Cylinder Liner Defect Detection and Classification based on Deep Learning

Abstract

The machine vision-based defect detection for cylinder liner is a challenging task due to irregular shape, various and small defects on the cylinder liner surface. To improve the accuracy of defect detection by machine vision a deep learning-based defect detection method for cylinder liner was explored in this paper. First, a machine vision system was designed based on the analysis of the causes and types of defects to obtain the field images for establishing an original dataset. Then the dataset was augmented by a modified augmentation method which combines the region of interest automatic extraction method with the traditional augmentation methods. Except for introduction of the anchor configuration optimization method, an XML file-based method of highlighting defect area was proposed to address the problem of tiny defect detection. The optimal model was experimentally determined by considering the network model, the training strategy and the sample size. Finally, the detection system was developed and the network model was deployed. Experiments are carried out and the results of the proposed method compared with those of the traditional methods. The results show that the detection accuracies of sand, scratch and wear defects are 77.5%, 70% and 66.3% which are improved by at least 26.3% compared with the traditional methods. The proposal can be used for field defect detection of cylinder liner.

Efficient packaging defect detection: leveraging pre-trained vision models through transfer learning

“…The dataset was then augmented using a modified augmentation technique incorporating the region of interest's automatic extraction technique with conventional augmentation techniques. The results indicate that the detection accuracies were superior to those of conventional methods [4], [11]. In addition, a Wavelet Transform Multiscale Filtering algorithm has been offered for the experiment of defect detection on glass bottles to reduce the influence of texture and increase the robustness to localization error [13].…”

“…The subfield of machine learning, known as "deep learning", focuses on creating and applying methods that let neural networks learn from and make predictions about big, complex datasets. Multiple layers of nodes, also called neurons, are interconnected in deep learning [4]. In computer vision, deep learning has shown exceptional effectiveness in recent years.…”

“…In addition, the parameters of the network that are learned while working with one network can be applied to other networks that are comparable in order to produce high success rates for surface defect detection. Consequently, a number of studies have utilized a computer vision strategy to support in classification, defect detection, and quality inspection for fabric defect [3], [10], cylinder and glass bottle [4], [11], [12] and box packaging [1], [8]. For instance, Jeyaraj and Nadar [3] had proposed developing a quick and efficient classification system for fabric defects.…”

“…Due to the irregular shape, variety, and small size of the surface defects, cylinder liner defect detection based on machine vision is a challenging task. Gao et al [4] used experiments to determine which model was the most effective. The dataset was then augmented using a modified augmentation technique incorporating the region of interest's automatic extraction technique with conventional augmentation techniques.…”

Towards High Quality PCB Defect Detection Leveraging State-of-the-Art Hybrid Models

“…In recent years, deep learning has revolutionized the field of artificial intelligence, leading to significant advancements in various domains such as computer vision, natural language processing, autonomous vehicles, and medical diagnostics [1,2]. At its core, deep learning utilizes neural networks with multiple layers to learn representations of data with multiple levels of abstraction, enabling the discovery of intricate structures in large datasets.…”