Analysis on DeepLabV3+ Performance for Automatic Steel Defects Detection

2020-04-09 · Zheng Nie, Jiachen Xu, Shengchang Zhang

Our works experimented DeepLabV3+ with different backbones on a large volume of steel images aiming to automatically detect different types of steel defects. Our methods applied random weighted augmentation to balance different defects types in the training set. And then applied DeeplabV3+ model three different backbones, ResNet, DenseNet and EfficientNet, on segmenting defection regions on the steel images. Based on experiments, we found that applying ResNet101 or EfficientNet as backbones could reach the best IoU scores on the test set, which is around 0.57, comparing with 0.325 for using DenseNet. Also, DeepLabV3+ model with ResNet101 as backbone has the fewest training time.

📄 PDF Abstract BibTeX arXiv:2004.04822

Code (1)

code-implementation1/Code9/tree/main/DeepLabv3 mindspore

Methods 이 논문이 사용한 방법론

RMSProp RMSProp is an unpublished adaptive learning rate optimizer proposed by Geoff Hinton. The motivation…

Bottleneck Residual Block A Bottleneck Residual Block is a variant of the residual block that utilises 1x1 convolutions to create a bottleneck. The…

Residual Connection 설명 없음

Residual Block Residual Blocks are skip-connection blocks that learn residual functions with reference to the layer inputs, instead of learning unreferenced functions. They were introduced…

Concatenated Skip Connection A Concatenated Skip Connection is a type of skip connection that seeks to reuse features by concatenating them to new layers, allowing more information to be retained from…

Bitcoin Customer Service Number +1-833-534-1729 설명 없음

Global Average Pooling Global Average Pooling is a pooling operation designed to replace fully connected layers in classical CNNs. The idea is to generate one feature map for each corresponding…

Dense Block A Dense Block is a module used in convolutional neural networks that connects *all layers* (with matching feature-map sizes) directly with each other. It was originally…

A deep-learning-based approach for fast and robust steel surface defects classification

2019-05-15 · elsevier journal 2019 5 · Guizhong Fu, Peize Sun a, Wenbin Zhu, Jiangxin Yang 외

Automatic visual recognition of steel surface defects provides critical functionality to facilitate quality control of steel strip production. In this paper, we present a compact yet effective convolutional neural networ…

Diversity General Classification

Recognition Of Surface Defects On Steel Sheet Using Transfer Learning

2019-09-07 · Jingwen Fu, Xiaoyan Zhu, Yingbin Li

Automatic defect recognition is one of the research hotspots in steel production, but most of the current methods mainly extract features manually and use machine learning classifiers to recognize defects, which cannot t…

Transfer Learning

Method of detecting the defects of the end face of steel coils

2019-10-16 · Seventh International Conference on Optical and Photonic Engineering (icOPEN 2019), Phuket, Thailand 2019 10 · Liting Zhu, Kinbo Chen, Jinjun Rao, Mei Liu

Due to the excessively dense steel coil layers on the end face of the rolled coil, it is difficult to extract the defect area of the end face of rolled steel by the image segmentation methods such as edge detection and t…

Edge Detection Image Segmentation Semantic Segmentation

Uneven illumination surface defects inspection based on convolutional neural network

2019-05-16 · Hao Wu, Yulong Liu, Wenbin Gao, Xiangrong Xu

Surface defect inspection based on machine vision is often affected by uneven illumination. In order to improve the inspection rate of surface defects inspection under uneven illumination condition, this paper proposes a…

Deep Learning Based Steel Pipe Weld Defect Detection

2021-04-30 · Dingming Yang, Yanrong Cui, Zeyu Yu, Hongqiang Yuan

Steel pipes are widely used in high-risk and high-pressure scenarios such as oil, chemical, natural gas, shale gas, etc. If there is some defect in steel pipes, it will lead to serious adverse consequences. Applying obje…

Deep Learning Defect Detection Object object-detection+1