Search

Abstract Light field imaging contains abundant scene structure information, which can improve the accuracy of salient object detection in challenging tasks and has received widespread attention. However, how to apply the abundant information of light field imaging to salient object detection still faces enormous challenges. In this paper, the lightweight attention and residual convLSTM network is proposed to address this issue, which is mainly composed of the lightweight attention-based feature enhancement module (LFM) and residual convLSTM-based feature integration module (RFM). The LFM can provide an attention map for each focal slice through the attention mechanism to focus on the features related to the object, thereby enhancing saliency features. The RFM leverages the residual mechanism and convLSTM to fully utilize the spatial structural information of focal slices, thereby achieving high-precision feature fusion. Experimental results on three publicly available light field datasets show that the proposed method surpasses the existing 17 state-of-the-art methods and achieves the highest score among five quantitative indicators.

A 54-year-old woman was admitted to the hospital with a left neck mass. Enhanced CT and ultrasound examinations revealed a lesion in the left sternocleidomastoid muscle. The patient undergone right thyroid lobe resection 8 years ago. Interestingly, the lesion on the sternocleidomastoid muscle, along with the left lobe of the patient’s thyroid, visually appears to form a displaced and complete thyroid in the early Tc-99m-MIBI parathyroid scintigraphy. Combined with Tc-99m-MIBI scintigraphy and abnormal PTH and blood calcium levels, the consideration was given to the lesion in the sternocleidomastoid muscle as an ectopic parathyroid adenoma. Subsequent surgical pathology confirmed this suspicion.

In the process of acquiring 3D point cloud data, due to environmental interference or unstable scanning equipment, the acquired data often have noisy points. Recently, with the development of neural networks for point clouds, great progress has been made in deep learning-based point cloud denoising. However, most of the existing methods adopt a pointnet-like structure to predict point offsets. Simple pooling operation loses much important information, such as local neighborhood information, and global information. The loss of information makes the algorithm ineffective when dealing with some complex cases. In order to solve the above problems, we propose a self-attention-based point cloud denoising network architecture, through the Transformer structure, to establish long-range dependencies of the points. In addition, we propose a local information embedding module to fast select meaningful points and serve as the input of the Transformer. We also consider the correlation between channels of point cloud features and further introduce a channel attention module. Extensive experiments prove that our method outperforms existing methods, and maintains a high running speed.

This paper mainly studies the motion and control of the drag-free spacecraft system for a high-precision mission. In high-precision missions, the influence of three-body gravitational perturbation on a drag-free spacecraft system is still lack of relevant research. To this end, we first established a relative motion model of the test mass and the spacecraft based on multibody dynamics, and the equations of the relative motion of the test mass and the spacecraft cavity were obtained through coordinate transformation. The difference between the sun orbit and the Earth orbit is comparatively analyzed by a numerical solution. The influence of gravitational forces from the Earth, the moon, and the sun on the relative motion is studied in detail. Secondly, a sliding mode controller is designed for the high-precision mission. Finally, the performance of the controller is analyzed from the time domain and the frequency domain. Numerical simulation results show that the controller can meet the requirements of the time domain and the frequency domain.

Laser displacement sensors are widely used in the aviation industry for the purpose of surface normal measurements. The measurement of a surface normal depends on prior knowledge of the poses and positions of the sensors, which are obtained through calibration. This paper introduces a new parameter to the traditional calibration procedure, to reduce the calibration error, and explores the factors affecting calibration using the Monte Carlo method. In the experiment, the normal measurement error of the probe consisted of four sensors after calibration was less than 0.1∘, which satisfied the established requirements. This paper indicates the boundary conditions for a successful calibration and validates the proposed method, which provides a new method for the pose and position calibration of laser displacement sensors and other similar sensors.

Materials of satellite integration truss frame are required to withstand temperature that range from about – 250 °C ~ + 150 °C. In order to reduce structural components deformation caused by such temperature change, material of truss frame mostly adopts laminated composite material tubes, whose linear thermal expansion coefficient (LTEC) is very small. Therefore, accurate measurement of LTEC of truss frame materials over a broad temperature range is essential for successful mission. To address this issue, this paper proposes a general experiment platform for measuring LTEC of laminated composite material specimen reaching length up to one meter in the temperature range from – 100 °C to +100 °C. The platform uses light-density optical fiber probe to measure length variation and thermocouple to record temperature variation. Thereafter, the thermal expansion coefficient and its measurement uncertainty can be obtained by establishing and solving mathematical model. Finally, LTEC measurement of a tubular composite materials specimen is conducted. The experiment result demonstrates the validity and practicality of the experiment platform and the measurement accuracy of LTEC which can reach up to 10-7/°C.DOI: http://dx.doi.org/10.5755/j01.ms.21.4.9708

Abstract The study proposed a novel multiple-image encryption scheme using spatial multiplexing under the illumination of a multi-petals structured light. The security is enhanced through the multi-petals structured light with coherent superposition of several vortex beams with different topological charges. Through the system, an image array is formed behind the sub-petals of the complex structured light. Using Fresnel diffraction, each image is encrypted into the ciphertext by the Fibonacci Lucas transform. The multi-petals structured light with different orbital angular momentum can integrate more system parameters as the additional keys resulting in rich key space with improved security level of the cryptosystem. The feasibility and security of the proposed methods are validated through both numerical simulation and experiments. The research has strong and practical applications for three-dimensional information encryption system.