KEYWORDS: Lawrencium, 3D modeling, Data modeling, Education and training, Machine learning, Light sources and illumination, 3D image processing, Image classification, Computer vision technology, Animal model studies
Fine-tuning a pretrained model with real data for a machine learning task requires many hours of manual work, especially for computer vision tasks, where collection and annotation of data can be very time-consuming. We present a framework and methodology for synthetic data collection that is not only efficient in terms of time taken to collect and annotate data, making use of free- and open-source software tools and 3D assets but also beats the state-of-the-art against real data, which is the ultimate test for any similar-to-real approach. We test our approach on a set of image classes from ObjectNet, which is a challenging image classification benchmark test dataset that is designed to be similar in many respects to ImageNet but with a wider variety of viewpoints, rotations, and backgrounds, which can make it more difficult for transfer learning problems. The novelty of our approach stems from the way we create complex backgrounds for 3D models using 2D images laid out as decals in a 3D game engine, where synthetic images are captured programmatically with a large number of systematic variations. We demonstrate that our approach is highly effective, resulting in a deep learning model with a top-1 accuracy of 72% on the ObjectNet data, which is a new state-of-the-art result. In addition, we present an efficient strategy for learning rate tuning that is an order of magnitude faster than regular grid search.
The quantitative analysis of illicit materials using Raman spectroscopy is of widespread interest for law enforcement and healthcare applications. One of the difficulties faced when analysing illicit mixtures is the fact that the narcotic can be mixed with many different cutting agents. This obviously complicates the development of quantitative analytical methods. In this work we demonstrate some preliminary efforts to try and account for the wide variety of potential cutting agents, by discrimination between the target substance and a wide range of excipients. Near-infrared Raman spectroscopy (785 nm excitation) was employed to analyse 217 samples, a number of them consisting of a target analyte (acetaminophen) mixed with excipients of different concentrations by weight. The excipients used were sugars (maltose, glucose, lactose, sorbitol), inorganic materials (talcum powder, sodium bicarbonate, magnesium sulphate), and food products (caffeine, flour). The spectral data collected was subjected to a number of pre-treatment statistical methods including first derivative and normalisation transformations, to make the data more suitable for analysis. Various methods were then used to discriminate the target analytes, these included Principal Component Analysis (PCA), Principal Component Regression (PCR) and Support Vector Machines.
The unambiguous identification and quantification of hazardous materials is of increasing importance in many sectors such as waste disposal, pharmaceutical manufacturing, and environmental protection. One particular problem in waste disposal and chemical manufacturing is the identification of solvents into chlorinated or non-chlorinated. In this work we have used Raman spectroscopy as the basis for a discrimination and quantification method for chlorinated solvents. Raman spectra of an extensive collection of solvent mixtures (200+) were collected using a JY-Horiba LabRam, infinity with a 488 nm excitation source. The solvent mixtures comprised of several chlorinated solvents: dichloromethane, chloroform, and 1,1,1-trichloroethane, mixed with solvents such as toluene, cyclohexane and/or acetone. The spectra were then analysed using a variety of chemometric techniques (Principal Component Analysis and Principal Component Regression) and machine learning (Neural Networks and Genetic Programming). In each case models were developed to identify the presence of chlorinated solvents in mixtures at levels of ~5%, to identify the type of chlorinated solvent and then to accurately quantify the amount of chlorinated solvent.
The automated identification and quantification of illicit materials using Raman spectroscopy is of significant importance for law enforcement agencies. This paper explores the use of Machine Learning (ML) methods in comparison with standard statistical regression techniques for developing automated identification methods. In this work, the ML task is broken into two sub-tasks, data reduction and prediction. In well-conditioned data, the number of samples should be much larger than the number of attributes per sample, to limit the degrees of freedom in predictive models. In this spectroscopy data, the opposite is normally true. Predictive models based on such data have a high number of degrees of freedom, which increases the risk of models over-fitting to the sample data and having poor predictive power. In the work described here, an approach to data reduction based on Genetic Algorithms is described. For the prediction sub-task, the objective is to estimate the concentration of a component in a mixture, based on its Raman spectrum and the known concentrations of previously seen mixtures. Here, Neural Networks and k-Nearest Neighbours are used for prediction. Preliminary results are presented for the problem of estimating the concentration of cocaine in solid mixtures, and compared with previously published results in which statistical analysis of the same dataset was performed. Finally, this paper demonstrates how more accurate results may be achieved by using an ensemble of prediction techniques.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.