Automated Defect Detection From Ultrasonic Images Using Deep Learning

Medak, Duje; Posilović, Luka; Subašić, Marko; Budimir, Marko; Lončarić, Sven

doi:10.1109/tuffc.2021.3081750

Cited by 66 publications

(17 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Even though Medak et al, in [18] agree that object detection algorithms require large amount of data to provide humanlevel accuracy, they prove EfficientDet to be able to perform SOTA results on realistic performance in Ultrasonic and Forensics defect detection. They introduce a novel anchors (sliding window) size finding mechanism for OSOD, a kind of hyperparameter search.…”

Section: 3-automated Defect Detection: Modifying Backbonementioning

confidence: 99%

Underwater Object Detection Based on Improved EfficientDet

Jia

et al. 2022

Remote Sensing

View full text Add to dashboard Cite

Intelligent detection of marine organism plays an important part in the marine economy, and it is significant to detect marine organisms quickly and accurately in a complex marine environment for the intelligence of marine equipment. The existing object detection models do not work well underwater. This paper improves the structure of EfficientDet detector and proposes the EfficientDet-Revised (EDR), which is a new marine organism object detection model. Specifically, the MBConvBlock is reconstructed by adding the Channel Shuffle module to enable the exchange of information between the channels of the feature layer. The fully connected layer of the attention module is removed and convolution is used to cut down the amount of network parameters. The Enhanced Feature Extraction module is constructed for multi-scale feature fusion to enhance the feature extraction ability of the network to different objects. The results of experiments demonstrate that the mean average precision (mAP) of the proposed method reaches 91.67% and 92.81% on the URPC dataset and the Kaggle dataset, respectively, which is better than other object detection models. At the same time, the processing speed reaches 37.5 frame per second (FPS) on the URPC dataset, which can meet the real-time requirements. It can provide a useful reference for underwater robots to perform tasks such as intelligent grasping.

show abstract

Section: 3-automated Defect Detection: Modifying Backbonementioning

confidence: 99%

Underwater Object Detection Based on Improved EfficientDet

Jia

et al. 2022

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…The results show a relatively high level of accuracy with the 92% defect recognition. Medak et al [42] proposed an automated defect detection network through a EfficientDet network (originally developed for object detection) that takes as input ultrasonic B-scan images. The proposed approach uses an image processing DL architecture (EfficientDet) that outperforms other DL models (YOLOv3 [43] and RetinaNet [44]) having better performance in new data, while still automating the defect detection.…”

Section: B-scan and Other Ultrasonic Imagesmentioning

confidence: 99%

Deep learning in automated ultrasonic NDE -- developments, axioms and opportunities

Cantero-Chinchilla,

Wilcox,

Croxford

2021

Preprint

View full text Add to dashboard Cite

The analysis of ultrasonic NDE data has traditionally been addressed by a trained operator manually interpreting data with the support of rudimentary automation tools. Recently, many demonstrations of deep learning (DL) techniques that address individual NDE tasks (data acquisition, data pre-processing, defect detection, and defect characterisation) have started to emerge in the research community. These methods have the potential to offer high flexibility, efficiency, and accuracy subject to the availability of sufficient training data; moreover, they enable the automation of complex processes that span one or more NDE steps (e.g. detection, characterisation, and sizing). There is, however, a lack of consensus on the direction and requirements that these new methods should follow. These elements are critical to help achieve full artificial intelligence driven automation of ultrasonic NDE so that the research community, industry, and regulatory bodies embrace them. This paper reviews the state-of-the-art of autonomous ultrasonic NDE enabled by DL methodologies. The review is organised by the NDE tasks that are addressed by means of DL approaches. Key remaining challenges for each task are noted. Basic axiomatic principles for DL methods in NDE are identified based on the literature review, relevant international regulations, and current industrial needs. By placing DL methods in the context of general NDE automation levels, this paper aims to provide a roadmap for future research and development in the area.

show abstract

“…Visual inspection [4]- [6] or more modern techniques such as ultrasonic testing [7], [8], magnetic particle inspection [9], [10], liquid penetration testing [11], and eddy current testing [12], which are non-destructive testing (NDT), can be used to examine railway track faults.…”

Section: Introductionmentioning

confidence: 99%

Fault Diagnosis in Railway Track using Efficient Net based CNN

Mordia¹,

Verma²

2022

YMER

View full text Add to dashboard Cite

Railway transportation is the most cost-effective and convenient means of passenger travel in India. The cress cross tracks are almost present in every part of India. Keeping in mind the security of people also the free running railway without any problem, we have to focus in the safety part of this system [1]. In India, the railway network accounts for over 80% of all transportation. Approximately 60% of accidents occur at crossings of railway tracks because of a railway track fracture, resulting in the loss of valuable life and economic damage. As a result, new technology is required for both fault detection and object detection in railway tracks. This technology must be resilient, efficient, and steady [2]. This paper presents a vision based method to find some common defects in railways. Some images have been collected of railway track and image processing method is used to preprocess these images and to detect the features related to defective parts. An EfficientNet based CNN model is developed to detect the defects which uses global average pooling, adam as optimizer, softmax as activation function and categorical_crossentropy as loss function. This research result consists of a classification report as defective and non-defective parts or image with accuracy of 91 percentages over 30 epochs. Keywords: Rail Track Images, Defects, Defect Detection Method, CNN and Defect Classification.

show abstract

Automated Defect Detection From Ultrasonic Images Using Deep Learning

Cited by 66 publications

References 23 publications

Underwater Object Detection Based on Improved EfficientDet

Underwater Object Detection Based on Improved EfficientDet

Deep learning in automated ultrasonic NDE -- developments, axioms and opportunities

Fault Diagnosis in Railway Track using Efficient Net based CNN

Contact Info

Product

Resources

About