Matches in SemOpenAlex for { <https://semopenalex.org/work/W4297797598> ?p ?o ?g. }
Showing items 1 to 82 of
82
with 100 items per page.
- W4297797598 abstract "Contemporary Artificial Intelligence technologies allow for the employment of Computer Vision to discern good crops from bad, providing a step in the pipeline of selecting healthy fruit from undesirable fruit, such as those which are mouldy or gangrenous. State-of-the-art works in the field report high accuracy results on small datasets (<1000 images), which are not representative of the population regarding real-world usage. The goals of this study are to further enable real-world usage by improving generalisation with data augmentation as well as to reduce overfitting and energy usage through model pruning. In this work, we suggest a machine learning pipeline that combines the ideas of fine-tuning, transfer learning, and generative model-based training data augmentation towards improving fruit quality image classification. A linear network topology search is performed to tune a VGG16 lemon quality classification model using a publicly-available dataset of 2690 images. We find that appending a 4096 neuron fully connected layer to the convolutional layers leads to an image classification accuracy of 83.77%. We then train a Conditional Generative Adversarial Network on the training data for 2000 epochs, and it learns to generate relatively realistic images. Grad-CAM analysis of the model trained on real photographs shows that the synthetic images can exhibit classifiable characteristics such as shape, mould, and gangrene. A higher image classification accuracy of 88.75% is then attained by augmenting the training with synthetic images, arguing that Conditional Generative Adversarial Networks have the ability to produce new data to alleviate issues of data scarcity. Finally, model pruning is performed via polynomial decay, where we find that the Conditional GAN-augmented classification network can retain 81.16% classification accuracy when compressed to 50% of its original size." @default.
- W4297797598 created "2022-10-01" @default.
- W4297797598 creator A5032454100 @default.
- W4297797598 creator A5076983719 @default.
- W4297797598 creator A5078013087 @default.
- W4297797598 creator A5080362193 @default.
- W4297797598 creator A5089453046 @default.
- W4297797598 date "2021-04-12" @default.
- W4297797598 modified "2023-09-30" @default.
- W4297797598 title "Fruit Quality and Defect Image Classification with Conditional GAN Data Augmentation" @default.
- W4297797598 doi "https://doi.org/10.48550/arxiv.2104.05647" @default.
- W4297797598 hasPublicationYear "2021" @default.
- W4297797598 type Work @default.
- W4297797598 citedByCount "0" @default.
- W4297797598 crossrefType "posted-content" @default.
- W4297797598 hasAuthorship W4297797598A5032454100 @default.
- W4297797598 hasAuthorship W4297797598A5076983719 @default.
- W4297797598 hasAuthorship W4297797598A5078013087 @default.
- W4297797598 hasAuthorship W4297797598A5080362193 @default.
- W4297797598 hasAuthorship W4297797598A5089453046 @default.
- W4297797598 hasBestOaLocation W42977975981 @default.
- W4297797598 hasConcept C108010975 @default.
- W4297797598 hasConcept C111472728 @default.
- W4297797598 hasConcept C115961682 @default.
- W4297797598 hasConcept C119857082 @default.
- W4297797598 hasConcept C138885662 @default.
- W4297797598 hasConcept C144024400 @default.
- W4297797598 hasConcept C149923435 @default.
- W4297797598 hasConcept C150899416 @default.
- W4297797598 hasConcept C153180895 @default.
- W4297797598 hasConcept C154945302 @default.
- W4297797598 hasConcept C199360897 @default.
- W4297797598 hasConcept C202444582 @default.
- W4297797598 hasConcept C22019652 @default.
- W4297797598 hasConcept C2779530757 @default.
- W4297797598 hasConcept C2908647359 @default.
- W4297797598 hasConcept C33923547 @default.
- W4297797598 hasConcept C41008148 @default.
- W4297797598 hasConcept C43521106 @default.
- W4297797598 hasConcept C50644808 @default.
- W4297797598 hasConcept C6557445 @default.
- W4297797598 hasConcept C86803240 @default.
- W4297797598 hasConcept C9652623 @default.
- W4297797598 hasConceptScore W4297797598C108010975 @default.
- W4297797598 hasConceptScore W4297797598C111472728 @default.
- W4297797598 hasConceptScore W4297797598C115961682 @default.
- W4297797598 hasConceptScore W4297797598C119857082 @default.
- W4297797598 hasConceptScore W4297797598C138885662 @default.
- W4297797598 hasConceptScore W4297797598C144024400 @default.
- W4297797598 hasConceptScore W4297797598C149923435 @default.
- W4297797598 hasConceptScore W4297797598C150899416 @default.
- W4297797598 hasConceptScore W4297797598C153180895 @default.
- W4297797598 hasConceptScore W4297797598C154945302 @default.
- W4297797598 hasConceptScore W4297797598C199360897 @default.
- W4297797598 hasConceptScore W4297797598C202444582 @default.
- W4297797598 hasConceptScore W4297797598C22019652 @default.
- W4297797598 hasConceptScore W4297797598C2779530757 @default.
- W4297797598 hasConceptScore W4297797598C2908647359 @default.
- W4297797598 hasConceptScore W4297797598C33923547 @default.
- W4297797598 hasConceptScore W4297797598C41008148 @default.
- W4297797598 hasConceptScore W4297797598C43521106 @default.
- W4297797598 hasConceptScore W4297797598C50644808 @default.
- W4297797598 hasConceptScore W4297797598C6557445 @default.
- W4297797598 hasConceptScore W4297797598C86803240 @default.
- W4297797598 hasConceptScore W4297797598C9652623 @default.
- W4297797598 hasLocation W42977975981 @default.
- W4297797598 hasLocation W42977975982 @default.
- W4297797598 hasOpenAccess W4297797598 @default.
- W4297797598 hasPrimaryLocation W42977975981 @default.
- W4297797598 hasRelatedWork W2942629287 @default.
- W4297797598 hasRelatedWork W2989932438 @default.
- W4297797598 hasRelatedWork W3012393889 @default.
- W4297797598 hasRelatedWork W3099765033 @default.
- W4297797598 hasRelatedWork W3175189414 @default.
- W4297797598 hasRelatedWork W3186919929 @default.
- W4297797598 hasRelatedWork W4254952533 @default.
- W4297797598 hasRelatedWork W4304128395 @default.
- W4297797598 hasRelatedWork W4313289428 @default.
- W4297797598 hasRelatedWork W4320737025 @default.
- W4297797598 isParatext "false" @default.
- W4297797598 isRetracted "false" @default.
- W4297797598 workType "article" @default.