Techniques for the Automatic Detection and Hiding of Sensitive Targets in Emergency Mapping Based on Remote Sensing Data

Abstract

Emergency remote sensing mapping can provide support for decision making in disaster assessment or disaster relief, and therefore plays an important role in disaster response. Traditional emergency remote sensing mapping methods use decryption algorithms based on manual retrieval and image editing tools when processing sensitive targets. Although these traditional methods can achieve target recognition, they are inefficient and cannot meet the high time efficiency requirements of disaster relief. In this paper, we combined an object detection model with a generative adversarial network model to build a two-stage deep learning model for sensitive target detection and hiding in remote sensing images, and we verified the model performance on the aircraft object processing problem in remote sensing mapping. To improve the experimental protocol, we introduced a modification to the reconstruction loss function, candidate frame optimization in the region proposal network, the PointRend algorithm, and a modified attention mechanism based on the characteristics of aircraft objects. Experiments revealed that our method is more efficient than traditional manual processing; the precision is 94.87%, the recall is 84.75% higher than that of the original mask R-CNN model, and the F1-score is 44% higher than that of the original model. In addition, our method can quickly and intelligently detect and hide sensitive targets in remote sensing images, thereby shortening the time needed for emergency mapping.

Residual Attention Mechanism for Remote Sensing Target Hiding

“…The image inpainting network Cont Atten [9] calculates the contribution of external features to each location in the missing region, and it was applied to the inpainting areas after removing airplanes [7]. This pioneered an automated approach for target hiding.…”

“…A new multisensor dataset of very-high-resolution satellite imagery from diverse landscapes was recently introduced for the super-resolution restoration of high-resolution (HR) remote sensing images from low-resolution (LR) inputs [6]. In addition, to automatically capture the positions and contours of sensitive targets, semantic segmentation was introduced [7]. Its process is the same as that of remote sensing interpretation tasks [8].…”

“…Its process is the same as that of remote sensing interpretation tasks [8]. After detecting sensitive targets, Qiu et al proposed an image inpainting model to remove and fill target areas [7]. This method combines object detection with image inpainting to achieve the fast, automatic hiding of sensitive targets.…”

Extraction of Rural Residential Land from Very-High Resolution UAV Images Using a Novel Semantic Segmentation Framework

“…By dilation convolution, Atrous Spatial Pyramid Pooling (ASPP) may efficiently increase the receptive field, enabling the model to understand multi-scale features, and with certain achievements in practical applications in the field of remote sensing, for example, it can be used for road extraction from remote sensing images [47,48] and for fast and accurate land cover classification [49] on medium-resolution remote sensing images. The PointRend [50] neural network module treats image segmentation as a rendering problem and performs adaptively selected point-based segmentation predictions at adaptively selected locations using an iterative subdivision algorithm, and it has good performance in remote sensing image instance segmentation; for example, the introduction of PointRend in MaskRcnn has improved the accuracy of the edge, and the accuracy and efficiency in emergency remote sensing mapping have been greatly improved compared with traditional methods [51].…”

Hierarchical Self-Attention Embedded Neural Network With Dense Connection for Remote-Sensing Image Semantic Segmentation

“…CNN is a trainable multilayer architecture composed of multiple feature-extraction stages, which learns the representative and discriminative features in a hierarchical manner from the data [15]. Thus, CNN-based techniques have opened a door to strengthen the ability to interpret remote-sensing data, including preprocessing [16][17][18], pixel-based classification [19][20][21], target recognition [22][23][24], scene understanding [25][26][27], and ocean observation [28].…”