Refine
Year of publication
Document Type
- Conference Proceeding (47)
- Article (peer reviewed) (8)
- Contribution to a Periodical (3)
- Part of a Book (2)
- Report (1)
Is part of the Bibliography
- no (61)
Keywords
- Fluorescence (9)
- Fluorescence lifetime imaging (9)
- FD-FLIM (8)
- Phase shift (8)
- fluorescence (8)
- Neural networks (7)
- Microelectronics (6)
- Diode lasers (5)
- Fluoreszenz (4)
- Microscopy (4)
Development of a Neural Network for Automatic Classification of Post-Consumer Wood Using Rapid-FLIM
(2023)
The economic use of wood is a growing sector, not only because of the significant advantage of wood to retain CO 2 . It is crucial to increase the material recycling of wood in several lifecycles, but currently, there is no reliable post-consumer wood sorting technique in line. This research measures 365 post-consumer wood samples of classes A1-4 four times with the frequency domain fluorescence lifetime imaging microscopy (FD-FLIM) subset method rapid fluorescence lifetime imaging microscopy (Rapid-FLIM). The data is analyzed on their statistical features. Four neural networks based on Multilayer perceptron are then trained and tested with twelve statistical features extracted from the Rapid-FLIM images. The best model for this application contains the optimizer RMSprop, the activation function SELU and the loss function binary crossentropy. The best model of this structure could achieve a false positive ratio of 4.79 % over the ten folds.
Environmental pollution by plastics is an increasing problem. However, state-of-the-art methods have significant disadvantages in detecting and identifying plastics directly in an environmental matrix. In this study, we propose a blob detection algorithm in combination with a neural network for fast and automated identification of plastics and non-plastics in a single fluorescence lifetime image. Therefore an artificial environmental matrix is prepared that contains soil, grass, spruce and HDPE (high density polyethylene) particles. Several FD-FLIM (frequency domain fluorescence lifetime imaging microscopy) images are taken, and the detection algorithm and the neural network are applied. We successfully demonstrated the suitability of the thresholding algorithm and the binary classification of the HDPE particles directly in the environmental matrix.
The FD-FLIM technique has a high potential for automated post-consumer wood sorting. A problem of analysing post-consumer wood fluorescence properties is the uncertainty of the post-consumer wood category as the origin of the samples are unknown. In this study, the fluorescence properties of actual post-consumer wood is compared with prepared wood samples. The post-consumer wood samples display slightly different fluorescence intensities and lifetimes due to environmental influences and a higher sample diversity. For improved training of evaluation algorithms for post-consumer wood sorting, the prepared sample set should be extended or the post-consumer wood should be additionally analysed in the laboratory.
Environmental pollution by plastics is a global issue of increasing concern. However, microplastic analysis in complex environmental matrices, such as soil samples, remains an analytical challenge. Destructive mass-based methods for microplastic analysis do not determine plastics’ shape and size, which are essential parameters for reliable ecological risk assessment. By contrast, nondestructive particle-based methods produce such data but require elaborate, time-consuming sample preparation. Thus, time-efficient and reliable methods for microplastic analysis are needed. The present study explored the potential of frequency-domain fluorescence lifetime imaging microscopy (FD-FLIM) for rapidly and reliably identifying as well as differentiating plastics and natural materials from terrestrial environments. We investigated the fluorescence spectra of ten natural materials from terrestrial environments, tire wear particles, and eleven different transparent plastic granulates <5 mm to determine the optimal excitation wavelength for identification and differentiation via FD-FLIM under laboratory conditions. Our comparison of different excitation wavelengths showed that 445 nm excitation exhibited the highest fluorescence intensities. 445 nm excitation was also superior for identifying plastic types and distinguishing them from natural materials from terrestrial environments with a high probability using FD-FLIM. We could demonstrate that FD-FLIM analysis has the potential to contribute to a streamlined and time-efficient direct analysis of microplastic contamination. However, further investigations on size-, shape-, color-, and material-type detection limitations are necessary to evaluate if the direct identification of terrestrial environmental samples of relatively low complexity, such as a surface inspection soil, is possible.
