Downloaded 39 times
Uploaded byBill Cardoso
New Algorithms to Improve X-Ray Inspection
The document discusses advancements in X-ray inspection, emphasizing the transition from hardware improvements to algorithm innovations due to increased computing power. It highlights various algorithms for real-time image processing, dual energy X-ray imaging, and the impact of these technologies on counterfeit detection. The overall conclusion is that while the first century of X-ray inspection was hardware-driven, the current and future advancements will be largely influenced by algorithms.
More Related Content
![UiPath Automation Suite Installation (Hands-On) [2/3]](https://cdn.slidesharecdn.com/ss_thumbnails/automationsuitecommunitysession2-251015095633-a6d862f1-thumbnail.jpg?width=600ounds&width=560&fit=bounds)
New Algorithms to Improve X-Ray Inspection
- 1. New Algorithms toImprove X-Ray Inspection
- 2.
- 3.
- 4. This*century,*algorithms*are*driving*x.ray*innovation • Significant*improvement*in*computing* power • Explosive*algorithm*focus*–see*Google • Marginal*improvement*in*hardware 4
- 5. Outline • SMT*Defects*identified*with*x.ray* inspection • WISE •Dual*Energy • ICARUS • Artificial*Neural*Networks 5
- 6.
- 7. WISE:*Wavelet*Image*Spectra* Enhancement* • X.ray*inspection*enhancement*algorithm*based* on*parametric*estimation*of*image • Runs*in*real*time*–not*an*offline*filter • What*is*parameter*estimation? PATENT PENDING Raw data s = A.sin(2πf + ϕ) A = 2 f = 4kHz ϕ = π Parameter Estimation 8
- 8.
- 9.
- 10.
- 11.
- 12.
- 13.
- 14.
- 15.
- 16.
- 17.
- 18. Dual*Energy*X.Ray*Imaging PATENT PENDING Low EnergyImage High Energy Image 19
- 19.
- 20.
- 21.
- 22.
- 23.
- 24.
- 25. ICARUS Image Comparison Algorithmsfor Radiography Unified Software 26
- 26.
- 27.
- 28. ICARUS Image*Comparison*Algorithms*for*Radiography*Unified*Software 1st – SELECT REFERENCE IMAGE 2nd– DEFINE COMPARISON MASK 29
- 29. ICARUS Image*Comparison*Algorithms*for*Radiography*Unified*Software 4th – CORRECTFOR ROTATION AND OFFSET TRANSLATION 3rd – COLLECT IMAGE FROM TAPE 30
- 30. ICARUS Image*Comparison*Algorithms*for*Radiography*Unified*Software 5th – MEASURESIMILARITY BETWEEN REFERENCE AND ACQUIRED IMAGE WITH MASK APPLIED TO BOTH IMAGES REFERENCE ACQUIRED 31
- 31. ICARUS Image*Comparison*Algorithms*for*Radiography*Unified*Software 6th – COMPAREMEASURED SIMILARITY WITH SET THRESHOLD REFERENCE ACQUIRED Example: 98% SIMILARITY: IF < THRESHOLD ! FAIL IF > THRESHOLD ! PASS DIFFERENCES 32
- 32.
- 33. ICARUS Image*Comparison*Algorithms*for*Radiography*Unified*Software 7th– REPEAT PROCESSFOR WHOLE TAPE. STOP ON FAIL 34
- 34. Conclusions • Innovation*in*the*first*century*of*the*x.ray* inspection*market*was*driven*by* hardware*improvements • Massive*computing*power*has*allowed* the*second*century*to*be*driven*by* algorithm*innovations •Expect*to*see*more*algorithm* improvements*this*century 35