scholarly journals Impact of Image Compression on the Performance of Steel Surface Defect Classification with a CNN

2021 ◽  
Vol 10 (4) ◽  
pp. 73
Author(s):  
Tajeddine Benbarrad ◽  
Lamiae Eloutouate ◽  
Mounir Arioua ◽  
Fatiha Elouaai ◽  
My Driss Laanaoui
Keyword(s):  
Machine Vision ◽  
Image Compression ◽  
High Speed ◽  
Steel Surface ◽  
Surface Defect ◽  
Surface Defects ◽  
High Accuracy ◽  
Classification Performance ◽  
Inspection Time ◽  
The Impact

Machine vision is increasingly replacing manual steel surface inspection. The automatic inspection of steel surface defects makes it possible to ensure the quality of products in the steel industry with high accuracy. However, the optimization of inspection time presents a great challenge for the integration of machine vision in high-speed production lines. In this context, compressing the collected images before transmission is essential to save bandwidth and energy, and improve the latency of vision applications. The aim of this paper was to study the impact of quality degradation resulting from image compression on the classification performance of steel surface defects with a CNN. Image compression was applied to the Northeastern University (NEU) surface-defect database with various compression ratios. Three different models were trained and tested with these images to classify surface defects using three different approaches. The obtained results showed that trained and tested models on the same compression qualities maintained approximately the same classification performance for all used compression grades. In addition, the findings clearly indicated that the classification efficiency was affected when the training and test datasets were compressed using different parameters. This impact was more obvious when there was a large difference between these compression parameters, and for models that achieved very high accuracy. Finally, it was found that compression-based data augmentation significantly increased the classification precision to perfect scores (98–100%), and thus improved the generalization of models when tested on different compression qualities. The importance of this work lies in exploiting the obtained results to successfully integrate image compression into machine vision systems, and as appropriately as possible.

scholarly journals Steel Surface Defect Classification Using Deep Residual Neural Network

Metals ◽  
2020 ◽  
Vol 10 (6) ◽  
pp. 846
Author(s):  
Ihor Konovalenko ◽  
Pavlo Maruschak ◽  
Janette Brezinová ◽  
Ján Viňáš ◽  
Jakub Brezina
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
High Speed ◽  
Steel Surface ◽  
Surface Defect ◽  
Surface Defects ◽  
Defect Classification ◽  
Flat Surfaces ◽  
Automated Method ◽  
Speed And Accuracy

An automated method for detecting and classifying three classes of surface defects in rolled metal has been developed, which allows for conducting defectoscopy with specified parameters of efficiency and speed. The possibility of using the residual neural networks for classifying defects has been investigated. The classifier based on the ResNet50 neural network is accepted as a basis. The model allows classifying images of flat surfaces with damage of three classes with the general accuracy of 96.91% based on the test data. The use of ResNet50 is shown to provide excellent recognition, high speed, and accuracy, which makes it an effective tool for detecting defects on metal surfaces.

Review of Vision Real-Time Inspection Algorithm for Rolling Steel Surface Defects

2011 ◽  
Vol 308-310 ◽  
pp. 1328-1332 ◽  
Author(s):  
Wu Bin Li ◽  
Chang Hou Lu ◽  
Jian Chuan Zhang
Keyword(s):  
Machine Vision ◽  
Real Time ◽  
Steel Surface ◽  
Surface Defect ◽  
Surface Defects ◽  
Defect Inspection ◽  
Development Trends ◽  
Key Issues ◽  
At Home

Rolling steel surface defect inspection technology based on machine vision is more and more widely used. The latest progress of vision-based real-time inspection algorithm for rolling steel surface defect at home and abroad is introduced, and several key issues are analyzed. Finally, the current domestic research emphases and development trends are proposed.

scholarly journals Strip Steel Surface Defects Recognition Based on SOCP Optimized Multiple Kernel RVM

2018 ◽  
Vol 2018 ◽  
pp. 1-8
Author(s):  
Hou Jingzhong ◽  
Xia Kewen ◽  
Yang Fan ◽  
Zu Baokai
Keyword(s):  
Steel Surface ◽  
Surface Defect ◽  
Surface Defects ◽  
Recognition Rate ◽  
Strip Surface ◽  
Strip Steel ◽  
Practical Applications ◽  
Multiple Kernel ◽  
Low Efficiency ◽  
Defect Recognition

