Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userDefect-GAN: High-Fidelity Defect Synthesis for Automated Defect Inspection
360
0
0.0
(
0
)
Added by
Gongjie Zhang
Publication date
2021
fields
Informatics Engineering
and research's language is
English
Ask ChatGPT about the research
No Arabic abstract
Automated defect inspection is critical for effective and efficient maintenance, repair, and operations in advanced manufacturing. On the other hand, automated defect inspection is often constrained by the lack of defect samples, especially when we adopt deep neural networks for this task. This paper presents Defect-GAN, an automated defect synthesis network that generates realistic and diverse defect samples for training accurate and robust defect inspection networks. Defect-GAN learns through defacement and restoration processes, where the defacement generates defects on normal surface images while the restoration removes defects to generate normal images. It employs a novel compositional layer-based architecture for generating realistic defects within various image backgrounds with different textures and appearances. It can also mimic the stochastic variations of defects and offer flexible control over the locations and categories of the generated defects within the image background. Extensive experiments show that Defect-GAN is capable of synthesizing various defects with superior diversity and fidelity. In addition, the synthesized defect samples demonstrate their effectiveness in training better defect inspection networks.
rate research
Read More
Previous works for PCB defect detection based on image difference and image processing techniques have already achieved promising performance. However, they sometimes fall short because of the unaccounted defect patterns or over-sensitivity about some hyper-parameters. In this work, we design a deep model that accurately detects PCB defects from an input pair of a detect-free template and a defective tested image. A novel group pyramid pooling module is proposed to efficiently extract features of a large range of resolutions, which are merged by group to predict PCB defect of corresponding scales. To train the deep model, a dataset is established, namely DeepPCB, which contains 1,500 image pairs with annotations including positions of 6 common types of PCB defects. Experiment results validate the effectiveness and efficiency of the proposed model by achieving $98.6%$ mAP @ 62 FPS on DeepPCB dataset. This dataset is now available at: https://github.com/tangsanli5201/DeepPCB.
With the rapid development of intelligent detection algorithms based on deep learning, much progress has been made in automatic road defect recognition and road marking parsing. This can effectively address the issue of an expensive and time-consuming process for professional inspectors to review the street manually. Towards this goal, we present RoadAtlas, a novel end-to-end integrated system that can support 1) road defect detection, 2) road marking parsing, 3) a web-based dashboard for presenting and inputting data by users, and 4) a backend containing a well-structured database and developed APIs.
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.
Using first-principles calculations, we explored all the 21 defect-pairs in GaN and considered 6 configurations with different defect-defect distances for each defect-pair. 15 defect-pairs with short defect-defect distances are found to be stable during structural relaxation, so they can exist in the GaN lattice once formed during the irradiation of high-energy particles. 9 defect-pairs have formation energies lower than 10 eV in the neutral state. The vacancy-pair VN-VN is found to have very low formation energies, as low as 0 eV in p-type and Ga-rich GaN, and act as efficient donors producing two deep donor levels, which can limit the p-type doping and minority carrier lifetime in GaN. VN-VN has been overlooked in the previous study of defects in GaN. Most of these defect-pairs act as donors and produce a large number of defect levels in the band gap. Their formation energies and concentrations are sensitive to the chemical potentials of Ga and N, so their influences on the electrical and optical properties of Ga-rich and N-rich GaN after irradiation should differ significantly. These results about the defect-pairs provide fundamental data for understanding the radiation damage mechanism in GaN and simulating the defect formation and diffusion behavior under irradiation.
We discuss some general properties of defect branes, i.e. branes of co-dimension two, in (toroidally compactified) IIA/IIB string theory. In particular, we give a full classification of the supersymmetric defect branes in dimensions 2 < D < 11 as well as their higher-dimensionalstring and M-theory origin as branes and a set of generalized Kaluza-Klein monopoles. We point out a relation between the generalized Kaluza-Klein monopole solutions and a particular type of mixed-symmetry tensors. These mixed-symmetry tensors can be defined at the linearized level as duals of the supergravity potentials that describe propagating degrees of freedom. It is noted that the number of supersymmetric defect branes is always twice the number of corresponding central charges in the supersymmetry algebra.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
