Data augmentation: A comprehensive survey of modern approaches

Mumuni, Alhassan; Mumuni, Fuseini

doi:10.1016/j.array.2022.100258

Cited by 237 publications

(75 citation statements)

References 423 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Data augmentation is a typical approach to data pre-processing used to expand the variety of training data, increasing the distribution of datasets, and improving deep learning models' generalization ability. [26][27][28] We applied eight data augmentation techniques are suitable for the camera module production environment, such as crop transformation, translational transformation, adding Gaussian noise, rotating the image angle, adjust-ing brightness and contrast, flipping the image horizontally, and mean blur. These techniques increase the diversity and distribution of the data set, making the augmented images more realistic and reliable.…”

Section: Automatic Data Set Expansion Methods Based On Data Augmentationmentioning

confidence: 99%

YOLOv7‐EAS: A Small Target Detection of Camera Module Surface Based on Improved YOLOv7

Zou

Yuan

et al. 2023

Advcd Theory and Sims

View full text Add to dashboard Cite

The camera module surface may introduce defects in the focusing process, and manual detection of these defects is inefficient and costly. To address this problem, a method to automatically expand the dataset for deep learning models with fewer original images is developed. A defect detection method is also presented, YOLOv7‐EAS, that enhances YOLOv7 for camera module images with complex backgrounds and small target defects. This method introduces NewEIOU Loss, a bounding box regression loss function that extends EIOU Loss with a nonlinear compensation term. This function improves the model's convergence rate to quadratic order. Moreover, the detection head based on ASFF is redesigned to adapt the layer features and spatially filter out irrelevant information to improve the detection rate of small targets. Finally, a 3D attention mechanism is applied to increase the model's ability to extract fine‐grained features without increasing the network parameters. Experimental results show that the improved YOLOv7 algorithm achieves 90.5% mAP for the camera module image dataset, which is 10.7% higher than the original algorithm, at 136 FPS. This method can perform the camera module surface defect detection task more accurately and meet the real‐time requirements of the production line.

show abstract

Section: Automatic Data Set Expansion Methods Based On Data Augmentationmentioning

confidence: 99%

YOLOv7‐EAS: A Small Target Detection of Camera Module Surface Based on Improved YOLOv7

Zou

Yuan

et al. 2023

Advcd Theory and Sims

View full text Add to dashboard Cite

show abstract

“…Deep learning is a branch of machine learning that uses multilayer neural networks to learn features and patterns from data and perform prediction and analysis, such as random forest (RF), feedforward neural network (FNN), back-propagation neural network (BPNN), convolutional neural network (CNN), reinforcement learning, and recurrent neural network (RNN). Deep learning has a wide range of applications in various fields, e.g., face recognition, medical image analysis, autonomous driving, sentiment analysis, dialog systems, machine translation, and speech recognition. − …”

Section: Literature Reviewmentioning

confidence: 99%

Optimization of a Gas–Liquid Dual-Impeller Stirred Tank Based on Deep Learning with a Small Data Set from CFD Simulation

Kang,

Feng,

Wang

2023

Ind. Eng. Chem. Res.

View full text Add to dashboard Cite

An optimization strategy based on deep learning was developed for dual-impeller design in a gas−liquid stirred tank. The optimization objective was to maximize the gas−liquid specific interfacial area and minimize power consumption. A small raw data set was obtained with time-consuming computational fluid dynamics (CFD) simulation for different dual-impeller designs. The noise injection method was employed as a data augmentation technique to generate the virtual data. Based on the augmented data set, feedforward neural network (FNN) was used to model the relationship between the six geometric variables of impellers and two objective functions, while Bayesian optimization was applied to tune the hyperparameters of FNN. A two-stage deep learning approach was introduced to enhance the prediction of power consumption based on the correlation between the specific interfacial area and power consumption. The deep learning model was used to calculate the entire solution space and, thereby, capture the Pareto optimal solutions. The proposed method validated by a data set of CFD simulations could achieve high prediction accuracy and robustness.

show abstract

“…The data augmentation methods are classified into two categories: (1) augmentation of original data (2) and generation of synthetic/artificial data ( Mumuni and Mumuni, 2022 ). The augmentation of original data generally includes pixel transformation, affine transformation, and elastic transformation.…”

Section: Related Workmentioning

confidence: 99%

Developing precision agriculture using data augmentation framework for automatic identification of castor insect pests

Nitin

Gupta²,

Yadav³

et al. 2023

Front. Plant Sci.

View full text Add to dashboard Cite

Castor (Ricinus communis L.) is an important nonedible industrial crop that produces oil, which is used in the production of medicines, lubricants, and other products. However, the quality and quantity of castor oil are critical factors that can be degraded by various insect pest attacks. The traditional method of identifying the correct category of pests required a significant amount of time and expertise. To solve this issue, automatic insect pest detection methods combined with precision agriculture can help farmers in providing adequate support for sustainable agriculture development. For accurate predictions, the recognition system requires a sufficient amount of data from a real-world situation, which is not always available. In this regard, data augmentation is a popular technique used for data enrichment. The research conducted in this investigation established an insect pest dataset of common castor pests. This paper proposes a hybrid manipulation-based approach for data augmentation to solve the issue of the lack of a suitable dataset for effective vision-based model training. The deep convolutional neural networks VGG16, VGG19, and ResNet50 are then adopted to analyze the effects of the proposed augmentation method. The prediction results show that the proposed method addresses the challenges associated with adequate dataset size and significantly improves overall performance when compared to previous methods.

show abstract

Data augmentation: A comprehensive survey of modern approaches

Cited by 237 publications

References 423 publications

YOLOv7‐EAS: A Small Target Detection of Camera Module Surface Based on Improved YOLOv7

YOLOv7‐EAS: A Small Target Detection of Camera Module Surface Based on Improved YOLOv7

Optimization of a Gas–Liquid Dual-Impeller Stirred Tank Based on Deep Learning with a Small Data Set from CFD Simulation

Developing precision agriculture using data augmentation framework for automatic identification of castor insect pests

Contact Info

Product

Resources

About