Strip steel surface defect recognition is a pattern recognition problem with wide applications. Previous works on strip surface defect recognition mainly focus on feature selection and dimension reduction. There are also approaches on real-time systems that mainly exploit the autocorrection within some given picture. However, the instances cannot be used in practical applications because of a bad recognition rate and low efficiency. In this paper, we study the intelligent algorithm of strip steel surface defect recognition, where the goal is to improve the accuracy and save running time. This problem is very important in various applications, especially the process testing of steel manufacturing. We propose an approach called the second-order cone programming (SOCP) optimized multiple kernel relevance vector machine (MKRVM), which can recognize strip surface defects much better than other methods. The method includes the model parameter estimation, training, and optimization of the model based on SOCP and the classification test. We compare our approach with existing methods on strip surface defect recognition. The results demonstrate that our proposed approach can improve the recognition accuracy and reduce the time costs of the strip surface defect.

Defect Detecting Technology Based on Machine Vision of Industrial Parts

2014 ◽  
Vol 641-642 ◽  
pp. 1275-1279 ◽  
Author(s):  
Xiao Jun He ◽  
Zhen Di Yi ◽  
Jing Liu ◽  
Yu Zheng Wang
Keyword(s):  
Machine Vision ◽  
Edge Detection ◽  
Surface Defects ◽  
Detection Method ◽  
Median Filter ◽  
High Accuracy ◽  
Testing Time

In order to reach and test the surface defects on industrial parts, based on Machine Vision this paper put forward a defective parts detection method. The method of median filter was adopted to eliminate the noise of image. The Ostu-method was used for the segmenting threshold. Pixel level and level edge detection were used to complete the precise components defects detection. Experiments show that this scheme is feasible, and can achieve high accuracy and shorter testing time.

Research on Surface Defect Detection Technique of Rolling Element Based on Computer Vision

2014 ◽  
Vol 1006-1007 ◽  
pp. 773-778 ◽  
Author(s):  
Chuan Ren ◽  
Xiao Yu Xiu ◽  
Guo Hui Zhou
Keyword(s):  
Computer Vision ◽  
Defect Detection ◽  
High Speed ◽  
Surface Defect ◽  
Surface Defects ◽  
Median Filter ◽  
Image Sensor ◽  
Threshold Method ◽  
Rolling Element ◽  
Surface Defect Detection

This paper proposed a new method of surface defect detection of rolling element based on computer vision, which adopted CCD digital camera as image sensor, and used digital image processing techniques to defect the surface defects of rolling element. The main steps include collect image, use an improved median filter to reduce the noise, increase or decrease the exposure to achieve the image enhancement, create a binary image with threshold method and detect the edge of the image, and use subtraction method for surface defects identification. The experiment indicates that the above methods the advantages of simple, the capability of noise resistance, high speed processing and better real-time.

scholarly journals Steel Surface Defect Classification Based on Small Sample Learning

2021 ◽  
Vol 11 (23) ◽  
pp. 11459
Author(s):  
Shiqing Wu ◽  
Shiyu Zhao ◽  
Qianqian Zhang ◽  
Long Chen ◽  
Chenrui Wu
Keyword(s):  
Feature Extraction ◽  
Steel Surface ◽  
Surface Defect ◽  
Nearest Neighbor ◽  
Surface Defects ◽  
Feature Space ◽  
Small Sample ◽  
Defect Classification ◽  
Sample Number ◽  
Degradation Problem

The classification of steel surface defects plays a very important role in analyzing their causes to improve manufacturing process and eliminate defects. However, defective samples are very scarce in actual production, so using very few samples to construct a good classifier is a challenge to be addressed. If the layer number of the model with proper depth is increased, the model accuracy will decrease (not caused by overfit), and the training error as well as the test error will be very high. This is called the degradation problem. In this paper, we propose to use feature extraction + feature transformation + nearest neighbors to classify steel surface defects. In order to solve the degradation problem caused by network deepening, the three feature extraction networks of Residual Net, Mobile Net and Dense Net are designed and analyzed. Experiment results show that in the case of a small sample number, Dense block can better solve the degradation problem caused by network deepening than Residual block. Moreover, if Dense Net is used as the feature extraction network, and the nearest neighbor classification algorithm based on Euclidean metric is used in the new feature space, the defect classification accuracy can reach 92.33% when only five labeled images of each category are used as the training set. This paper is of some guiding significance for surface defect classification when the sample number is small.

