A pre‐processing technique to decrease inspection time in glass tube production lines
Abstract In case of glass tube for pharmaceutical applications, high‐quality defect detection is made via inspection systems based on computer vision. The processing must guarantee real‐time inspection and meet increasing rate and quality requirements. Defect detection in glass tubes is complicated...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2021-08-01
|
| Series: | IET Image Processing |
| Subjects: | |
| Online Access: | https://doi.org/10.1049/ipr2.12186 |
| _version_ | 1798035528133640192 |
|---|---|
| author | Gabriele Antonio De Vitis Pierfrancesco Foglia Cosimo Antonio Prete |
| author_facet | Gabriele Antonio De Vitis Pierfrancesco Foglia Cosimo Antonio Prete |
| author_sort | Gabriele Antonio De Vitis |
| collection | DOAJ |
| description | Abstract In case of glass tube for pharmaceutical applications, high‐quality defect detection is made via inspection systems based on computer vision. The processing must guarantee real‐time inspection and meet increasing rate and quality requirements. Defect detection in glass tubes is complicated by aspects that hamper the efficiency of state‐of‐the‐art techniques. This paper presents a pre‐processing algorithm which excludes portions of the image where defects are surely absent. The approach decreases the time for defect detection and classification phases (any detection algorithm can be applied), as they are applied only in high‐probability candidate sub‐image. We derive a methodology to get robust values of algorithm's parameters during production. The algorithm relies on detrended standard deviation and double threshold hysteresis, which solve issues related to the misalignment between illuminator and acquisition camera, and enable a robust detection despite rotation, vibration, and irregularities of tubes. We consider Canny, MAGDDA, and Niblack algorithms. The solution keeps the detection quality of such algorithms and reaches a 4.69× throughput gain. It represents a methodology to obtain defect detection in time‐constrained environments through a software‐only approach, and can be exploited in parallel/accelerated solutions and in contexts where a linear camera is utilized on both flat and uneven surfaces. |
| first_indexed | 2024-04-11T20:59:26Z |
| format | Article |
| id | doaj.art-b7068861a68b4a78b98eca37658d9b21 |
| institution | Directory Open Access Journal |
| issn | 1751-9659 1751-9667 |
| language | English |
| last_indexed | 2024-04-11T20:59:26Z |
| publishDate | 2021-08-01 |
| publisher | Wiley |
| record_format | Article |
| series | IET Image Processing |
| spelling | doaj.art-b7068861a68b4a78b98eca37658d9b212022-12-22T04:03:34ZengWileyIET Image Processing1751-96591751-96672021-08-0115102179219110.1049/ipr2.12186A pre‐processing technique to decrease inspection time in glass tube production linesGabriele Antonio De Vitis0Pierfrancesco Foglia1Cosimo Antonio Prete2Dipartimento di Ingegneria dell'Informazione Università di Pisa Via Diotisalvi 2 Pisa ItalyDipartimento di Ingegneria dell'Informazione Università di Pisa Via Diotisalvi 2 Pisa ItalyDipartimento di Ingegneria dell'Informazione Università di Pisa Via Diotisalvi 2 Pisa ItalyAbstract In case of glass tube for pharmaceutical applications, high‐quality defect detection is made via inspection systems based on computer vision. The processing must guarantee real‐time inspection and meet increasing rate and quality requirements. Defect detection in glass tubes is complicated by aspects that hamper the efficiency of state‐of‐the‐art techniques. This paper presents a pre‐processing algorithm which excludes portions of the image where defects are surely absent. The approach decreases the time for defect detection and classification phases (any detection algorithm can be applied), as they are applied only in high‐probability candidate sub‐image. We derive a methodology to get robust values of algorithm's parameters during production. The algorithm relies on detrended standard deviation and double threshold hysteresis, which solve issues related to the misalignment between illuminator and acquisition camera, and enable a robust detection despite rotation, vibration, and irregularities of tubes. We consider Canny, MAGDDA, and Niblack algorithms. The solution keeps the detection quality of such algorithms and reaches a 4.69× throughput gain. It represents a methodology to obtain defect detection in time‐constrained environments through a software‐only approach, and can be exploited in parallel/accelerated solutions and in contexts where a linear camera is utilized on both flat and uneven surfaces.https://doi.org/10.1049/ipr2.12186Inspection and quality controlImage and video codingImage sensorsPower applications in glass, ceramic, brick and cement industriesComputer vision and image processing techniquesProduction engineering computing |
| spellingShingle | Gabriele Antonio De Vitis Pierfrancesco Foglia Cosimo Antonio Prete A pre‐processing technique to decrease inspection time in glass tube production lines IET Image Processing Inspection and quality control Image and video coding Image sensors Power applications in glass, ceramic, brick and cement industries Computer vision and image processing techniques Production engineering computing |
| title | A pre‐processing technique to decrease inspection time in glass tube production lines |
| title_full | A pre‐processing technique to decrease inspection time in glass tube production lines |
| title_fullStr | A pre‐processing technique to decrease inspection time in glass tube production lines |
| title_full_unstemmed | A pre‐processing technique to decrease inspection time in glass tube production lines |
| title_short | A pre‐processing technique to decrease inspection time in glass tube production lines |
| title_sort | pre processing technique to decrease inspection time in glass tube production lines |
| topic | Inspection and quality control Image and video coding Image sensors Power applications in glass, ceramic, brick and cement industries Computer vision and image processing techniques Production engineering computing |
| url | https://doi.org/10.1049/ipr2.12186 |
| work_keys_str_mv | AT gabrieleantoniodevitis apreprocessingtechniquetodecreaseinspectiontimeinglasstubeproductionlines AT pierfrancescofoglia apreprocessingtechniquetodecreaseinspectiontimeinglasstubeproductionlines AT cosimoantonioprete apreprocessingtechniquetodecreaseinspectiontimeinglasstubeproductionlines AT gabrieleantoniodevitis preprocessingtechniquetodecreaseinspectiontimeinglasstubeproductionlines AT pierfrancescofoglia preprocessingtechniquetodecreaseinspectiontimeinglasstubeproductionlines AT cosimoantonioprete preprocessingtechniquetodecreaseinspectiontimeinglasstubeproductionlines |