선택한 단어 수는 입니다.
최소 단어 이상 선택하여야 합니다.
최소 단어 이상 선택하여야 합니다.
선택한 단어 수는 30입니다.
최대 10 단어까지만 선택 가능합니다.
최대 10 단어까지만 선택 가능합니다.
제안 방법
- In this study, the active infrared thermography system with a baseline-free defect detection algorithm is developed to automatically and instantaneously detect and visualize subsurface (below 150 µm ~ 300 µm from the inspection surface) micro (150 µm ~ 300 µm in diameter) voids in epoxy molding compounds (EMCs) used in electronic chip packaging.
대상 데이터
- 1 shows an overall schematic of the active infrared thermography system composed of excitation, sensing, and control units. The excitation unit consists of an arbitrary wave generator (AWG) and a quartz halogen lamp. AWG generates a modulated voltage signal according to a control signal sent by the control unit.
- Note that the quartz halogen lamp which has the wavelength range of near infrared is useful for penetrating much deeper into a specimen because shorter wavelengths are likely to lead to more efficient thermal wave generation in the thickness direction [12]. The sensing unit is comprised of an infrared camera with a close-up lens. The infrared camera with an uncooled micro bolometer detector is used to capture 2D raw thermal images of the radiated heats from the EMC surface.
- 4 (a) and (b). Total 36 EMC specimens are tested and labeled as EMC 1 to EMC 36. EMC 1 to EMC 18 have subsurface micro voids, and EMC 19 to EMC 36 are intact.
이론/모형
- The extreme value statistics is a branch of statistics that models the statistical properties of extreme values in either the upper or lower tails of sampled data [16]. Among available extreme value distributions, a Gumbel distribution is selected in this study to characterize the upper tail distribution of the processed lock-in amplitude values based on goodness-of-fit analysis. Once the best-fit Gumbel distribution is estimated, the threshold value corresponding to a one-sided 99% confidence interval is determined.
- The infrared camera is 50 mm apart from the test specimen, and acquires raw thermal images with a sampling rate of 50 Hz. Once the thermal responses are measured, the data is transmitted to the control computer and processed by the proposed baseline-free subsurface micro void visualization algorithm described in the following section.
- This temperature variation over the surface is reflected in the measured raw thermal images. The raw thermal images are processed by the proposed baseline-free subsurface micro void visualization algorithm to extract and isolate void-induced perturbation. Such baseline-free diagnostic is able to provide instantaneous diagnosis and minimize false alarms caused by irregular surface conditions of the test specimen and arbitrary noise components of the raw thermal images, making it suitable for online inspection.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.