Search

In public spaces, conducting security checks to detect concealed objects carried on the human body is crucial for enhancing global anti-terrorist measures. Terahertz imaging has recently played a pivotal role in concealed object detection. However, previous studies have faced significant challenges in achieving superior accuracy and performance. To address these issues, we propose a YOLOv5m model for detecting hidden objects beneath human clothing. We employ the CSPDarknet53 block to reduce noise and enhance discriminative power. Object location and size are identified using a PANet and the prediction head. To reduce computational complexity and obtain highly relevant features, we utilize multi-convolutional layers. Duplicate boxes are eliminated and high-quality bounding boxes are accurately detected using the NMS block. Hyper parameter tuning is performed using the Mutation Enabled Salp Swarm Algorithm, resulting in improved detection accuracy and reduced processing time. Our proposed model achieves impressive metrics, including a precision of 98.99%, recall of 97.80%, F1 score of 98.05%, detection rate of 96.50% and execution time of 135 s. Comparatively, our method outperforms existing approaches such as CNN, YOLO3, AC-SDBSCAN, YOLO-v2, RaadNet and SPFAN. We train and test our proposed method using a terahertz video dataset, demonstrating excellent results with high precision. [ABSTRACT FROM AUTHOR]

In video surveillance, real-time detection of hidden or concealed objects under the clothes of humans is a challenging task. Very few research works have been carried out on this hidden object detection and recognition with the aid of various coherent technologies namely Millimeter wave (MMW) imaging, Infrared (IR) imaging, and Terahertz (THz) imaging systems. Hidden objects that include weapons like axe, knife, gun, bomb, pistol, etc., are a major threat to security surveillance as it needs to be recognized in a few seconds irrespective of their size. This paper proposes an efficient hidden object recognition method that focuses on the detection and recognition of concealed weapons in humans using Modified Weighted You Only Look Once v5 (MWYOLOv5) model. The probable occurrence of inaccurate forecast results due to the sense of a low confidence score is mitigated by the use of the weighted boxes fusion (WBF) method. As a result, higher-confidence boxes contribute more to fused box coordinates than lower-confidence boxes. Meanwhile, it is significant to select optimal hyperparameter values for training YOLOv5 model with CSPDarknet53 based feature extractor and Path Aggregation Network (PANet) based feature aggregation procedures to predict concealed objects. To achieve this, a new crossover salp swarm algorithm (CSSA) is developed to tune the YOLO hyperparameters such as learning rate, momentum, weight decay and batch size. This provides more accuracy in recognizing hidden objects on using THz and MMW images on comparison with existing methods. The proposed hazardous weapon recognition model is trained and tested on both MMW and THz Imagery datasets and it shows that the proposed methodology is showing good results with high mAP@.5 and mAP@.5:95. [ABSTRACT FROM AUTHOR]

The study presents the comparison of detection and recognition of concealed objects covered with various types of clothing by using passive imagers operating in a terahertz (THz) range at 1.2 mm (250 GHz) and a mid-wavelength infrared (MWIR) at 3–6 μm (50–100 THz). During this study, large dataset of images presenting various items covered with various types of clothing has been collected. The detection and classification algorithms aimed to operate robustly at high processing speed across these two spectrums. Properties of both spectrums, theoretical limitations, performance of imagers and physical properties of fabrics in both spectral domains are described. The paper presents a comparison of two deep learning–based processing methods. The comparison of the original results of various experiments for the two spectrums is presented. [ABSTRACT FROM AUTHOR]

The use of microwave imaging is becoming more prevalent for detection of interior hidden defects in manufactured and packaged materials. In applications for detection of hidden moisture, microwave tomography can be used to image the material and then perform an inverse calculation to derive an estimate of the variability of the hidden material, such internal moisture, thereby alerting personnel to damaging levels of the hidden moisture before material degradation occurs. One impediment to this type of imaging occurs with nearby objects create strong reflections that create destructive and constructive interference, at the receiver, as the material is conveyed past the imaging antenna array. In an effort to remove the influence of the reflectors, such as metal bale ties, research was conducted to develop an algorithm for removal of the influence of the local proximity reflectors from the microwave images. This research effort produced a technique, based upon the use of ultra-wideband signals, for the removal of spurious reflections created by local proximity reflectors. This improvement enables accurate microwave measurements of moisture in such products as cotton bales, as well as other physical properties such as density or material composition. The proposed algorithm was shown to reduce errors by a 4:1 ratio and is an enabling technology for imaging applications in the presence of metal bale ties.

The use of microwave imaging is becoming more prevalent for detection of interior hidden defects in manufactured and packaged materials. In applications for detection of hidden moisture, microwave tomography can be used to image the material and then perform an inverse calculation to derive an estimate of the variability of the hidden material, such internal moisture, thereby alerting personnel to damaging levels of the hidden moisture before material degradation occurs. One impediment to this type of imaging occurs with nearby objects create strong reflections that create destructive and constructive interference, at the receiver, as the material is conveyed past the imaging antenna array. In an effort to remove the influence of the reflectors, such as metal bale ties, research was conducted to develop an algorithm for removal of the influence of the local proximity reflectors from the microwave images. This research effort produced a technique, based upon the use of ultra-wideband signals, for the removal of spurious reflections created by local proximity reflectors. This improvement enables accurate microwave measurements of moisture in such products as cotton bales, as well as other physical properties such as density or material composition. The proposed algorithm was shown to reduce errors by a 4:1 ratio and is an enabling technology for imaging applications in the presence of metal bale ties. [ABSTRACT FROM AUTHOR]