cao, D. (2026) · Knowledge-Based Systems · doi:10.1016/j.knosys.2026.116267 · Cited by 4
Data Augmentation for DOA Estimation Under Limited Snapshots Scenarios
Zeng, H., Lan, X., & Wang, X. (2026) · IEEE Sensors Journal · doi:10.1109/jsen.2026.3695644 · Cited by 1
“At present, data augmentation is most widely used in the field of computer vision [20], [21], [22].”
Dual Active Sampling on Batch-Incremental Active Learning
Phan, J., Ruocco, M., & Scibilia, F. (2019) · Communications in computer and information science · doi:10.1007/978-3-030-35664-4_12 · Cited by 1
View Sub-sampling and Reconstruction for Efficient Light Field Compression
Chen, Y., Alain, M., & Smolić, A. (2022) · International Machine Vision and Image Processing Conference (IMVIP) · doi:10.56541/swcc3620 · Cited by 1
Lee, T., McKeever, S., & Courtney, J. (2022) · International Machine Vision and Image Processing Conference (IMVIP) · doi:10.56541/poya9239 · Cited by 1
An Improved Method of Image Recognition with Deep Learning Combined with Attention Mechanism
Fang, X., Wang, L., Liu, C., & Tao, H. (2022) · International Conference on Image, Vision and Computing (ICIVC) · doi:10.1109/icivc55077.2022.9887045 · Cited by 1
“as color jitter, Cutout, adding noise and random clipping [19] to enhance the data of flower images, so that the limited data can generate value equivalent to more data.”
BackgroundBias oriented unbiased data augmentation for cross-bias representation learning
Li, L., Tang, F., Cao, J., Li, X., & Wang, D. (2022) · Multimedia Systems · doi:10.1007/s00530-022-01013-6 · Cited by 1
“The former artificially inflates the training data-set by either data warping or oversampling, such as geometric and color transformations, noise injection, and random patching [24–26].”
MethodSTAug: Copy-Paste Based Image Augmentation Technique Using Salient Target
Kang, J., & Chung, K. (2022) · IEEE Access · doi:10.1109/access.2022.3224141 · Cited by 3
“As a data augmentation technique, random image cropping and patching (RICAP) randomly selects four images, cuts them randomly, and generates one image.”
BackgroundMethodInfluentialAn Extended Visual Intelligence Scheme for Disassembly in Automated Recycling Routines
Yildiz, E., Renaudo, E., Hollenstein, J., Piater, J., & Wörgötter, F. (2022) · Communications in computer and information science · doi:10.1007/978-3-031-19650-8_2 · Cited by 2
Investigating Image Augmentation for Classification of Chest X-Ray Images
Almattar, W., & Algherairy, A. S. (2022) · International Conference on Computational Intelligence and Communication Networks (CICN) · doi:10.1109/cicn56167.2022.10008268 · Cited by 1
“Moreover, in this study, they investigated the use of three types of data augmentation : Conventional methods (i.e. rotation, x-axis shift, y-axis shift, horizontal flipping, scaling and Shear transformation) , Mixup [19] and Random Image Cropping and Patching (RICAP) [20].”
MethodJoint Distribution and Class-based Data Augmentation for Wildlife Detection
Pan, Y., Sui, C., Jiang, F., Yang, G., Zang, A., & Zhou, S. (2022) · International Conference on Automation, Robotics and Computer Engineering (ICARCE) · doi:10.1109/icarce55724.2022.10046567 · Cited by 2
“…generate multiple augmentation samples by performing different transformations on the original samples, such as random inversion, color channel change [1], rotation [2], cropping [3], noise [4], RandomErasing [5], Cutout [6], GridMask [7], and a combination of methods such as RandAugmentation [8].”
MethodAMGAN: An Attribute-Matched Generative Adversarial Network for UAV Virtual Sample Generation
Yang, Z., Jia, X., Shen, Y., Yang, Y., Li, H., & Zhang, W. E. (2023) · Neural Processing Letters · doi:10.1007/s11063-023-11304-2 · Cited by 1
Talukder, M. H., Ota, S., Takanokura, M., & Ishii, N. (2023) · Communications in computer and information science · doi:10.1007/978-3-031-37320-6_7 · Cited by 1
Towards Adaptable and Interactive Image Captioning with Data Augmentation and Episodic Memory
Anagnostopoulou, A., Hartmann, M., & Sonntag, D. (2023) · Workshop on Simple and Efficient Natural Language Processing (SustaiNLP) · doi:10.18653/v1/2023.sustainlp-1.19 · Cited by 1
Improvement of a Rapid Method of Detecting Gasoline Detergency Based on the Image Recognition
Zhang, R., Zhu, R., Jia, M., Pang, Y., Zhang, B., Bao, X., & Wang, Y. (2023) · ACS Omega · doi:10.1021/acsomega.3c05350 · Cited by 1
Texture-Based Data Augmentation for Small Datasets
Dash, A., & Albu, A. B. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-45382-3_29 · Cited by 1
Enhancing Object Detection in YouTube Thumbnails Forensics with YOLOv4
Sayyafzadeh, S., Chi, H., Ho, S. M., & Mkpong-Ruffin, I. (2023) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata59044.2023.10386427 · Cited by 1
“Data augmentation [13] is a technique that involves creating additional training data by applying various transformations to the original dataset.”