Microplastic pollution of the environment has been extensively studied, with recent studies focusing on the prevalence of microplastics in the environment and their effects on various organisms. Identification methods that simplify the extraction and analysis process to the point where the extraction can be omitted are being investigated, thus enabling the direct identification of microplastic particles. Currently, microplastic samples from environmental matrices can only be identified using time-consuming extraction, sample processing, and analytical methods. Various spectroscopic methods are currently employed, such as micro Fourier-transform infrared, attenuated total reflectance, and micro Raman spectroscopy. However, microplastics in environmental matrices cannot be directly identified using these spectroscopic methods. Investigations using frequency-domain fluorescence lifetime imaging microscopy (FD-FLIM) to identify and differentiate plastics from environmental materials have yielded promising results for directly identifying microplastics in an environmental matrix. Herein, two artificially prepared environmental matrices that included natural soil, grass, wood, and high-density polyethylene were investigated using FD-FLIM. Our first results showed that we successfully identified one plastic type in the two artificially prepared matrices using FD-FLIM. However, further research must be conducted to improve the FD-FLIM method and explore its limitations for directly identifying microplastics in environmental samples.
Previous research on identifying plastic types and differentiating plastics from environmental material is promising by utilizing the specific fluorescence lifetime, but the evaluation still has to be automated. Therefore, an automated Gaussian analysis is developed for evaluating frequency-domain fluorescence lifetime images of plastics and environmental materials. Furthermore, we applied a “Multilayer Perceptron” and “Random Forest Classifier” to the data resulting from the Gaussian analysis of the frequency domain fluorescence lifetime imaging microscopy data. The classification results show high F1-scores, whereby the best “Multilayer Perceptron” and “Random Forest Classifier” achieved an F1-score of 90%. Thus, identifying and differentiating plastics and environmental materials is possible by applying a “Multilayer Perceptron” or “Random Forest Classifier” to the Gaussian-analyzed imaged fluorescence lifetime data.
Texas Instruments' MSP430 Launch Pad is used as a key component for remote labs that complement an introduction course to microcomputer technology in the bachelor program Mechatronic at the University of Applied Sciences in Rosenheim. The remote labs use breakout boards and include a timer experiment with generation of pulse width modulation signals, I2C communication and analog to digital conversion.
The System HAILsens implements a device for Real-Time sensor data processing, running on a Texas Instruments Concerto M3 core. An I2C interface is used for sensor communications. Ethernet with UDP on a custom developed three layered frame organisation of sensor data serves as a link to the main system on board of an aeroplane. For backing up the acquired data, all frames are locally stored on an SD Card. To have the possibility of easily checking the current system status, an LCD module and a push-button provide a minimal user interface.
Comprehensive guide to the process and test methods used for electronic device failure analysis. Contains dozens of articles—contributed by industry experts—covering a wide range of topics involving the failure analysis of microelectronics. It places the most important and up-to-date information on this subject at your fingertips.
Topic coverage includes:
Failure Analysis Process Flow
Failure Verification
Failure Modes and Failure Classification
Special Devices (MEMS, Optoelectronics, Passives)
Fault Localization Techniques: Package Level (NDT)
Die Level (Depackaging, Photon Emission, Microthermography, Laser-Based Methods, Particle Beam Methods)
Deprocessing & Imaging Techniques: Deprocessing
General Imaging Techniques
Local Deprocessing & Imaging
Circuit Edit and Design Modification
Material Analysis Techniques
Reference Information: Important Topics for Semiconductor Devices
Failure Analysis Techniques Roadmap
Failure Analysis Operations and Management
Appendices: Failure Analysis Terms, Definitions, and Acronyms
Industry Standards