Lychee Surface Defect Detection Based on Deep Convolutional Neural Networks with GAN-Based Data Augmentation

Wang, Chenglong; Xiao, Zhifeng

doi:10.3390/agronomy11081500

Cited by 43 publications

(27 citation statements)

References 37 publications

(16 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These architectures do not represent the full landscape of GANs in literature, but they have open-sourced packages and have been employed or adapted for expanding agricultural image data, as reviewed in Section 5. In addition to the tabular summary, readers are referred to dedicated literature (Wang and Xiao, 2021;Huang et al, 2018;Creswell et al, 2018) and "The GAN Zoo" in the Github (https://github.com/ hindupuravinash/the-gan-zoo) for a more comprehensive list of GAN variants as well as a GAN toolbox in the PyTorch environment at https://github. com/facebookresearch/pytorch_GAN_zoo for algorithm implementation.…”

Section: Discussionmentioning

confidence: 99%

“…Transformer-based GANs have been proposed recently for improving the state of the art in image generation (Jiang et al, 2021;Hudson and Zitnick, 2021;Zhao et al, 2021a). Wang and Xiao (2021) reported on using Trans-GAN (Jiang et al, 2021) (Section 4.11) to generate lychee images for surface defect detection. Images were preferentially generated for two defect classes to rebalance training data, and the augmented-data achieved improvements of 0.58%-2.86% in mAP for defect detection, depending on object detectors, compared to training without GAN-based data augmentation.…”

Section: Postharvest Quality Assessmentmentioning

confidence: 99%

See 1 more Smart Citation

Generative Adversarial Networks for Image Augmentation in Agriculture: A Systematic Review

Olaniyi¹,

Chen²,

Lu³

et al. 2022

Preprint

View full text Add to dashboard Cite

In agricultural image analysis, optimal model performance is keenly pursued for better fulfilling visual recognition tasks (e.g., image classification, segmentation, object detection and localization), in the presence of challenges with biological variability and unstructured environments. Large-scale, balanced and ground-truthed image datasets, however, are often difficult to obtain to fuel the development of advanced, high-performance models. As artificial intelligence through deep learning is impacting analysis and modeling of agricultural images, data augmentation plays a crucial role in boosting model performance while reducing manual efforts for data preparation, by algorithmically expanding training datasets. Beyond traditional data augmentation techniques, generative adversarial network (GAN) invented in 2014 in the computer vision community, provides a suite of novel approaches that can learn good data representations and generate highly realistic samples. Since 2017, there has been a growth of research into GANs for image augmentation or synthesis in agriculture for improved model performance. This paper presents an overview of the evolution of GAN architectures followed by a systematic review of their application to agriculture (https://github.com/Derekabc/GANs-Agriculture), involving various vision tasks for plant health, weeds, fruits, aquaculture, animal farming, plant phenotyping as well as postharvest inspection of fruits. Challenges and opportunities of GANs are discussed for future research.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Postharvest Quality Assessmentmentioning

confidence: 99%

Generative Adversarial Networks for Image Augmentation in Agriculture: A Systematic Review

Olaniyi¹,

Chen²,

Lu³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Characteristic for CNNs is the extraction of features from raw data input with specific patterns, without having to apply manual feature design. Within the recent years, DL has found its ways into agricultural studies for which fruit detection, ripeness classification and detection of diseases has been of great interest [30][31][32]. Sa et al [33] developed a fruit grading system by evaluating the fruit ripeness by surface defects using hyperspectral imaging.…”

Section: Introductionmentioning

confidence: 99%

Interactive Deep Learning for Shelf Life Prediction of Muskmelons Based on an Active Learning Approach

Albert-Weiss

Osman

2022

Sensors

View full text Add to dashboard Cite

A pivotal topic in agriculture and food monitoring is the assessment of the quality and ripeness of agricultural products by using non-destructive testing techniques. Acoustic testing offers a rapid in situ analysis of the state of the agricultural good, obtaining global information of its interior. While deep learning (DL) methods have outperformed state-of-the-art benchmarks in various applications, the reason for lacking adaptation of DL algorithms such as convolutional neural networks (CNNs) can be traced back to its high data inefficiency and the absence of annotated data. Active learning is a framework that has been heavily used in machine learning when the labelled instances are scarce or cumbersome to obtain. This is specifically of interest when the DL algorithm is highly uncertain about the label of an instance. By allowing the human-in-the-loop for guidance, a continuous improvement of the DL algorithm based on a sample efficient manner can be obtained. This paper seeks to study the applicability of active learning when grading ‘Galia’ muskmelons based on its shelf life. We propose k-Determinantal Point Processes (k-DPP), which is a purely diversity-based method that allows to take influence on the exploration within the feature space based on the chosen subset k. While getting coequal results to uncertainty-based approaches when k is large, we simultaneously obtain a better exploration of the data distribution. While the implementation based on eigendecomposition takes up a runtime of O(n3), this can further be reduced to O(n·poly(k)) based on rejection sampling. We suggest the use of diversity-based acquisition when only a few labelled samples are available, allowing for better exploration while counteracting the disadvantage of missing the training objective in uncertainty-based methods following a greedy fashion.

show abstract

“…The rapid identification of ginger shoots and seeds through edge computing devices (ECD) are essential to promote the automation of ginger sowing machines and improve ginger yield and the economic efficiency of ginger farmers. In recent years, with the rapid development of deep learning technology, object detection has been widely used in the biosystems engineering domain [2][3][4][5][6][7]. It overcomes the insufficient representation capability of traditional machine vision and automatically extracts image feature information by building deep convolutional neural networks (CNN) [8].…”

Section: Introductionmentioning

confidence: 99%

Using Channel and Network Layer Pruning Based on Deep Learning for Real-Time Detection of Ginger Images

Fang

et al. 2021

Agriculture

View full text Add to dashboard Cite

Consistent ginger shoot orientation helps to ensure consistent ginger emergence and meet shading requirements. YOLO v3 is used to recognize ginger images in response to the current ginger seeder’s difficulty in meeting the above agronomic problems. However, it is not suitable for direct application on edge computing devices due to its high computational cost. To make the network more compact and to address the problems of low detection accuracy and long inference time, this study proposes an improved YOLO v3 model, in which some redundant channels and network layers are pruned to achieve real-time determination of ginger shoots and seeds. The test results showed that the pruned model reduced its model size by 87.2% and improved the detection speed by 85%. Meanwhile, its mean average precision (mAP) reached 98.0% for ginger shoots and seeds, only 0.1% lower than the model before pruning. Moreover, after deploying the model to the Jetson Nano, the test results showed that its mAP was 97.94%, the recognition accuracy could reach 96.7%, and detection speed could reach 20 frames·s−1. The results showed that the proposed method was feasible for real-time and accurate detection of ginger images, providing a solid foundation for automatic and accurate ginger seeding.

show abstract

Lychee Surface Defect Detection Based on Deep Convolutional Neural Networks with GAN-Based Data Augmentation

Cited by 43 publications

References 37 publications

Generative Adversarial Networks for Image Augmentation in Agriculture: A Systematic Review

Generative Adversarial Networks for Image Augmentation in Agriculture: A Systematic Review

Interactive Deep Learning for Shelf Life Prediction of Muskmelons Based on an Active Learning Approach

Using Channel and Network Layer Pruning Based on Deep Learning for Real-Time Detection of Ginger Images

Contact Info

Product

Resources

About