Hyperspectral Imaging for Bloodstain Identification

Abstract

Blood is key evidence to reconstruct crime scenes in forensic sciences. Blood identification can help to confirm a suspect, and for that reason, several chemical methods are used to reconstruct the crime scene however, these methods can affect subsequent DNA analysis. Therefore, this study presents a non-destructive method for bloodstain identification using Hyperspectral Imaging (HSI, 397-1000 nm range). The proposed method is based on the visualization of heme-components bands in the 500-700 nm spectral range. For experimental and validation purposes, a total of 225 blood (different donors) and non-blood (protein-based ketchup, rust acrylic paint, red acrylic paint, brown acrylic paint, red nail polish, rust nail polish, fake blood, and red ink) samples (HSI cubes, each cube is of size 1000 × 512 × 224, in which 1000 × 512 are the spatial dimensions and 224 spectral bands) were deposited on three substrates (white cotton fabric, white tile, and PVC wall sheet). The samples are imaged for up to three days to include aging. Savitzky Golay filtering has been used to highlight the subtle bands of all samples, particularly the aged ones. Based on the derivative spectrum, important spectral bands were selected to train five different classifiers (SVM, ANN, KNN, Random Forest, and Decision Tree). The comparative analysis reveals that the proposed method outperformed several state-of-the-art methods.

Keywords: ANNs; SVM; bloodstains identification; hyperspectral imaging; weak bands.

Figure 1

Blood sample deposited on a wall sheet.

Figure 2

Blood spectrum.

Figure 3

Spectral signature of few pixels: (a) Reflectance spectra with noise. (b) Reflectance spectra after smoothing.

Figure 4

Mean reflectance spectra of different blood donor samples and Savitzky Golay second-order derivative spectra comparison over White Fabric, Whtile Tile, and Wall Sheet.

Figure 5

Mean reflectance spectra of aged blood samples of donor 1 and Savitzky Golay second order derivative spectra comparison over white fabric, white tile, and wall sheet.

Figure 6

Mean reflectance spectra and Savitzky Golay second order derivative spectra of blood and non-blood samples comparison on white fabric, white tile, and wall sheet.

Figure 7

Comparison of spectra of aged blood and different non-blood samples on wall sheet.

Figure 8

Prediction results of test samples on white fabric: (1) True color image, (2) pre-processed image, (3) labeling, and (4) classifier results: (a) SVM, (b) DT, (c) KNN, and (d) ANN.

Figure 9

Prediction results of test samples on white tile: (1) True color image, (2) pre-processed image, (3) labeling, and (4) classifier results: (a) SVM, (b) DT (c) KNN, and (d) ANN.

Figure 10

Prediction results of test samples on wall sheet: (1) True color image, (2) pre-processed image, (3) labeling, (4) classifier results: (a) SVM, (b) DT, (c) KNN, and (d) ANN.