Concealed object segmentation in terahertz imaging via adversarial learning

Abstract

Terahertz imaging (frequency between 0.1 to 10 THz) is a modern technique for public security check. Due to poor imaging quality, traditional machine vision methods often fail to detect concealed weapons in Terahertz samples, while modern instance segmentation approaches have complex multiple-stage concatenation and often hunger for massive and accurate training data. In this work, we realize a novel Conditional Generative Adversarial Nets (CGANs), named as Mask-CGANs to segment weapons in such a challenging imaging quality. The Mask-Generator network employs a "selected-connection U-Net" to restrain false alarms and speed up training convergence. The loss function takes reconstruction errors and sparse priors into consideration to preserve precise segmentation. Such a learning architecture works well with a small training dataset. Experiments show that the proposed model outperforms CGANs (more than 16-32 % in Recall, Precision and Accuracy) and Mask-RCNN (more than 3-6 %). Moreover, its testing speed (69.7 FPS ) is fast enough to be implemented in a real-time security check system, which is 44 times faster than Mask-RCNN.In the experiments for mammographic mass segmentation on INBreast dataset, the Dice index of the proposed method is 91.29, surpasses the-state-of-the-art

Color Image Generation from LiDAR Reflection Data by Using Selected Connection UNET

“…First, ED-FCN-based image-generation system [ 10 , 12 ] is analyzed with respect to sparseness and similarity. Then, the conventional SC-UNET architectures used for terahertz image segmentation [ 23 ] is re-purposed and adapted to heterogenous image generation based on the analyses.…”

“…The UNET is also used for the semantic segmentation for the LiDAR reflection data [ 9 ]. Recently selected connection UNET (SC-UNET), shown in Figure 1 c, is proposed to detect the concealed object in the THz image for military purposes [ 23 ], resulting in additional improvement over UNET. Modified UNET and ED-FCN, i.e., UNET++ and RTFNet, are proposed for medical and urban scene semantic segmentation, respectively [ 13 , 14 ].…”

Study on Data Partition for Delimitation of Masses in Mammography

“…In comparison to Liang et al [ 22 ], the method underperformed in segmentation. One of the reasons may be that Liang et al used a very small and imbalanced dataset, consisting of only images with tumoral masses.…”

“…0.90 Zhang et al [21] INbreast n.a. 0.85 Liang et al [22] INbreast n.a. 0.91 Dhungel et al [27] INbreast 0.90 ± 0.02 @ 1.3 0.85 ± 0.02 Al-antari et al [28] INbreast n.a.…”

“…Liang et al [ 22 ] proposed a Conditional Generative Adversarial Network (CGAN) for segmentation of the tumoral masses in a very small dataset using only images with masses. The CGAN consists of two networks, the Mask-Generator and the Discriminator.…”