scholarly journals A Lightweight Deep Network for Defect Detection of Insert Molding Based on X-ray Imaging

Sensors ◽  
2021 ◽  
Vol 21 (16) ◽  
pp. 5612
Author(s):  
Benwu Wang ◽  
Feng Huang
Keyword(s):  
Defect Detection ◽  
Molding Process ◽  
X Ray ◽  
Deep Network ◽  
X Ray Imaging ◽  
Common Causes ◽  
Model Size ◽  
The One ◽  
Image Defect ◽  
Insert Molding

Aiming at the abnormality detection of industrial insert molding processes, a lightweight but effective deep network is developed based on X-ray images in this study. The captured digital radiography (DR) images are firstly fast guide filtered, and then a multi-task detection dataset is constructed using an overlap slice in order to improve the detection of tiny targets. The proposed network is extended from the one-stage target detection method of yolov5 to be applicable to DR defect detection. We adopt the embedded Ghost module to replace the standard convolution to further lighten the model for industrial implementation, and use the transformer module for spatial multi-headed attentional feature extraction to perform improvement on the network for the DR image defect detection. The performance of the proposed method is evaluated by consistent experiments with peer networks, including the classical two-stage method and the newest yolo series. Our method achieves a mAP of 93.6%, which exceeds the second best by 3%, with robustness sufficient to cope with luminance variations and blurred noise, and is more lightweight. We further conducted ablation experiments based on the proposed method to validate the 32% model size reduction owing to the Ghost module and the detection performance enhancing effect of other key modules. Finally, the usability of the proposed method is discussed, including an analysis of the common causes of the missed shots and suggestions for modification. Our proposed method contributes a good reference solution for the inspection of the insert molding process.

Defect detection in apples by means of x-ray imaging

1996 ◽  
Author(s):  
Thomas F. Schatzki ◽  
Ron P. Haff ◽  
Richard Young ◽  
Ilkay Can ◽  
Lan Chau Le ◽  
...  
Keyword(s):  
Defect Detection ◽  
X Ray ◽  
X Ray Imaging

DEFECT DETECTION IN APPLES BY MEANS OF X-RAY IMAGING

1997 ◽  
Vol 40 (5) ◽  
pp. 1407-1415 ◽  
Author(s):  
T. F. Schatzki ◽  
R. P. Haff ◽  
R. Young ◽  
I. Can ◽  
L-C. Le ◽  
...  
Keyword(s):  
Defect Detection ◽  
X Ray ◽  
X Ray Imaging

Analysis of the Mechanism for the Appearance of Bubbles in Rubber during Vulcanization

2018 ◽  
Vol 45 (4) ◽  
pp. 175-182
Author(s):  
Yuji Sato
Keyword(s):  
Moisture Content ◽  
Crosslink Density ◽  
Imaging Method ◽  
Pixel Size ◽  
X Ray ◽  
X Ray Imaging ◽  
Cure Time ◽  
The One ◽  
Curing Pressure ◽  
Radiation Research

Appearance of bubbles in the rubber has been observed in real time by using X-ray imaging method in SPring-8. The behavior of diameter and quantity of bubbles has been measured for the samples which have the various contents of cure agent and moisture. The babbles appear after 30 to 50 sec from releasing curing pressure at the place where there are no any feature in the X-ray images pixel size is 0.5 μm. The diameter of bubbles increases linearly with time at first, and then it grows up to be closer to the limit diameter finally. This final diameter changes with cure time and amount of cure agent. And the quantity of bubbles changes with amount of moisture. The result shows the one of origin of bubbles is moisture, and even if the moisture content changes, the size of moisture particles does not change, the quantity of particles changes. The crosslink density participates in whether this particle changes to a bubble or keeps that state. That also participates in the final diameter of bubble. X-Ray experiments were performed at BL19B2, BL46XU in the SPring-8 with approval of the Japan Radiation Research Institute (JASRI) Proposal No. 2014A1571,2014A1572).

scholarly journals Dose optimization approach to fast X-ray microtomography of the lung alveoli

2013 ◽  
Vol 46 (4) ◽  
pp. 856-860 ◽  
Author(s):  
Goran Lovric ◽  
Sébastien F. Barré ◽  
Johannes C. Schittny ◽  
Matthias Roth-Kleiner ◽  
Marco Stampanoni ◽  
...  
Keyword(s):  
Radiation Dose ◽  
Radiation Risk ◽  
Experiment Planning ◽  
Optimization Approach ◽  
Worst Case ◽  
X Ray ◽  
X Ray Imaging ◽  
The One

A basic prerequisite for in vivo X-ray imaging of the lung is the exact determination of radiation dose. Achieving resolutions of the order of micrometres may become particularly challenging owing to increased dose, which in the worst case can be lethal for the imaged animal model. A framework for linking image quality to radiation dose in order to optimize experimental parameters with respect to dose reduction is presented. The approach may find application for current and future in vivo studies to facilitate proper experiment planning and radiation risk assessment on the one hand and exploit imaging capabilities on the other.

scholarly journals X-ray Imaging Investigation on the Gilding Technique of an Ancient Egyptian Taweret Wooden Statuette

