A dataset for evaluating blood detection in hyperspectral images.

The sensitivity of imaging spectroscopy to haemoglobin derivatives makes it a promising tool for detecting blood. However, due to complexity and high dimensionality of hyperspectral images, the development of hyperspectral blood detection algorithms is challenging. To facilitate their development, we present a new hyperspectral blood detection dataset. This dataset, published under an open access license, consists of multiple detection scenarios with varying levels of complexity. It allows to test the performance of Machine Learning methods in relation to different acquisition environments, types of background, age of blood and presence of other blood-like substances. We have explored the dataset with blood detection experiments, for which we have used a hyperspectral target detection algorithm based on the well-known Matched Filter detector. Our results and their discussion highlight the challenges of blood detection in hyperspectral data and form a reference for further works.

Blood Stain Classification with Hyperspectral Imaging and Deep Neural Networks.

In recent years, growing interest in deep learning neural networks has raised a question on how they can be used for effective processing of high-dimensional datasets produced by hyperspectral imaging (HSI). HSI, traditionally viewed as being within the scope of remote sensing, is used in non-invasive substance classification. One of the areas of potential application is forensic science, where substance classification on the scenes is important. An example problem from that area-blood stain classification-is a case study for the evaluation of methods that process hyperspectral data. To investigate the deep learning classification performance for this problem we have performed experiments on a dataset which has not been previously tested using this kind of model. This dataset consists of several images with blood and blood-like substances like ketchup, tomato concentrate, artificial blood, etc. To test both the classic approach to hyperspectral classification and a more realistic application-oriented scenario, we have prepared two different sets of experiments. In the first one, Hyperspectral Transductive Classification (HTC), both a training and a test set come from the same image. In the second one, Hyperspectral Inductive Classification (HIC), a test set is derived from a different image, which is more challenging for classifiers but more useful from the point of view of forensic investigators. We conducted the study using several architectures like 1D, 2D and 3D convolutional neural networks (CNN), a recurrent neural network (RNN) and a multilayer perceptron (MLP). The performance of the models was compared with baseline results of Support Vector Machine (SVM). We have also presented a model evaluation method based on t-SNE and confusion matrix analysis that allows us to detect and eliminate some cases of model undertraining. Our results show that in the transductive case, all models, including the MLP and the SVM, have comparative performance, with no clear advantage of deep learning models. The Overall Accuracy range across all models is 98-100% for the easier image set, and 74-94% for the more difficult one. However, in a more challenging inductive case, selected deep learning architectures offer a significant advantage; their best Overall Accuracy is in the range of 57-71%, improving the baseline set by the non-deep models by up to 9 percentage points. We have presented a detailed analysis of results and a discussion, including a summary of conclusions for each tested architecture. An analysis of per-class errors shows that the score for each class is highly model-dependent. Considering this and the fact that the best performing models come from two different architecture families (3D CNN and RNN), our results suggest that tailoring the deep neural network architecture to hyperspectral data is still an open problem.

Application of hyperspectral imaging and machine learning methods for the detection of gunshot residue patterns.

Advanced image processing algorithms can support the forensic analyst to make tasks like detection, pattern comparison or identification more objective. In the case of the gunshot residue (GSR) analysis, the automatic detection of potential GSR samples can support the task of evidence collection or analysis of residue needed e.g. for a muzzle-to-target firing distance estimation. In this paper we investigate the application of a hyperspectral camera and two well-known Machine Learning algorithms to automatically indicate the potential presence of GSR samples in a scene containing cloth fabrics. For this study we have created and annotated a hyperspectral image dataset consisting of GSR samples present on multiple fabric types. The GSR samples were obtained using two types of ammunition, discharged from two shooting distances. We have investigated two detection scenarios: an unsupervised anomaly detection (with the RX detector) and a supervised pixel classification (with the SVM classifier). Our results show that an accurate detection is possible in both cases. We also note that in this setting the anomaly detection approach usually requires an image normalisation, while the classifier does not require a fabric-specific information. As an addition, we show that the hyperspectral imaging generally outperforms the RGB imaging in terms of GSR detection accuracy. While the actual verification on presence of GSR on the scene requires an analyst and secondary identification methods, the hyperspectral camera with image processing algorithms can be a valuable tool supporting the evidence collection and analysis.