MethodSynthesized Image Training Techniques: On Improving Model Performance Using Confusion
Idris, A., Khaleel, M., Tavanapong, W., & de Groen, P. C. (2024) · ACM Transactions on Multimedia Computing Communications and Applications · doi:10.1145/3641856 · Cited by 1
“Besides having low computation overhead, EQM is based on MixUp [ 31 ], whereas RCP is based on RICAP [ 42 ].”
Knowledge Distillation based on Monoclass Teachers for Edge Infrastructure
Maron, C., Fresse, V., Morand, K., & Havart, F. (2024) · doi:10.5121/csit.2024.140107 · Cited by 1
“Takashi et al. [3] proposed a method combining random cropping (RICAP) which randomly crops four images and patches them to create a new training image.”
MethodZhang, H., Jiang, Y., & Zhao, L. P. (2024) · IEEE International Conference on Electronic Information and Communication Technology (ICEICT) · doi:10.1109/iceict61637.2024.10671144 · Cited by 1
“4(a without augmentation cropping augmentation [11] flipping augmentation [12] sharpening augmentation [13] CWRI augmentation Specifically, according to Fig.”
BackgroundHighway Anomaly Detection Based on Deep Learning
Yan, Z., Yin, W., & Zhai, J. (2024) · Lecture notes in electrical engineering · doi:10.1007/978-981-97-8650-3_33 · Cited by 1
Traversi, N. D. F., Evald, P. J. D. D. O., Santos, J. V. D., Drews, P., & Botelho, S. S. D. C. (2024) · IEEE International Conference on Industrial Informatics (INDIN) · doi:10.1109/indin58382.2024.10774499 · Cited by 1
Mialland, A., Fukunaga, S., Katsuki, R., Dong, Y., Yamaguchi, H., & Saitō, Y. (2026) · Bioinformatics · doi:10.1093/bioinformatics/btag415 · Cited by 2
“data augmentation methods have been commonly used in various modern machine learning approaches, demonstrating their effectiveness in fields such as computer vision (Shorten and Khoshgoftaar 2019, Takahashi et al. 2020) or natural language processing (NLP) (Feng et al. 2021, Li et al. 2022).”
Boudouh, N., & Mokhtari, B. (2024) · International Conference on Advances in Electrical and Communication Technologies (ICAECOT) · doi:10.1109/icaecot62402.2024.10828593 · Cited by 1
NucleiMix: Realistic data augmentation for nuclei instance segmentation
Wang, J., & Kwak, J. T. (2025) · Computers in Biology and Medicine · doi:10.1016/j.compbiomed.2025.110901 · Cited by 2
Zhang, J., Hu, Y., & Lazarian, A. (2025) · The Astrophysical Journal · doi:10.3847/1538-4357/adeb7a · Cited by 4
“To further enhance the model’s generalization capabilities and ensure robust performance on unseen data, we have implemented a data augmentation strategy during the training phase (Takahashi et al. 2020).”
A New Backbone Network for Instance Segmentation: Application on a Semiconductor Process Inspection
Han, J., & Hong, S. (2020) · IEEE Access · doi:10.1109/access.2020.3039356 · Cited by 2
Naoya, R., 北風, 裕., & Matsumura, R. (2021) · International Conference on Intelligent Systems and Image Processing (ICISIP) · doi:10.12792/icisip2021.032 · Cited by 1
Huang, L., Yi-gong, Z., & Yang, T. (2022) · Multimedia Tools and Applications · doi:10.1007/s11042-022-12651-3 · Cited by 1
“Data augmentation [24] was performed on all the images in the dataset to increase the number of images and thus provide more reliable information.”
MethodHizukuri, A., Kunieda, S., & Nakayama, R. (2022) · IEEJ Transactions on Electronics Information and Systems · doi:10.1541/ieejeiss.142.586 · Cited by 1
Kommidi, M., Chinta, A., Dachepally, T. K., & Chebrolu, S. (2022) · International Conference on Paradigm Shifts in Communications Embedded Systems, Machine Learning and Signal Processing (PCEMS) · doi:10.1109/pcems55161.2022.9807972 · Cited by 2
“brightness, shear, scaling, RandomCrop [26] proposed by Takahashi et al.”
