Detection and Modeling of Alcohol Intoxication Dynamic from IR Images Based on Clustering Driven by ABC Algorithm

Alcohol detection is a challenging issue due to many aspects, especially to security reasons. Conventional measuring systems usually utilize a direct contact with the human body to obtain on spot alcohol level estimation. Nevertheless, it is well known that there are several side effects including the facial temperature distribution for alcohol detection. Since the facial temperature map is observable from the infrared (IR) records, we have performed a set of experimental measurements allowing for dynamical tracking of time-dependent effect of the alcohol intoxication. In this paper, we have proposed the clustering multiregional segmentation driven by the genetic optimization, particularly the Artificial Bee Colony (ABC) algorithm for the facial IR segmentation. The genetic optimization determines an optimal distribution of the initial cluster’s centroids, which represent the main part of a proper clustering. Based on the segmentation procedure, we have proposed a dynamical model allowing for prediction of time-dependent alcohol intoxication features.

[1]  Jan Kubicek,et al.  Analysis and modeling of alcohol intoxication from IR images based on multiregional image segmentation and correlation with breath analysis , 2017, 2017 IEEE Conference on Big Data and Analytics (ICBDA).

[2]  K. Fromme,et al.  GABRA2, Alcohol, and Illicit Drug Use: An Event-Level Model of Genetic Risk for Polysubstance Use , 2018, Journal of abnormal psychology.

[3]  M. Gamer,et al.  Alcohol facilitates detection of concealed identity information , 2018, Scientific Reports.

[4]  Xi Zhu,et al.  Random forest based classification of alcohol dependence patients and healthy controls using resting state MRI , 2018, Neuroscience Letters.

[5]  R. Boom,et al.  A predictive model for flavor partitioning and protein-flavor interactions in fat-free dairy protein solutions. , 2018, Food research international.

[6]  Silvia Conforto,et al.  Pre-Processing Effect on the Accuracy of Event-Based Activity Segmentation and Classification through Inertial Sensors , 2015, Sensors.

[7]  Pengfei Wang,et al.  Infrared and visible image fusion based on NSCT and stacked sparse autoencoders , 2018, Multimedia Tools and Applications.

[8]  Amel Benazza-Benyahia,et al.  Efficient transform-based texture image retrieval techniques under quantization effects , 2016, Multimedia Tools and Applications.

[9]  Yu. B. Blokhinov,et al.  РАЗРАБОТКА АЛГОРИТМА СЕМАНТИЧЕСКОЙ СЕГМЕНТАЦИИ АЭРОФОТОСНИМКОВ РЕАЛЬНОГО ВРЕМЕНИ , 2018 .

[10]  Ming Zhu,et al.  Multiscale Fully Convolutional Network for Foreground Object Detection in Infrared Videos , 2018, IEEE Geoscience and Remote Sensing Letters.

[11]  J. Rehm,et al.  Unrecorded alcohol use: a global modelling study based on nominal group assessments and survey data , 2018, Addiction.

[12]  Paweł Pawłowski,et al.  Efficient pedestrian detection with enhanced object segmentation in far IR night vision , 2017, 2017 Signal Processing: Algorithms, Architectures, Arrangements, and Applications (SPA).

[13]  Silvia Conforto,et al.  Analysis of different image-based biofeedback models for improving cycling performances , 2012, Electronic Imaging.

[14]  M. Cao,et al.  New models for predicting workability and toughness of hybrid fiber reinforced cement-based composites , 2018, Construction and Building Materials.

[15]  Marek Penhaker,et al.  Detection and Dynamical Tracking of Temperature Facial Distribution Caused by Alcohol Intoxication with Using of Modified OTSU Regional Segmentation , 2018, ACIIDS.

[16]  Yanfeng Wu,et al.  A novel fast detection method of infrared LSS-Target in complex urban background , 2017, Int. J. Wavelets Multiresolution Inf. Process..