MFA-net: Object detection for complex X-ray cargo and baggage security imagery

Abstract

Deep convolutional networks have been developed to detect prohibited items for automated inspection of X-ray screening systems in the transport security system. To our knowledge, the existing frameworks were developed to recognize threats using only baggage security X-ray scans. Therefore, the detection accuracy in other domains of security X-ray scans, such as cargo X-ray scans, cannot be ensured. We propose an object detection method for efficiently detecting contraband items in both cargo and baggage for X-ray security scans. The proposed network, MFA-net, consists of three plug-and-play modules, including the multiscale dilated convolutional module, fusion feature pyramid network, and auxiliary point detection head. First, the multiscale dilated convolutional module converts the standard convolution of the detector backbone to a conditional convolution by aggregating the features from multiple dilated convolutions using dynamic feature selection to overcome the object-scale variant issue. Second, the fusion feature pyramid network combines the proposed attention and fusion modules to enhance multiscale object recognition and alleviate the object and occlusion problem. Third, the auxiliary point detection head adopts an auxiliary head to predict the new keypoints of the bounding box to emphasize the localizability without requiring further ground-truth information. We tested the performance of the MFA-net on two large-scale X-ray security image datasets from different domains: a Security Inspection X-ray (SIXray) dataset in the baggage domain and our dataset, named CargoX, in the cargo domain. Moreover, MFA-net outperformed state-of-the-art object detectors in both domains. Thus, adopting the proposed modules can further increase the detection capability of the current object detectors on X-ray security images.

Fig 1. Representative image examples of the GDXray dataset [1].

Fig 2. Two-stage object detector incorporated with the three proposed components (i.e., the multiscale dilated convolutional module, fusion feature pyramid network, and auxiliary point detection head).

Fig 3. Multiscale dilated convolutional module architecture.

Fig 4. Comparison of the architecture of the feature pyramid network (FPN) and the proposed FusionFPN: a) architecture of the FPN, b) architecture of FusionFPN, c) attention module architecture, and d) fusion module architecture.

Fig 5

Architectural and bounding box detection-output comparison between the two-stage object detector shared head and proposed auxiliary point detection head: a) two-stage object detector shared head architecture, b) proposed auxiliary point detection architecture, c) two-stage object detectors: bounding box detection-output points, and d) auxiliary point detection head output points.

Fig 6. Prohibited items and X-ray security images in the CargoX dataset: a) examples of prohibited items for the CargoX dataset and b) image examples for the CargoX dataset.

Fig 7. Visualization of the effect of the proposed MDConv, FusionFPN, point head, and MFA-net with all the proposed modules on CargoX.

Fig 8. Visualization of the effect of the proposed MDConv, FusionFPN, point head, and MFA-net with all the proposed modules on SIXray [2].