An auto‐classification procedure for concealed weapon detection in millimeter‐wave radiometric imaging systems

Işıker, Hakan; Unal, Ilhami; Tekbaş, Mustafa; Özdemir, Caner

doi:10.1002/mop.31005

Cited by 15 publications

(11 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The image threshold-based methods mostly use the image gray histogram to determine the segmentation threshold, according to which the body and objects can be separated from each other, and then classify the objects. For example, both [ 20 , 21 ] proposed a two-level thresholding method for estimating the size of the concealed objects, where the lower threshold determines the regions of the human body and the higher one is used to segment the concealed objects. The advantage of these methods is the simple operation and low computational complexity.…”

Section: Related Workmentioning

confidence: 99%

CNN with Pose Segmentation for Suspicious Object Detection in MMW Security Images

Meng

Zhang

Wang

2020

Sensors

View full text Add to dashboard Cite

Millimeter-wave (MMW) imaging scanners can see through clothing to form a three-dimensional holographic image of the human body and suspicious objects, providing a harmless alternative for non-contacting searches in security check. Suspicious object detection in MMW images is challenging, since most of them are small, reflection-weak, shape, and reflection-diverse. Conventional detectors with artificial neural networks, like convolution neural network (CNN), usually take the problem of finding suspicious objects as an object recognition task, yielding difficulties in developing large-amount and complete sample sets of objects. In this paper, a new algorithm is developed using the human pose segmentation followed by the deep CNN detection. The algorithm is emphasized to learn the similarity with humans’ body clutter applied to training corresponding CNNs after the image segmentation base of the pose estimation. Moreover, the suspicious object recognition in the MMW image is converted to a binary classification task. Instead of recognizing all sorts of suspicious objects, the CNN detector determines whether the body part images present the abnormal patterns containing suspicious objects. The proposed algorithm that is based on CNN with the pose segmentation has concise configuration, but optimal performance in the suspicious object detection. Extensive experiments confirm the effectiveness and superiority of the proposal.

show abstract

Section: Related Workmentioning

confidence: 99%

CNN with Pose Segmentation for Suspicious Object Detection in MMW Security Images

Meng

Zhang

Wang

2020

Sensors

View full text Add to dashboard Cite

show abstract

“…Although OTSU method is an effective method in image processing, it is mostly used in Hough-Saturation-Value (HSV) regions of (Red, Green, and Blue) RGB images (Sabancı et al2018). In many radar imaging applications where the OTSU method was used, the operations were executed through the RGB image files (Işıker and Özdemir, 2019;Işıker et al, 2018;Işıker et al, 2015;Khoukhi et al2019;Lu and Hu 2012). However, converting SAR data into image files and then filtering these files again with image processing causes high computation time and data loss.…”

Section: Introductionmentioning

confidence: 99%

Clutter removal in millimeter wave GB-SAR images using OTSU’s thresholding method

Yiğit

Demirci

Özdemir

2022

International Journal of Engineering and Geosciences

View full text Add to dashboard Cite

In this study, a filtering method based on the threshold value of normalized synthetic aperture radar (SAR) data is proposed to eliminate clutter in millimeter wave ground based synthetic aperture radar (GB-SAR) images. In the proposed method, first, stepped frequency continuous wave SAR data are reconstructed by using the back-projection algorithm and focused complex SAR data are obtained. Then, the amplitude values of the complex SAR data are normalized and the best threshold values to distinguish the target from clutter is determined by the OTSU's thresholding method. Next, a filter mask is created that cancels all data below the computed threshold values. The mask matrix is finally multiplied with the resulted GB-SAR data to eliminate all clutter from the image. With the proposed technique, the best threshold value is determined automatically by directly processing the raw data without converting the SAR data into any RGB images. The proposed technique is validated through real GB-SAR experiments that were carried out in the frequency band of 78-81 GHz. In the experiments, challenging GB-SAR data are obtained using high cluttered background materials, and very successful filtering operations are performed with the proposed technique.

show abstract

“…Upon the introduction of the concept of Multiple-Input Multiple-Output (MIMO), SAR imagingbased security inspection systems are also in development, which greatly improves the imaging accuracy and rate [4][5][6][7][8][9]. Different from X-ray, active millimeter-wave security equipment does not cause harm to the human body [10].…”

Section: Introductionmentioning

confidence: 99%

A Suspicious Multi-Object Detection and Recognition Method for Millimeter Wave SAR Security Inspection Images Based on Multi-Path Extraction Network

Yuan

Zhang

et al. 2021

Remote Sensing

View full text Add to dashboard Cite

There are several major challenges in detecting and recognizing multiple hidden objects from millimeter wave SAR security inspection images: inconsistent clarity of objects, similar objects, and complex background interference. To address these problems, a suspicious multi-object detection and recognition method based on the Multi-Path Extraction Network (MPEN) is proposed. In MPEN, You Only Look Once (YOLO) v3 is used as the base network, and then the Multi-Path Feature Pyramid (MPFP) module and modified residual block distribution are proposed. MPFP is designed to output the deep network feature layers separately. Then, to distinguish similar objects more easily, the residual block distribution is modified to improve the ability of the shallow network to capture details. To verify the effectiveness of the proposed method, the millimeter wave SAR images from the laboratory’s self-developed security inspection system are utilized in conducting research on multi-object detection and recognition. The detection rate (probability of detecting a target) and average false alarm (probability of error detection) rate of our method on the target are 94.6% and 14.6%, respectively. The mean Average Precision (mAP) of recognizing multi-object is 82.39%. Compared with YOLOv3, our method shows a better performance in detecting and recognizing similar targets.

show abstract

An auto‐classification procedure for concealed weapon detection in millimeter‐wave radiometric imaging systems

Cited by 15 publications

References 32 publications

CNN with Pose Segmentation for Suspicious Object Detection in MMW Security Images

CNN with Pose Segmentation for Suspicious Object Detection in MMW Security Images

Clutter removal in millimeter wave GB-SAR images using OTSU’s thresholding method

A Suspicious Multi-Object Detection and Recognition Method for Millimeter Wave SAR Security Inspection Images Based on Multi-Path Extraction Network

Contact Info

Product

Resources

About