Publication:
Object Detection Model Training Framework for Very Small Datasets Applied to Outdoor Industrial Structures

Simple item page
dc.citedby1
dc.contributor.authorBaharuddin M.Z.en_US
dc.contributor.authorHow D.N.T.en_US
dc.contributor.authorSahari K.S.M.en_US
dc.contributor.authorAbas A.Z.en_US
dc.contributor.authorRamlee M.K.en_US
dc.contributor.authorid35329255600en_US
dc.contributor.authorid57212923888en_US
dc.contributor.authorid57218170038en_US
dc.contributor.authorid57361520800en_US
dc.contributor.authorid57361480700en_US
dc.date.accessioned2023-05-29T09:10:32Z
dc.date.available2023-05-29T09:10:32Z
dc.date.issued2021
dc.descriptionComputer vision; Convolutional neural networks; Deep neural networks; Digital devices; Feature extraction; Image enhancement; Inspection; Object recognition; Convolutional neural network; Deep learning; Detection models; Industrial inspections; Industrial structures; Model training; Objects detection; Retinanet; Small data set; Training framework; Object detectionen_US
dc.description.abstractVisual inspection of electrical utility assets is crucial in ensuring the continuous operation of a system or plant. With the advent of digital imagery using mobile devices, it has become easy to collect a vast amount of asset pictures from sites. To further enhance inspection efficiency, we propose RetinaNet, a deep learning-based object detection model that can be trained to automatically detect specific objects and features from images of outdoor industrial structures. The model is capable of detecting features such as intrusions, tree or bushes in the vicinity of the lattice towers. We also introduce a model training framework for use with very small datasets which consists of rigorous data augmentation, image pre-sizing, focal loss function, progressive resizing, learning rate finder, and the Ranger optimizer. Experiment results show that the proposed model used in conjunction with the aforementioned training framework results in the lowest validation loss and highest mean average precision of 31.36 � 2021, Springer Nature Switzerland AG.en_US
dc.description.natureFinalen_US
dc.identifier.doi10.1007/978-3-030-90235-3_47
dc.identifier.epage551
dc.identifier.scopus2-s2.0-85120535591
dc.identifier.spage540
dc.identifier.urihttps://www.scopus.com/inward/record.uri?eid=2-s2.0-85120535591&doi=10.1007%2f978-3-030-90235-3_47&partnerID=40&md5=79270ebc59d894f4345b18d068536a90
dc.identifier.urihttps://irepository.uniten.edu.my/handle/123456789/26439
dc.identifier.volume13051 LNCS
dc.publisherSpringer Science and Business Media Deutschland GmbHen_US
dc.sourceScopus
dc.sourcetitleLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
dc.titleObject Detection Model Training Framework for Very Small Datasets Applied to Outdoor Industrial Structuresen_US
dc.typeConference Paperen_US
dspace.entity.typePublication
Files
Collections
SCOPUS