2021 ◽  
Vol 7 (11) ◽  
pp. 229
Author(s):  
Luisa Vigorelli ◽  
Alessandro Re ◽  
Laura Guidorzi ◽  
Tiziana Cavaleri ◽  
Paola Buscaglia ◽  
...  
Keyword(s):  
Imaging Techniques ◽  
Imaging Study ◽  
Ancient Egypt ◽  
X Ray ◽  
Ancient Egyptian ◽  
X Ray Imaging ◽  
Investigation Methods ◽  
The One ◽  
Technical Features ◽  
Tomographic Analysis

Diagnostic physical methods are increasingly applied to Cultural Heritage both for scientific investigations and conservation purposes. In particular, the X-ray imaging techniques of computed tomography (CT) and digital radiography (DR) are non-destructive investigation methods to study an object, being able to give information on its inner structure. In this paper, we present the results of the X-ray imaging study on an ancient Egyptian statuette (Late Period 722–30 BCE) belonging to the collection of Museo Egizio in Torino and representing an Egyptian goddess called Taweret, carved on wood and gilded with some colored details. Since few specific studies have been focused on materials and techniques used in Ancient Egypt for gilding, a detailed investigation was started in order to verify the technical features of the decoration in this sculpture. Specifically, DR and CT analyses have been performed at the Centro Conservazione e Restauro “La Venaria Reale” (CCR), with a new high resolution flat-panel detector, that allowed us to perform tomographic analysis reaching a final resolution better than the one achievable with the previous apparatus operating in the CCR.

scholarly journals Effectiveness of Ultrasound in Verification of the Mucus Plugs and Sialoliths of the Wharton`s Duct

2019 ◽  
pp. 144-153
Keyword(s):  
Salivary Gland ◽  
Color Doppler ◽  
X Ray ◽  
Clinical Appearance ◽  
Endoscopic View ◽  
Salivary Calculi ◽  
Common Causes ◽  
Obstructive Sialadenitis ◽  
Salivary Gland Disease ◽  
The One

The pathological changes in 467 submandibular glands were identified both endoscopically and radiographically, and endoscopic findings showed three types: calculus (91 percent), mucus plug (3 percent), and stenosis (6 percent). —Yu Chuangqi et al, 2013 China Mucus plugs (synonyms: mucous plugs, mucin plugs, fibromucinous plugs and mucosal plugs) and sialoliths (synonyms: salivary stones, salivary calculi, and concrements) belong to the one of the common causes of the obstructive salivary gland disease (synonyms: obstructive sialadenitis and obstructive sialadenopathy). Among other etiologies of obstructive sialadenitis are: foreign bodies, inflammation, kinks, strictures, anatomic malformations, polyps or even tumors. Those causes are found in different percentages. The radiographic investigation e.g. X-ray and computed tomography (CT) are very useful in detection of the salivary stones. Nevertheless, as approximately 80-90 percent of the sialoliths are opaque on a standard review X-ray and CT, and in 10-20% radiolucent. But these methods are not useful in the detection of mucus plugs due to the non-contrast features of the last. There are a lot of studies which described ultrasound features of the sialoliths. Also, there are some studies that demonstrate endoscopic view of the mucosal plugs in a ductal system and in some cases the authors during sialendoscopy noted the floating mucous plugs. But we cannot find articles in PubMed which demonstrate ultrasound and clinical appearance of the obstructive salivary gland disease caused by sialoliths with mucus plugs simultaneously. The purpose of our article is to describe a first and precise description of ultrasound pattern of the mucus plugs comparing with sialolith and their clinical presentation after removal. We report the consecutive gray scale and color Doppler sonograms with a supplemental video.

scholarly journals Deep learning based solder joint defect detection on industrial printed circuit board X-ray images

2022 ◽  
Author(s):  
Qianru Zhang ◽  
Meng Zhang ◽  
Chinthaka Gamanayake ◽  
Chau Yuen ◽  
Zehao Geng ◽  
...  
Keyword(s):  
Quality Control ◽  
Deep Learning ◽  
Defect Detection ◽  
Printed Circuit Board ◽  
Circuit Board ◽  
Image Size ◽  
X Ray ◽  
Component Size ◽  
Image Dimension ◽  
X Ray Imaging

AbstractWith the improvement of electronic circuit production methods, such as reduction of component size and the increase of component density, the risk of defects is increasing in the production line. Many techniques have been incorporated to check for failed solder joints, such as X-ray imaging, optical imaging and thermal imaging, among which X-ray imaging can inspect external and internal defects. However, some advanced algorithms are not accurate enough to meet the requirements of quality control. A lot of manual inspection is required that increases the specialist workload. In addition, automatic X-ray inspection could produce incorrect region of interests that deteriorates the defect detection. The high-dimensionality of X-ray images and changes in image size also pose challenges to detection algorithms. Recently, the latest advances in deep learning provide inspiration for image-based tasks and are competitive with human level. In this work, deep learning is introduced in the inspection for quality control. Four joint defect detection models based on artificial intelligence are proposed and compared. The noisy ROI and the change of image dimension problems are addressed. The effectiveness of the proposed models is verified by experiments on real-world 3D X-ray dataset, which saves the specialist inspection workload greatly.

Sign in / Sign up

Export Citation Format

Share Document