Research on On-Line Surface Defect Detection for Steel Strip Based on Sparse Coding

2012 ◽  
Vol 548 ◽  
pp. 749-752 ◽  
Author(s):  
Zhao Liu ◽  
Jia Hu ◽  
Li Hu ◽  
Xiao Long Zhang ◽  
Jian Yi Kong
Keyword(s):  
Machine Vision ◽  
Sparse Coding ◽  
Visual Information ◽  
Surface Defect ◽  
Surface Defects ◽  
Steel Strip ◽  
Feature Representation ◽  
Surface Defect Detection ◽  
On Line ◽  
Object Feature

In the field of metallurgy, surface defects detection for steel plate based on machine vision is a new key technology. In order to improve the accuracy and speed of machine vision in real-time surface defects detection, taking into account the neurons selectivity and sparseness to visual information, we present a flexible data selection mechanism in the layer of photoreceptors and a new sparse coding model for object feature representation and object recognition. Experiments show that the new method is more effective and more effective in the process of training and classification. The key finding of this study is that, the effective sparse coding mechanism not only could have occurred in the data input stage, but also could be in a new way.

A Real-Time Steel Surface Defect Detection Approach with High Accuracy

2021 ◽  
pp. 1-1
Author(s):  
Wenyan Wang ◽  
Chunfeng Mi ◽  
Ziheng Wu ◽  
Kun Lu ◽  
Hongming Long ◽  
...  
Keyword(s):  
Real Time ◽  
Defect Detection ◽  
Steel Surface ◽  
Surface Defect ◽  
High Accuracy ◽  
Detection Approach ◽  
Surface Defect Detection

Recent Development of Equal Channel Angular Rolling (ECAR)

2010 ◽  
Vol 638-642 ◽  
pp. 1917-1922
Author(s):  
Young Hoon Chung
Keyword(s):  
Processing Speed ◽  
High Speed ◽  
Surface Defect ◽  
Surface Defects ◽  
Thin Sheet ◽  
Grain Structure ◽  
Shear Texture ◽  
Fine Grain ◽  
Plastic Deformation Process ◽  
Technical Control

Equal Channel Angular Rolling (ECAR), the severe plastic deformation process, is suitable for shear deforming long and thin sheet continuously. An interesting issue is that thickness of a sheet is not reduced during ECAR. Although shear texture and fine grain structure in Al alloys are easily obtained by ECAR, yet the ECAR process’s difficulties in terms of technical control still remain, such as surface defect, low ductility and low processing speed. The surface defects and processing speed are partially improved by applying a series deformation of rolling and ECAR. A high-speed solution heat-treatment is developed for restoring the ductility of Al 6061 alloy.

Research on Rail Surface Defect Inspection System

2012 ◽  
Vol 229-231 ◽  
pp. 1389-1393
Author(s):  
Yu Hu ◽  
Jian Xu Mao ◽  
Jian Pin Mao
Keyword(s):  
High Speed ◽  
Surface Defect ◽  
Surface Defects ◽  
Imaging System ◽  
Detection System ◽  
Inspection System ◽  
System Control ◽  
Visual Imaging ◽  
Automatic Detection System ◽  
Rail Surface Defect

In order to realize the inspection of rail surface defects with high speed and high precision, an automatic detection system based on machine vision is presented. The hardware structure of the system consists of the mechanical system, control system and visual imaging system. The software structure using histogram threshold segmentation, multi-structural morphological edge detection and other image processing methods to detect and identify defects automatically, and also built the simulation rail detection platform. The experimental results show that the cracks, scars and other detects can be accurately detected and extracted in real time, and meet the requirement of the rail surface inspection.

Sign in / Sign up

Export Citation Format

Share Document