Foreground detection, which extracts moving objects from videos, is an important and fundamental problem of video analysis. Classic methods often build background models based on some hand-craft features. Recent deep neural network (DNN) based methods can learn more effective image features by training, but most of them do not use temporal feature or use simple hand-craft temporal features. In this paper, we propose a new dual multi-scale 3D fully-convolutional neural network for foreground detection problems. It uses an encoder–decoder structure to establish a mapping from image sequences to pixel-wise classification results. We also propose a two-stage training procedure, which trains the encoder and decoder separately to improve the training results. With multi-scale architecture, the network can learning deep and hierarchical multi-scale features in both spatial and temporal domains, which is proved to have good invariance for both spatial and temporal scales. We used the CDnet dataset, which is currently the largest foreground detection dataset, to evaluate our method. The experiment results show that the proposed method achieves state-of-the-art results in most test scenes, comparing to current DNN based methods.

The efficiency and the effectiveness of railway intrusion detection are crucial to the safety of railway transportation. Most current methods of railway intrusion detection or obstacle detection are inappropriate for large-scale applications due to their high cost or limited coverage. In this study, we present a fast and low-cost solution to intrusion detection of high-speed railways. As the solution to heavy computational burdens in the current convolutional-neural-network-based detection methods, the proposed method is mainly a novel neural network based on the SSD framework, which includes a feature extractor using an improved MobileNet and a lightweight and efficient feature fusion module. In addition, aiming to improve the detection accuracy of small objects, the feature map weights are introduced through convolution operation to fuse features at different scales. TensorRT is employed to optimize and deploy the proposed network in the low-cost embedded GPU platform, NVIDIA Jetson TX2, to enhance the efficiency. The experimental results show that the proposed methods achieved 89% mAP on the railway intrusion detection dataset, and the average processing time for a single frame was 38.6 ms on the Jetson TX2 module, which satisfies the need of real-time processing.

Foreground detection for infrared (IR) videos is an important and fundamental problem in many applications, e.g., IR surveillance, IR object tracking, and so on. Conventional foreground detection algorithms developed for visible videos do not focus on the problems for IR videos, e.g., low contrast, coarse texture, lack of color information, and so on. Recent foreground detection methods based on deep neural network (DNN) demonstrated significant improvement, but most of them still use only spatial features, which is less obvious in IR images. In this letter, we add deeply learned multiscale temporal features to improve the performance of background subtraction for IR videos. We propose a novel multiscale 3-D fully convolutional network (MFC3-D) to establish a mapping from image sequences to pixelwise classification results and to learn deep and hierarchical multiscale spatial-temporal features of the input images sequence. The experimental results show that the MFC3-D can learn spatial-temporal features effectively and achieved state-of-the-art results on the test data set, comparing to other DNN-based methods and traditional background subtraction methods.

As the primary safety protection system for particle accelerators, the Personnel Protection System (PPS) not only prioritizes safety but also recognizes the importance of system reliability to ensure the stable operation of the accelerator and foster trust among system stakeholders. When designing the PPS, a distinction can be made between critical safety functions and common functions. There are at least three critical safety functions: firstly, ensuring that personnel are prohibited from accessing the tunnel during accelerator operation; secondly, preventing accelerator activation in the presence of personnel within the tunnel; and thirdly, enabling prompt termination of accelerator operations during emergency situations. To meet the demanding safety integrity level requirement of SIL3, multiple layers of defense and safety redundancy are employed. In order to mitigate the potential risks arising from human errors, significant emphasis is placed on improving the automation and intelligence of the system. This research focuses on the development of a comprehensive novel system that incorporates multiple defense mechanisms to meet the stringent safety and reliability demands of large particle accelerators represented by High Energy Proton Source (HEPS). It designs novel local optimization schemes, including the semi-redundant control system, zero-count interlock, intelligent search and secure, and electromagnetic control key, among others. These pioneering approaches are seamlessly integrated into the comprehensive system, effectively enhancing its capabilities and overall efficiency. Furthermore, this study conducts an analysis of the reliability of interlocking equipment, employing measures such as hierarchical DC power, parallel redundancy, and optimized signal processing to enhance its overall reliability. Finally, a method for determining the Safety Integrity Level (SIL) and assessing the reliability of PPS is developed, and the system's performance is validated.