Passive millimeter wave image based security system for concealed threat identification using focus measures

Rishi Raj Sharma,S S Hemanth K Rao Adigarla,Om Mishra

DOI: https://doi.org/10.1007/s11042-024-20449-8

IF: 2.577

2024-11-20

Multimedia Tools and Applications

Abstract:Concealed object identification is very essential for security at the check point. In such scenarios passive millimeter wave (PMMW) imaging can be employed to detect the hidden objects without intruding the privacy of the person being scanned. Due to low resolution, lack of position, size, and shape information of the concealed object, the threat identification process becomes extremely challenging. This paper presents a technique for concealed objects detection using focus measure operators (FMO) studied for normal and threat regions. The probability density function of concealed object area-based patch size optimization method is suggested. Patch information of a PMMW image-based number of threats identification method is developed. The FMOs are computed for every patch and given to classifier to train the model. Quantum support vector machine (QSVM) and K-nearest neighbour (KNN) classifiers are employed in the proposed model. Results obtained using optimal patch are compared with the varying patch size and justified. The proposed method identifies the threat patch with 92.5% accuracy using ten features with KNN classifier and 75% accuracy using two features with QSVM classifier. The proposed system performs better than the compared methods.

computer science, information systems, theory & methods,engineering, electrical & electronic, software engineering