BackgroundAboghazalah, M. M., Elkafrawy, P., Torkey, H., & Elsayed, A. A. (2022) · Menoufia Journal of Electronic Engineering Research · doi:10.21608/mjeer.2022.218823 · Cited by 1
Yao, W., Varkarakis, V., Lemley, J., & Corcoran, P. (2022) · International Machine Vision and Image Processing Conference (IMVIP) · doi:10.56541/fevr2516 · Cited by 1
A machine vision system for avian song classification with CNN’s
Palma, G., Aquino, A., Monticelli, P., Verdade, L. M., Markham, C., & Moral, R. D. A. (2022) · International Machine Vision and Image Processing Conference (IMVIP) · doi:10.56541/mhzn4111 · Cited by 2
Untapped Potential of Data Augmentation: A Domain Generalization ViewpointPreprint
Piratla, V., & Shankar, S. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2007.04662 · Cited by 3
“Data Augmentation Various techniques like random erasure (Zhong et al., 2020), random replacement (Takahashi et al., 2018), noise patching (Lopes et al., 2019), and image interpolation (Tokozume et al., 2018) have been ex- 3Augmix severity scale is from 0 to 10 plored for creation of augmented data…”
BackgroundMethodDiscriminative Cross-Modal Data Augmentation for Medical Imaging ApplicationsPreprint
Yang, Y., & Xie, P. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2010.03468 · Cited by 2
“Takahashi et al. (2019) propose to randomly crop four images and patch them to create a new training image.”
BackgroundMethodClassification of Chest X-ray Images Using Machine Learning TechniquesPreprint
Ahmed, A. M., Tharwat, G., Bouallègue, B., Khattab, M. M., Moustafa, A., & Matter, S. S. (2021) · Preprints.org · doi:10.20944/preprints202103.0408.v1 · Cited by 1
“In [13] a method of X-ray image compression is presented based on a Recurrent Neural Networks Convolution (RNN-Conv).”
MethodAutonomous Driving Implementation in an Experimental EnvironmentPreprint
Aliyev, N., GÜZEL, M. T., & Sezer, O. (2021) · Preprints.org · doi:10.20944/preprints202105.0568.v1 · Cited by 1
The Art of Labeling: Task Augmentation for Private(Collaborative) Learning on Transformed Data.Preprint
Xiao, H., & Devadas, S. (2021) · IACR Cryptology ePrint Archive · Cited by 4
Improving Lungs Cancer Detection Based on Hybrid Features and Employing Machine Learning TechniquesPreprint
Yang, J., Yee, P. L., Khan, A. A., Khan, M. S., Karamti, H., Aldweesh, A. Y., Hussain, L., & Omar, A. (2023) · Research Square · doi:10.21203/rs.3.rs-2929395/v1 · Cited by 1
Lin, C., Killeen, P., Li, F., Yeap, T. H., & Kiringa, I. (2023) · doi:10.36227/techrxiv.24328609 · Cited by 1
Lin, C., Killeen, P., Li, F., Yeap, T. H., & Kiringa, I. (2023) · doi:10.36227/techrxiv.24328609.v1 · Cited by 1
A Comparative Analysis of Local Binary Pattern(LBP) Variants for Image Tamper DetectionPreprint
., S., & Kaur, M. (2023) · Research Square · doi:10.21203/rs.3.rs-3608580/v1 · Cited by 1
SelectAugment: Hierarchical Deterministic Sample Selection for Data Augmentation
Lin, S., Zhang, Z., Li, X. S., & Chen, Z. (2023) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v37i2.25247 · Cited by 3
Properties and Applications of Functions that Deep Learning Approximates
Matsubara, T. (2018) · The Brain & Neural Networks · doi:10.3902/jnns.25.175
Self-citationIdentificação de Composições da Paisagem Urbana: Uma abordagem de Deep Learning
Leão, A. L. F., Abonizio, H., Barbon, S., & Kanashiro, M. (2020) · Revista de Morfologia Urbana · doi:10.47235/rmu.v8i1.140 · Cited by 3
Peer Review #2 of "Distribution-preserving data augmentation (v0.2)"
(2021) · PeerJ Computer Science · doi:10.7287/peerj-cs.571v0.2/reviews/2
Peer Review #2 of "Distribution-preserving data augmentation (v0.1)"
(2021) · PeerJ Computer Science · doi:10.7287/peerj-cs.571v0.1/reviews/2
Autonomous Driving Implementation in an Experimental EnvironmentPreprint
Aliyev, N., Sezer, O., & GÜZEL, M. T. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2106.15274
“For this reason, overfitting is prevented by generating synthetic data with the data augmentation technique [21].”
MethodFerri, M., Mantegazza, D., Cereda, E., Zimmerman, N., Gambardella, L. M., Palossi, D., Guzzi, J., & Giusti, A. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2110.14491
“In [7] and [8], augmented images are created as a collage of other randomly transformed samples.”
BackgroundUsing Fictitious Class Representations to Boost Discriminative Zero-Shot LearnersPreprint
Dabbah, M., & El‐Yaniv, R. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2111.13550
Realization of the Autonomous Driving System on the Experimental Vehicle
Aliyev, N., GÜZEL, M. T., & Sezer, O. (2022) · Academic Platform Journal of Engineering and Smart Systems · doi:10.21541/apjess.1060763
An Intelligent Visual Analysis Scheme for Automatic Disassembly Processes in the Recycling Industry
Yildiz, E. (2021) · eDiss (Georg-August-Universität Göttingen) · doi:10.53846/goediss-8573
Peer Review #3 of "Distribution-preserving data augmentation (v0.2)"
(2021) · PeerJ Computer Science · doi:10.7287/peerj-cs.571v0.2/reviews/3
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.