A computer vision based machine learning approach for fatigue crack initiation sites recognition

Published on Jan 1, 2020in Computational Materials Science2.863
· DOI :10.1016/J.COMMATSCI.2019.109259
Shi Yuan Wang2
Estimated H-index: 2
(NUAA: Nanjing University of Aeronautics and Astronautics),
Pingze Zhang18
Estimated H-index: 18
(NUAA: Nanjing University of Aeronautics and Astronautics)
+ 3 AuthorsFengkun Li3
Estimated H-index: 3
(NUAA: Nanjing University of Aeronautics and Astronautics)
Abstract Fatigue crack initiation sites (FCISs) are crucial for fatigue investigations in metallic compounds, since they are closely related to the fatigue failure mechanisms and inform methods against crack initiation. However, the identification of FCISs requires professional knowledge and can be demanding and time consuming. Computer vision techniques show excellent performance for image texture recognition tasks and have been applied to some material domains. In order to explore more possibilities and widen the applications of computer vision techniques on material domains, as well as reduce the labor and knowledge requirements for FCISs recognition tasks, a state-of-the-art object detector – deeply supervised object detector which can train models from scratch, was introduced. The models trained from limited data show reasonable capacity for recognizing FCISs. Results demonstrate that increasing training dataset size can improve the accuracies of the models, while raising the number of epochs can result in a superior ability to recognize delicate features. Moreover, most of the images that cannot be recognized for FCISs possess common characteristics, such as, poor image quality, unclear features, insufficient training data and etc. Based on the above findings, several proposals are made for the future work to improve the performance of the same or relevant studies, as a reference.
📖 Papers frequently viewed together
4 Citations
1 Author (Jure Kovač)
1 Citations
Oct 8, 2016 in ECCV (European Conference on Computer Vision)
#1Wei Liu (UNC: University of North Carolina at Chapel Hill)H-Index: 17
#2Dragomir AnguelovH-Index: 31
Last. Alexander C. Berg (UNC: University of North Carolina at Chapel Hill)H-Index: 55
view all 7 authors...
We present a method for detecting objects in images using a single deep neural network. Our approach, named SSD, discretizes the output space of bounding boxes into a set of default boxes over different aspect ratios and scales per feature map location. At prediction time, the network generates scores for the presence of each object category in each default box and produces adjustments to the box to better match the object shape. Additionally, the network combines predictions from multiple featu...
10.3k CitationsSource
#1Aritra Chowdhury (RPI: Rensselaer Polytechnic Institute)H-Index: 5
#2Elizabeth J. Kautz (RPI: Rensselaer Polytechnic Institute)H-Index: 8
Last. Daniel Lewis (RPI: Rensselaer Polytechnic Institute)H-Index: 11
view all 4 authors...
Abstract Computer vision and machine learning methods were applied to the challenge of automatic microstructure recognition. Here, a case study on dendritic morphologies was performed. Two classification tasks were completed, and involved distinguishing between micrographs that depict dendritic morphologies from those that do not contain this particular microstructural feature (Task 1), and from those micrographs identified as depicting dendrites, different cross-sectional views (longitudinal or...
131 CitationsSource
Jun 27, 2016 in CVPR (Computer Vision and Pattern Recognition)
#1Joseph Redmon (UW: University of Washington)H-Index: 12
#2Santosh K. Divvala (Allen Institute for Artificial Intelligence)H-Index: 14
Last. Ali Farhadi (UW: University of Washington)H-Index: 57
view all 4 authors...
We present YOLO, a new approach to object detection. Prior work on object detection repurposes classifiers to perform detection. Instead, we frame object detection as a regression problem to spatially separated bounding boxes and associated class probabilities. A single neural network predicts bounding boxes and class probabilities directly from full images in one evaluation. Since the whole detection pipeline is a single network, it can be optimized end-to-end directly on detection performance....
9,872 CitationsSource
#1Brian L. DeCost (CMU: Carnegie Mellon University)H-Index: 13
#2Elizabeth A. Holm (CMU: Carnegie Mellon University)H-Index: 36
Abstract The ‘bag of visual features’ image representation was applied to create generic microstructural signatures that can be used to automatically find relationships in large and diverse microstructural image data sets. Using this representation, a support vector machine (SVM) was trained to classify microstructures into one of seven groups with greater than 80% accuracy over 5-fold cross validation. In addition, the bag of visual features was implemented as the basis for a visual search engi...
152 CitationsSource
#1Shuiwang Ji (ODU: Old Dominion University)H-Index: 43
#2Wei Xu (Facebook)H-Index: 58
Last. Kai Yu (Baidu)H-Index: 59
view all 4 authors...
We consider the automated recognition of human actions in surveillance videos. Most current methods build classifiers based on complex handcrafted features computed from the raw inputs. Convolutional neural networks (CNNs) are a type of deep model that can act directly on the raw inputs. However, such models are currently limited to handling 2D inputs. In this paper, we develop a novel 3D CNN model for action recognition. This model extracts features from both the spatial and the temporal dimens...
