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

Abstract

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.

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

“…The current related studies mainly focus on the detection of caches in the human body, and there are relatively few studies on the identification of cache types. Compared with the [14,16,17] network, we found that the three researchers mainly detected hidden objects, and did not detect the shape and size of the hidden objects, so the detection accuracy of this type of research is relatively high. MPEN can detect multi-object at the same time, and similar objects (hammer and wrench are listed in this article) can be well distinguished again.…”

“…Chen et al [12,13] use multi-scale fusion to enhance the method of extracting features to achieve a high accuracy detection of bridges and aircrafts. Meng et al [14] propose a human pose segmentation algorithm based on deep Convolutional Neural Network (CNN) detection, which enables human images to be divided into several parts for recognition. Lopez-Tapia et al [15] enhance the passive millimeter wave image to improve the detection rate of the target.…”

Coded-Aperture Computational Millimeter-Wave Image Classifier Using Convolutional Neural Network

“…Although most of the work has shown that CNN is a very useful approach in mmW object detection, almost all of the work available in the literature focuses solely on the learning algorithm and not on the data collection for training. For instance, in [51], CNN was used for object detection in human-scaled targets. To achieve this, an experimental radar system consisting of an array of antennas and operating at 27 GHz with a bandwidth of 5 GHz was leveraged to collect approximately 3000 multiangle mmW human images.…”

Modulation Recognition of Radar Signals Based on Adaptive Singular Value Reconstruction and Deep Residual Learning

“…With the rapid development of artificial intelligence, deep learning has contributed to significant breakthroughs in speech recognition, natural language processing and image recognition [ 10 , 11 , 12 , 13 , 14 ]. This is mainly due to its ability to automatically learning features from the input.…”