Model based visual inspection of pharmaceutical tablets with photometric stereo

Podrekar, Gregor; Tomaževič, Dejan; Likar, Boštjan; Usenik, Peter

doi:10.23919/mva.2017.7986819

Cited by 7 publications

(3 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition to the orientation of the surface normals, height information of the component surface 3D depth map can be used to detect defects. Podrekar [Pod+17] et al use a 3D depth map for quality inspection on tablets, which is calculated from the surface normal using path integration procedure. Defects can be detected by comparing the maps with a tablet model.…”

Section: Introductionmentioning

confidence: 99%

Detection of Printed Circuit Board Defects with Photometric Stereo and Convolutional Neural Networks

Hable¹,

E.²,

Scherr³

et al. 2023

Computer Science Research Notes

View full text Add to dashboard Cite

The quality inspection of printed circuit boards (PCBs) is no longer feasible by human inspectors due to accuracy requirements and the processing volume. Automated optical inspection systems must be specifically designed to meet the various inspection requirements. A photometric stereo setup is suitable for the inspection of highly reflective gold areas on PCBs. In this process, several image acquisitions are performed under different illumination directions. This can reveal defects that are not visible under other illumination systems. In this paper, we use a segmented ring light so that image acquisition is possible under four different illumination directions. Using these images, a normal map and a mean image are calculated. The defects on the gold areas of PCBs are detectable in either the normal map, the mean image, or both images. A CNN for classification detects the defects. The input is a 6-dimensional image from normal map and mean image. An accuracy up to 95% can be achieved with the available dataset.

show abstract

Section: Introductionmentioning

confidence: 99%

Detection of Printed Circuit Board Defects with Photometric Stereo and Convolutional Neural Networks

Hable¹,

E.²,

Scherr³

et al. 2023

Computer Science Research Notes

View full text Add to dashboard Cite

show abstract

“…That is, the shadow feature changes with different light angles. Further, the characteristics of surface gradient and height can be obtained by using shadow analysis in [28], [29] or photometric stereo in [30]- [32]. The photometric stereo is based on the ideal theoretical assumptions [33], [34]: (1) the light sources are at infinity; (2) the surface is a Lambert reflector; (3) the camera imaging model is orthogonal model.…”

Section: Introductionmentioning

confidence: 99%

Research on Surface Defect Detection Technology Based on the Zonal and Time-Sharing Computational Imaging

Yin

Yuan

et al. 2022

IEEE Access

View full text Add to dashboard Cite

In order to solve the misjudgment problem of different information in common gray images due to the very similar features such as morphology and gray value, and to avoid the high cost of using the 3-D sensor in defect detection, a zonal and time-sharing computational imaging (ZTSCI) is proposed in this paper. Firstly, a four-zone and time-sharing exposure visual imaging system is constructed by monocular telecentric imaging and lights in different directions. Then, according to the gray images under four groups of illumination conditions, and the direction vectors of incident lights, five feature images are calculated, such as fused images of maximum and average, enhanced images of normal vector, gradient and relative height. On the one hand, the image background fluctuation caused by the calculation error introduced by the ideal hypothesis is weakened, and the interference of surface texture to defect detection is reduced. On the other hand, the difference of easily-confused information can be increased in different feature spaces. In this paper, the ZTSCI is further applied to defect detection on the surface of new energy batteries. It solves the problem that defects and interference information are difficult to distinguish in gray images captured with common machine vision methods, reduces the difficulty of subsequent defect recognition, and lays a foundation for improving the accuracy of recognition.

show abstract

“…This method is capable of deriving the normal vectors of a surface from multiple images of that surface under dierent lighting conditions. Currently, photometric stereo is used to detect defects in many dierent products, including steel strips [13], and pharmaceutical tablets [14]. Deep learning was introduced into defect detection with photometric stereo to increase performance by improving the accuracy of the model and by decreasing the run time of the photometric stereo algorithm.…”

Section: Introductionmentioning

confidence: 99%

Online Metallic Surface Defect Detection Using Deep LearningOnline Metallic Surface Defect Detection Using Deep Learning

Çerezci¹,

Kazan²,

Oz³

et al. 2020

Preprint

View full text Add to dashboard Cite

Across a range of manufacturing contexts, automated quality control has been gaining significantattention because it offers competitive advantages such as cost reduction, high accuracy in defect detection, and system stability over time. Although computer vision has historically been the most commonly applied method in this context, novel approaches such as deep learning have recently become more frequent and are used in cases where traditional methods cannot be applied. Because of the surface texture and curvature of many metallic parts, detection of defects such as scratches, cracks, and dents can be challenging for traditional computer vision methods. In this study, an image acquisition system supported by a special lighting device that provides processable images from an extremely reflective cylindrical metallic surface has been developed. Multiple images obtained from a single lateral line of the surface, which is rotated at a specified speed, are combined using photometric stereo and given as input to a convolutional neural network that is employed to classify defective and non-defective samples. The results obtained from this method are close to 98.5% accurate.

show abstract

Model based visual inspection of pharmaceutical tablets with photometric stereo

Cited by 7 publications

References 8 publications

Detection of Printed Circuit Board Defects with Photometric Stereo and Convolutional Neural Networks

Detection of Printed Circuit Board Defects with Photometric Stereo and Convolutional Neural Networks

Research on Surface Defect Detection Technology Based on the Zonal and Time-Sharing Computational Imaging

Online Metallic Surface Defect Detection Using Deep LearningOnline Metallic Surface Defect Detection Using Deep Learning

Contact Info

Product

Resources

About