A Comprehensive Survey of Image Augmentation Techniques for Deep Learning

Xu, Mingle; Yoon, Sook; Fuentes, Alvaro; Park, Dong Sun

doi:10.48550/arxiv.2205.01491

Cited by 5 publications

(5 citation statements)

References 70 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Limited data is one main challenge in achieving high performance in the computer vision field (Xu et al, 2022a) and plant disease recognition (Lu et al, 2022;Xu et al, 2022b). Through our experimental results, we argue that the required amount of training dataset is partly dependent on the model or pre-trained model.…”

Section: Discussionmentioning

confidence: 86%

“…height and width, may incur challenges for recognition tasks as the disease phenomenon may not be clear enough in small-size images. To summarize, we emphasize that image variations (Xu et al, 2022a) in the dataset have an influence on training models and their corresponding performance, and thus, recognizing the image variations is significant to understanding the dataset.…”

Section: Plant Disease Datasetsmentioning

confidence: 99%

See 1 more Smart Citation

Transfer learning for versatile plant disease recognition with limited data

Yoon

Jeong

et al. 2022

Front. Plant Sci.

View full text Add to dashboard Cite

Deep learning has witnessed a significant improvement in recent years to recognize plant diseases by observing their corresponding images. To have a decent performance, current deep learning models tend to require a large-scale dataset. However, collecting a dataset is expensive and time-consuming. Hence, the limited data is one of the main challenges to getting the desired recognition accuracy. Although transfer learning is heavily discussed and verified as an effective and efficient method to mitigate the challenge, most proposed methods focus on one or two specific datasets. In this paper, we propose a novel transfer learning strategy to have a high performance for versatile plant disease recognition, on multiple plant disease datasets. Our transfer learning strategy differs from the current popular one due to the following factors. First, PlantCLEF2022, a large-scale dataset related to plants with 2,885,052 images and 80,000 classes, is utilized to pre-train a model. Second, we adopt a vision transformer (ViT) model, instead of a convolution neural network. Third, the ViT model undergoes transfer learning twice to save computations. Fourth, the model is first pre-trained in ImageNet with a self-supervised loss function and with a supervised loss function in PlantCLEF2022. We apply our method to 12 plant disease datasets and the experimental results suggest that our method surpasses the popular one by a clear margin for different dataset settings. Specifically, our proposed method achieves a mean testing accuracy of 86.29over the 12 datasets in a 20-shot case, 12.76 higher than the current state-of-the-art method’s accuracy of 73.53. Furthermore, our method outperforms other methods in one plant growth stage prediction and the one weed recognition dataset. To encourage the community and related applications, we have made public our codes and pre-trained model1.

show abstract

Section: Discussionmentioning

confidence: 86%

Section: Plant Disease Datasetsmentioning

confidence: 99%

Transfer learning for versatile plant disease recognition with limited data

Yoon

Jeong

et al. 2022

Front. Plant Sci.

View full text Add to dashboard Cite

show abstract

“…This survey is an attempt to provide a structured and broad overview of the recent works on DDA, spanning many research and application domains. Previous surveys, Shorten et al [35] and Xu et al [36], have comprehensively investigated augmentation techniques, including traditional augmentation transformations, GAN-based synthetic methods, and automatic augmentation algorithms. Yang et al [37] further elaborated automatic augmentation methods in details.…”

Section: Introductionmentioning

confidence: 99%

Differentiable Image Data Augmentation and Its Applications: A Survey

Shi¹,

Ghazzai²,

Massoud³

2023

Preprint

View full text Add to dashboard Cite

<p>Data augmentation is an effective method to improve model robustness and generalization. Conventional routines of data augmentation pipeline are used as a preprocessing modules for neural networks with predefined heuristics and restricted differentiability. Some recent works indicated that the differentiable data augmentation (DDA) could effectively contribute to the training of neural networks and the augmentation policy searching strategies. This survey provides a comprehensive and structured overview of the advances in DDA, with a focus on fundamental elements including differentiable operations, operation relaxations, and gradient estimations. We then categorize existing DDA works accordingly. Afterwards, the survey investigates the utilization of DDA in selected of practical applications, specifically neural augmentation and differentiable augmentation search. Finally, we discuss current challenges of DDA and some promising future research directions.</p>

show abstract

“…Second, a large-scale dataset is entailed for deep learning-based algorithms in the training process to obtain competitive performance, however, collecting data is time-consuming and expensive in most applications and more inconvenient for plant growth prediction as the time-series character. Although many data augmentation methods have been proposed and verified to address this challenge (DeVries and Taylor, 2017 ; Zhang et al, 2018 ; Yun et al, 2019 ; Xu et al, 2022 ), the time-series data augmentation algorithm seems underdeveloped. Since plants grow in three-dimensional space over time, three key points can be considered to do data augmentation for plant growth prediction.…”

Section: Introductionmentioning

confidence: 99%

“…Two popular non-time-series methods, Cutout (DeVries and Taylor, 2017 ) and Cutmix (Yun et al, 2019 ), conflict with this requirement in that they may split one leaf and spatially combine two leaves. Third, useful variations are embraced to make the trained model robust (Xu et al, 2022 ), such as different backgrounds, locations of leaves, and relative positions among leaves. Embracing the three points, we propose two time-series data augmentation, time-series Mixup (T-Mixup) and time-series Copy-Paste (T-Copy-Paste) based on Zhang et al ( 2018 ), Ghiasi et al ( 2021 ).…”

Section: Introductionmentioning

confidence: 99%

Flexible and high quality plant growth prediction with limited data

Yoon

et al. 2022

Front. Plant Sci.

Self Cite

View full text Add to dashboard Cite

Predicting plant growth is a fundamental challenge that can be employed to analyze plants and further make decisions to have healthy plants with high yields. Deep learning has recently been showing its potential to address this challenge in recent years, however, there are still two issues. First, image-based plant growth prediction is currently taken either from time series or image generation viewpoints, resulting in a flexible learning framework and clear predictions, respectively. Second, deep learning-based algorithms are notorious to require a large-scale dataset to obtain a competing performance but collecting enough data is time-consuming and expensive. To address the issues, we consider the plant growth prediction from both viewpoints with two new time-series data augmentation algorithms. To be more specific, we raise a new framework with a length-changeable time-series processing unit to generate images flexibly. A generative adversarial loss is utilized to optimize our model to obtain high-quality images. Furthermore, we first recognize three key points to perform time-series data augmentation and then put forward T-Mixup and T-Copy-Paste. T-Mixup fuses images from a different time pixel-wise while T-Copy-Paste makes new time-series images with a different background by reusing individual leaves extracted from the existing dataset. We perform our method in a public dataset and achieve superior results, such as the generated RGB images and instance masks securing an average PSNR of 27.53 and 27.62, respectively, compared to the previously best 26.55 and 26.92.

show abstract

A Comprehensive Survey of Image Augmentation Techniques for Deep Learning

Cited by 5 publications

References 70 publications

Transfer learning for versatile plant disease recognition with limited data

Transfer learning for versatile plant disease recognition with limited data

Differentiable Image Data Augmentation and Its Applications: A Survey

Flexible and high quality plant growth prediction with limited data

Contact Info

Product

Resources

About