Search

A lightweight semantic segmentation model Faster DeeplabV3+ based on DeeplabV3+ network was improved to address the issues of large parameter count, slow computation speed, and low efficiency in traditional semantic segmentation models. The Faster DeeplabV3 + model used lightweight MobilenetV2 instead of Xception as the backbone feature extraction network, significantly reducing the number of parameters and improving computational speed. Using the combination of deep separable convolution (DSC) and air space pyramid expansion convolution (ASPP), a new depth separable dilated convolution (DSD-Conv) was developed, namely, the depth separable empty space pyramid module (DP-ASPP), which expanded the receptive field while reducing the original convolution parameters and facilitating faster operation. In order to enhance the sensitivity and accuracy of the network in extracting feature information from different dimensions, a dual attention mechanism module was added for processing low-level and high-level feature maps generated in the coding region. Two loss functions, cross entropy and Dice Loss, were integrated to provide more comprehensive and diverse optimization for the model. The improved model was tested on the PASCAL VOC 2012 dataset. The experimental results show that the average intersection to union ratio increases from 76.57% to 79.07%, and the segmentation accuracy increases from 91.2% to 94.3%. The network parameter count (parameters) of the improved model is reduced by 3.86×106, and the floating-point computational load (GFLOPs) is decreased by 117.98 G. The proposed algorithm significantly reduces parameter count and improves computational speed, while also optimizing the segmentation effect. [ABSTRACT FROM AUTHOR]

In order to investigate the effects of freeze-thaw action on physical and mechanical properties and internal fracture damage of sandstone, freeze-thaw cycle tests were carried out on sandstone at different temperature range, and acoustic emission tests were carried out on freeze-thaw damaged rocks under uniaxial loading, so as to study the influence of freeze-thaw temperature range on physical and mechanical parameters and acoustic emission characteristics of rock. The results show that the temperature range increases with the freeze-thaw cycle, the mass loss rate and the change rate of P-wave velocity on the whole show a trend of first steep and then slowly rising, declining peak intensity and peak strain of the sandstone gradually increase, the uniaxial compressive strength and elastic modulus decreases, and the same freezing and thawing temperature range of the influence of elastic modulus under uniaxial compressive strength is more significant, and Poisson's ratio shows a trend of gradual increase. With the increase of freeze-thaw temperature range, the acoustic emission ringing count showed a gradual increasing trend, the cumulative ringing count gradually steepened, the silence period became smaller, the acoustic emission distribution time range gradually decreased, and the peak frequency showed a trend of low frequency development. The research results provide experimental basis for predicting engineering stability of rock mass in cold area. [ABSTRACT FROM AUTHOR]

Human engineering activities often lead to slope disasters, significantly impacting the safety of construction personnel and project progress. To obtain real-time status information of slopes, based on optoelectronic joint monitoring technology, multi-field neural sensing such as slope deformation, temperature, seepage, and geoelectricity was carried out to grasp the spatiotemporal continuous state of slopes accurately. By citing technologies such as the internet of things, big data, and artificial intelligence, data cleaning, association mining, and clustering analysis were carried out on multi-fields of information to identify control factors that affect slope stability and analyze the superposition effect of each control factor. On this basis, a slope multi-field safety guarantee system that integrates anomaly perception, information transmission, data processing, and decision warning was established. To identify, evaluate, and warn the potential slope risks, then the scientific decisions and implement emergency plans were made, which provided a reference basis for ensuring the safety of slope engineering, disaster prevention and reduction. [ABSTRACT FROM AUTHOR]

Most remote sensing image data is inevitably contaminated by clouds, resulting in data failure. Therefore, detecting clouds is a necessary preprocessing step. With the rapid development of space technology, more portable satellites have been designed to equip these micro satellites with remote sensing image preprocessing models. Designing a high-precision and low computational power requirement lightweight cloud and cloud shadow detection network model is of great significance. In response to the above issues, a lightweight M-shaped network (L-MNet) based on deep separable convolution was proposed. The L-MNet network model was introduced depthwise separable convolution (DS-Conv) on the basis of the M-shaped network (M-Net) model. A depthwise separable convolutional block (DS Conv Block) was designed to reduce the complexity and computational complexity of the algorithm. The results show that the proposed method can effectively reduce the model size and computational complexity of pixel level cloud detection while ensuring detection accuracy, and is helpful in achieving the task of small satellite in orbit cloud detection. [ABSTRACT FROM AUTHOR]

To solve the problem that the mini-van is easy to roll over during high-speed steering, a feedforward feedback-Smith predictive-model predictive control combined with differential braking for anti-rollover control method was proposed. Firstly, a three-de- gree-of-freedom vehicle model was established to obtain the actual values of longitudinal velocity, transverse sway angular velocity and spring-loaded roll angle, and the state deviation was obtained by subtracting them from the theoretical values obtained from Carsim. Then, the MPC (model predictive control) controller was designed to calculate the additional transverse sway moment with the normalized zero moment point lateral offset of the vehicle as the warning indicator, and the braking moment was distributed by the differential braking method to finally realize the vehicle rollover prevention. The vehicle turning motion experiments were conducted through the joint simulation of Carsim and MATLAB / Simulink. The experimental results show that the control method can effectively improve the stability and the safety of the minivan in the high-speed turning process. [ABSTRACT FROM AUTHOR]

Copyright of Science Technology & Engineering is the property of Chinese Society of Technology Economics and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)