A Fast and Compact Hybrid CNN for Hyperspectral Imaging-based Bloodstain Classification

Innovative Vibrational Spectroscopy Research for Forensic Application

“…Moreover, to determine which statistical method works best for analyzing bloodstains, all chemometrics should be compared. Therefore, Butt et al further analyzed the data set compiled by Romaszewski . To differentiate blood from blood-like substances in hyperspectral images, they assessed two convolutional neural network (CNN) models: a fast, compact 3D CNN model and a hybrid (3D and 2D) CNN model.…”

Teaching old presumptive tests new digital tricks with computer vision for forensic applications

“…8 It is with these colorimetric frameworks that, within chemistry alone, computer vision has blossomed as an independent analytical approach for non-contact determination of many phenomena. face detection, 13 security screening, 13 blood stain characterisation, [14][15][16] identication of chemical warfare agents, 17 quantifying explosives, 18 and ballistics. 19 More specically within presumptive testing, smart phoneenabled single imaging approaches have been applied to, for example, recreational drug detection in drinks, 20 risk-mitigating false positives in cocaine detection, 21 bodily uid identication, 15 digitalising amphetamine spot tests, 22 and micro-device fabrication for more objective spot testing.…”

“…The use of digital imaging in forensic applications is well established across a range of use cases. Applications span handwriting recognition, 12 face detection, 13 security screening, 13 blood stain characterisation, 14–16 identification of chemical warfare agents, 17 quantifying explosives, 18 and ballistics. 19…”

Teaching Old Presumptive Tests New Digital Tricks with Computer Vision for Forensic Applications

“…The use of digital imaging in forensic applications is well established across a range of use cases. Applications span handwriting recognition, 12 face detection, 13 security screening, 13 blood stain characterisation, [14][15][16] identification of chemical warfare agents, 17 quantifying explosives, 18 and ballistics. 19 More specifically within presumptive testing, smart phoneenabled single imaging approaches have been applied to, for example, recreational drug detection in drinks, 20 risk-mitigating false positives in cocaine detection, 21 bodily fluid identification, 15 digitalising amphetamine spot tests, 22 and micro-device fabrication for more objective spot testing.…”