Fabric image inspection using deep learning approach

Sizyakin, Roman; Voronin, Viacheslav; Gapon, Nikolay; Semenishchev, Evgeny A.; Zelensky, Alehander; Ilyukhin, Yu. V.

doi:10.1117/12.2646485

Cited by 3 publications

(7 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The general scheme of the proposed method is shown in Figure 4. For image enhancement we combined local and global transform based on multi-scale block-rooting processing [1,2].…”

Section: The Defect Detection Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Defect detection and removal for depth map quality enhancement in manufacturing with deep learning

Gapon

Voronin

Sizyakin

et al. 2023

Dimensional Optical Metrology and Inspection for Practical Applications XII

View full text Add to dashboard Cite

A system for determination the distance from the robot to the scene is useful for object tracking, and 3-D reconstruction may be desired for many manufacturing and robotic tasks. While the robot is processing materials, such as welding parts, milling, drilling, fragments of materials fall on the camera installed on the robot, introducing unnecessary information when building a depth map, as well as the emergence of new lost areas, which leads to incorrect determination of the size of objects. There is a problem comprising a decrease in the accuracy of planning the movement trajectory caused by incorrect sections on the depth map because of incorrect distance determination of objects. We present a two-stage approach combining defect detection and depth reconstruction algorithms. The first step is image defects detection based on convolutional auto-encoder (U-Net) and deep feature fusion network (DFFN-Net). The second step is a depth map reconstruction with the exemplar-based and the anisotropic gradient concepts. The proposed modified block fusion algorithm uses a local image descriptor obtained by an automatic encoder for image reconstruction, which extracts image features and depth maps using a decoding network. Our technique outperforms the state-of-the-art methods quantitatively in reconstruction accuracy on RGB-D benchmark for evaluating manufacturing vision systems.

show abstract

“…The general scheme of the proposed method is shown in Figure 4. For image enhancement we combined local and global transform based on multi-scale block-rooting processing [1,2].…”

Section: The Defect Detection Methodsmentioning

confidence: 99%

“…In hidden layers, an exponential rectified linear unit (ELU) is used as the activation function [1]. For loss estimation, we use Sorensen-Dice coefficient [1,2] showing the measure of the area of correctly marked segments [2]. The architecture of the proposed neural network includes fourteen hidden layers.…”

Section: U-net-based Defects Detection Modelmentioning

confidence: 99%

Defect detection and removal for depth map quality enhancement in manufacturing with deep learning

Gapon

Voronin

Sizyakin

et al. 2023

Dimensional Optical Metrology and Inspection for Practical Applications XII

View full text Add to dashboard Cite

show abstract

“…The general scheme of the proposed method is shown in Figure 2. To improve the image, we combined the local and global transformation based on multiscale processing of block rooting [27,28].…”

Section: General Scheme Of the Proposed Methodsmentioning

confidence: 99%

“…Using ELU allows you to achieve convergence of the neural network faster and with greater accuracy, as well as avoid the process of batch normalization. To estimate losses, we use the Sorensen-Dice coefficient [28], which shows the measure of the area of correctly marked segments:…”

Section: U-net-based Defects Detectionmentioning

confidence: 99%

See 1 more Smart Citation

Deep learning-based image defect detection and removal in manufacturing

Gapon¹,

Voronin²,

Zhdanova³

et al. 2023

Digital Optical Technologies 2023

View full text Add to dashboard Cite

Image inpaintnig in textile manufacturing is a new emerging research topic in preprocessing for jacquard CAD systems. One of the most important aspects of a jacquard CAD system is the simulation of the appearance of a jacquard texture during inference. Jacquard image inpainting has become an indispensable process for the Jacquard CAD system. Jacquard image reconstruction is designed to restore a damaged image with missing information, so it is necessary to determine which parts of the image need to be repaired. Thus, this task includes two processing stages: the detection of defects and their recovery. This article presents a two-stage approach that combines new and traditional algorithms for detecting defects and repairing damaged areas. The first stage is a defect detection method based on a convolutional autoencoder (U-Net). The second stage is image inpainting based on exemplar-based concepts and the anisotropic gradient. Our system quantitatively outperforms state-of-the-art methods regarding reconstruction accuracy in the benchmark.

show abstract

Deep learning-based depth map defect removal for industrial applications

Voronin,

Gapon,

Zhdanova

et al. 2023

Optical Metrology and Inspection for Industrial Applications X

View full text Add to dashboard Cite

A system for determining the distance from the robot to the scene is useful for object tracking, and 3-D reconstructions may be desired for many manufacturing and robotic tasks. While the robot is processing materials, such as welding parts, milling, drilling, etc., fragments of materials fall on the camera installed on the robot, introducing unnecessary information when building a depth map, as well as the emergence of new lost areas, which leads to incorrect determination of the size of objects. There is a problem comprising a decrease in the accuracy of planning the movement trajectory caused by wrong sections on the depth map because of erroneous distance determination to objects. We present an approach combining defect detection and depth reconstruction algorithms. The first step for image defect detection is based on a convolutional auto-encoder (U-Net). The second step is a depth map reconstruction using a spatial reconstruction based on a geometric model with contour and texture analysis. We apply contour restoration and texture synthesis for image reconstruction. A method is proposed for restoring the boundaries of objects in an image based on constructing a composite curve by cubic splines. Our technique outperforms the state-of-the-art methods quantitatively in reconstruction accuracy on the RGB-D benchmark for evaluating manufacturing vision systems.

show abstract

Fabric image inspection using deep learning approach

Cited by 3 publications

References 14 publications

Defect detection and removal for depth map quality enhancement in manufacturing with deep learning

Defect detection and removal for depth map quality enhancement in manufacturing with deep learning

Deep learning-based image defect detection and removal in manufacturing

Deep learning-based depth map defect removal for industrial applications

Contact Info

Product

Resources

About