3,512 CitationsSource
Dec 3, 2012 in NeurIPS (Neural Information Processing Systems)
#1Alex Krizhevsky (U of T: University of Toronto)H-Index: 16
#2Ilya Sutskever (U of T: University of Toronto)H-Index: 72
Last. Geoffrey E. Hinton (U of T: University of Toronto)H-Index: 157
view all 3 authors...
We trained a large, deep convolutional neural network to classify the 1.2 million high-resolution images in the ImageNet LSVRC-2010 contest into the 1000 different classes. On the test data, we achieved top-1 and top-5 error rates of 37.5% and 17.0% which is considerably better than the previous state-of-the-art. The neural network, which has 60 million parameters and 650,000 neurons, consists of five convolutional layers, some of which are followed by max-pooling layers, and three fully-connect...
44.6k Citations
#1Vijay Garla (Yale University)H-Index: 7
#2Cynthia Brandt (Yale University)H-Index: 41
Graphical abstractDisplay Omitted Highlights? Leveraged the structure of the UMLS to rank concepts for text classification. ? Used document class labels to define a context-sensitive semantic similarity measure. ? Classified clinical text with SVMs using a semantic kernel. ? Improved the top machine-learning system from I2B2 2008 challenge. In this study we present novel feature engineering techniques that leverage the biomedical domain knowledge encoded in the Unified Medical Language System (U...
60 CitationsSource
#1Sinno Jialin Pan (HKUST: Hong Kong University of Science and Technology)H-Index: 37
#2Qiang Yang (HKUST: Hong Kong University of Science and Technology)H-Index: 121
A major assumption in many machine learning and data mining algorithms is that the training and future data must be in the same feature space and have the same distribution. However, in many real-world applications, this assumption may not hold. For example, we sometimes have a classification task in one domain of interest, but we only have sufficient training data in another domain of interest, where the latter data may be in a different feature space or follow a different data distribution. In...
10.5k CitationsSource
#1M. Barrabés (ETSEIB)H-Index: 1
#2Pablo Sevilla (ETSEIB)H-Index: 11
Last. Francisco Javier Gil (ETSEIB)H-Index: 40
view all 4 authors...
Abstract Porous NiTi materials were obtained in order to allow the ingrowth of living tissue, thus increasing the mechanical anchorage of implants. This promotes long-term fixation without the need for bone grafting. This study focuses mainly on the determination of mechanical properties: compression, bending and compressive fatigue. The foams obtained had an appropriate range of pore sizes and interconnectivity, which enabled a morphology similar to that of bone to be achieved. Their mechanical...
65 CitationsSource
Most of us have been exposed to fatigue failures since we first started looking at broken pieces. In many cases, the first explanation of a failure began with a well-intentioned person explaining that a component had “crystallized” because it was a piece of defective steel. However, as we know, virtually all structural metals are crystalline, and after working with fatigue analysis for a while, it becomes apparent that defective materials are not common failure causes. Additionally we learn that...
72 CitationsSource
Cited By4
#2Yao ChenH-Index: 18
Last. Pooya SarehH-Index: 12
view all 7 authors...
#1Lun Zhao (CAS: Chinese Academy of Sciences)
#2Yunlong Pan (Kunming University of Science and Technology)
Last. Shafiqul Islam (BTH: Blekinge Institute of Technology)H-Index: 1
view all 5 authors...
The scanning electron microscope (SEM) is widely used in the analysis and research of materials, including fracture analysis, microstructure morphology, and nanomaterial analysis. With the rapid development of materials science and computer vision technology, the level of detection technology is constantly improving. In this paper, the deep learning method is used to intelligently identify microcracks in the microscopic morphology of SEM image. A deep learning model based on image level is selec...
#1Susheel Dharmadhikari (PSU: Pennsylvania State University)H-Index: 1
#2Eric Keller (PSU: Pennsylvania State University)H-Index: 10
Last. Amrita Basak (PSU: Pennsylvania State University)H-Index: 1
view all 4 authors...
Abstract This paper presents a dual-imaging methodology to investigate fatigue crack initiation & propagation along the lateral and transverse directions in Al7075 flat specimens having a one-sided V-notch. The framework consists of a confocal and a digital microscope. The confocal microscope can detect the short crack initiation at a length-scale of 3 micron while the digital microscope has the capability of reliably measuring cracks at a scale of 0.5 millimeter. The apparatus provides a unique...
#1Ivan KuricH-Index: 11
#1Ivan KuricH-Index: 2
Last. Dariusz Więcek (University of Bielsko-Biała)H-Index: 3
view all 5 authors...
Nowadays, when high industrial productivity is connected with high quality and low product faults, it is common practice to use 100% product quality control. Since the quantities of products are high in mass production and inspection time must be as low as possible, the solution may be to use visual inspection of finished parts via camera systems and subsequent image processing using artificial intelligence. Recently, deep learning has shown itself to be the most appropriate and effective method...
5 CitationsSource