Research on Recognition Technology of Aluminum Profile Surface Defects Based on Deep Learning

Wei, Ruofeng; Bi, Yunbo

doi:10.20944/preprints201904.0322.v1

Cited by 8 publications

(5 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To detect the surface type defects of the silicone rubber gaskets which have a rough and dim texture, a multi-exposure technique is used to enhance the illumination and highlight the defects. These multi-exposure images are then included in the dataset and trained in a 50-layer ResNet network [19][20]. The pixel size range of the defects is limited to avoid the feature vanishing during the convolution operations of the network and to front back allow for more accurate discernment between the background and the defects.…”

Section: Deep Learning Approachmentioning

confidence: 99%

Machine Vision and Deep Learning Based Rubber Gasket Defect Detection

et al. 2020

Adv. technol. innov.

View full text Add to dashboard Cite

This study develops an automated optical inspection system for silicone rubber gaskets using traditional rule-based and deep learning detection techniques. The specific object of interest is a 5 mm × 10 mm × 5 mm mobile device power supply connector gasket that provides protection against foreign body inclusion and water ingression. The proposed system can detect a total of five characteristic defects introduced during the mold-based manufacture process, which range from 10-100 μm. The deep learning detection strategies in this system employ convolutional neural networks (CNN) developed using the TensorFlow open-source library. Through both high dynamic range image capture and image generation techniques, accuracies of 100% and 97% are achieved for notch and residual glue defect predictions, respectively.

show abstract

Section: Deep Learning Approachmentioning

confidence: 99%

Machine Vision and Deep Learning Based Rubber Gasket Defect Detection

et al. 2020

Adv. technol. innov.

View full text Add to dashboard Cite

show abstract

“…The second category of surface defect recognition methods is based on an object detection network which provides object location and classification information. For instance, Wei and Bi [ 19 ] proposed a surface defect detection network based on Faster RCNN [ 20 ] to perform multi-scale detection on defects of various sizes and types on the surface of aluminum profiles. He et al [ 21 ] proposed a CNN-based surface defect detection approach, which uses a Multilevel-feature Fusion Network (MFN) to combine multiple hierarchical features into a multilevel feature.…”

Section: Literature Reviewmentioning

confidence: 99%

SurfNetv2: An Improved Real-Time SurfNet and Its Applications to Defect Recognition of Calcium Silicate Boards

Tsai

Chen

2020

Sensors

View full text Add to dashboard Cite

This paper presents an improved Convolutional Neural Network (CNN) architecture to recognize surface defects of the Calcium Silicate Board (CSB) using visual image information based on a deep learning approach. The proposed CNN architecture is inspired by the existing SurfNet architecture and is named SurfNetv2, which comprises a feature extraction module and a surface defect recognition module. The output of the system is the recognized defect category on the surface of the CSB. In the collection of the training dataset, we manually captured the defect images presented on the surface of the CSB samples. Then, we divided these defect images into four categories, which are crash, dirty, uneven, and normal. In the training stage, the proposed SurfNetv2 is trained through an end-to-end supervised learning method, so that the CNN model learns how to recognize surface defects of the CSB only through the RGB image information. Experimental results show that the proposed SurfNetv2 outperforms five state-of-the-art methods and achieves a high recognition accuracy of 99.90% and 99.75% in our private CSB dataset and the public Northeastern University (NEU) dataset, respectively. Moreover, the proposed SurfNetv2 model achieves a real-time computing speed of about 199.38 fps when processing images with a resolution of 128 × 128 pixels. Therefore, the proposed CNN model has great potential for real-time automatic surface defect recognition applications.

show abstract

“…Various artificial intelligence techniques have been implemented in the material science field to carry out different types of analysis or predictions about the properties and behavior of industrial components and materials [ 20 , 22 ]: prediction of elastic properties of metals [ 10 , 61 ], prediction of metallic components behavior [ 19 , 62 , 63 , 64 ], optimization of alloy composition [ 25 , 65 ], or early prediction of the degradation of metallic materials [ 53 , 66 ]. As already stated, artificial intelligence and neural networks can be applied to almost all science fields [ 57 , 67 ].…”

Section: Introductionmentioning

confidence: 99%

Prediction of Mechanical Properties by Artificial Neural Networks to Characterize the Plastic Behavior of Aluminum Alloys

2020

View full text Add to dashboard Cite

In metal forming, the plastic behavior of metallic alloys is directly related to their formability, and it has been traditionally characterized by simplified models of the flow curves, especially in the analysis by finite element simulation and analytical methods. Tools based on artificial neural networks have shown high potential for predicting the behavior and properties of industrial components. Aluminum alloys are among the most broadly used materials in challenging industries such as aerospace, automotive, or food packaging. In this study, a computer-aided tool is developed to predict two of the most useful mechanical properties of metallic materials to characterize the plastic behavior, yield strength and ultimate tensile strength. These prognostics are based on the alloy chemical composition, tempers, and Brinell hardness. In this study, a material database is employed to train an artificial neural network that is able to make predictions with a confidence greater than 95%. It is also shown that this methodology achieves a performance similar to that of empirical equations developed expressly for a specific material, but it provides greater generality since it can approximate the properties of any aluminum alloy. The methodology is based on the usage of artificial neural networks supported by a big data collection about the properties of thousands of commercial materials. Thus, the input data go above 2000 entries. When the relevant information has been collected and organized, an artificial neural network is defined, and after the training, the artificial intelligence is able to make predictions about the material properties with an average confidence greater than 95%.

show abstract

Research on Recognition Technology of Aluminum Profile Surface Defects Based on Deep Learning

Cited by 8 publications

References 21 publications

Machine Vision and Deep Learning Based Rubber Gasket Defect Detection

Machine Vision and Deep Learning Based Rubber Gasket Defect Detection

SurfNetv2: An Improved Real-Time SurfNet and Its Applications to Defect Recognition of Calcium Silicate Boards

Prediction of Mechanical Properties by Artificial Neural Networks to Characterize the Plastic Behavior of Aluminum Alloys

Contact Info

Product

Resources

About