Recurrent neural networks (RNNs) have been widely adopted in research areas concerned with sequential data, such as text, audio, and video. However, RNNs consisting of sigma cells or tanh cells are ...unable to learn the relevant information of input data when the input gap is large. By introducing gate functions into the cell structure, the long short-term memory (LSTM) could handle the problem of long-term dependencies well. Since its introduction, almost all the exciting results based on RNNs have been achieved by the LSTM. The LSTM has become the focus of deep learning. We review the LSTM cell and its variants to explore the learning capacity of the LSTM cell. Furthermore, the LSTM networks are divided into two broad categories: LSTM-dominated networks and integrated LSTM networks. In addition, their various applications are discussed. Finally, future research directions are presented for LSTM networks.
Full text
Available for:
DOBA, IZUM, KILJ, NUK, PILJ, PNG, SAZU, SIK, UILJ, UKNU, UL, UM, UPUK
Abstract
Moisture migration in unsaturated soils is a result of the interaction between temperature and soil microstructure. In order to reveal the mechanism of moisture increase of subgrade soils ...under diurnal cycle conditions, a series of macro and microscopic tests were carried out on the unsaturated silty clay and sand soil, including liquid–vapor mixed migration tests simulating a one-dimensional subgrade, environmental scanning electron microscope (ESEM), and matrix suction test. Then, the soil microstructure in microscopic images was investigated using the particle (pores) and cracks analysis system (PACS). Next, the relationship between the thermal effects of the soil–water characteristic curve (SWCC) and changes in soil microstructure was analyzed. Finally, the change mechanism of liquid–vapor mixed migration based on the change in soil microstructural under thermal effects was analyzed. The results showed under the diurnal cycle, both the silty clay and sand soil columns appeared in the phenomenon of a “diurnal cycle of water vapor migration”, which led to moisture accumulation at the top of the soil layer. In silty clay soil column, moisture was primarily driven by water vapor pressure and migrated upwards. Additionally, moisture redistribution led to changes in soil microstructure, which in turn influenced the process of moisture migration. The moisture content in the upper soil layer increased making both inter-aggregate and intra-aggregate pores decrease. The moisture content in the lower soil later decreased, leading to the water-holding capacity of the lower soil layer to increase. So, the moisture migration gradually decreased at night. In the sand soil column, moisture migration was mainly driven by gravity potential and migrated downwards. Moisture redistribution made inter-aggregate pore and matrix suction of the upper soil layer increase, leading to an increase in moisture migration at night.
Full text
Available for:
IZUM, KILJ, NUK, PILJ, PNG, SAZU, UL, UM, UPUK
Target detection in satellite images is an essential topic in the field of remote sensing and computer vision. Despite extensive research efforts, accurate and efficient target detection in remote ...sensing images remains unsolved due to the large target scale span, dense distribution, and overhead imaging and complex backgrounds, which result in high target feature similarity and serious occlusion. In order to address the above issues in a comprehensive manner, within this paper, we first propose a Centralised Visual Processing Center (CVPC), this structure is a parallel visual processing center for Transformer encoder and CNN, employing a lightweight encoder to capture broad, long-range interdependencies. Pixel-level Learning Center (PLC) module is used to establish pixel-level correlations and improve the depiction of detailed features. CVPC effectively improves the detection efficiency of remote sensing targets with high feature similarity and severe occlusion. Secondly, we propose a centralised feature cross-layer fusion pyramid structure to fuse the results with the CVPC in a top-down manner to enhance the detailed feature representation capability at each layer. Ultimately, we present a Context Enhanced Adaptive Sparse Convolutional Network (CEASC), which improves the accuracy while ensuring the detection efficiency. Based on the above modules, we designed and conducted a series of experiments. These experiments are conducted on three challenging public datasets, DOTA-v1.0, DIOR, and RSDO, showing that our proposed 3CNet achieves a more advanced detection accuracy while balancing the detection speed (78.62% mAP for DOTA-v1.0, 79.12% mAP for DIOR, and 95.50% mAP for RSOD).
Full text
Available for:
IZUM, KILJ, NUK, PILJ, PNG, SAZU, UL, UM, UPUK
Effective mechanical properties of an orthogonal corrugated sandwich structure have been investigated analytically and numerically. A necessary condition for geometrical design of orthogonal ...corrugated sandwich structure is obtained. The elastic properties and the collapse surfaces are identified and explored. Finite element simulations are carried out to validate the analytical model. It is shown that the analytical predations are in good agreement with numerical results. Moreover, the orthogonal corrugated sandwich structure has better shear performance than the conventional corrugated sandwich structure by appropriate design.
Display omitted
•Effective mechanical properties of an orthogonal corrugated sandwich structure have been obtained.•The collapse surfaces of an orthogonal corrugated sandwich structure are developed.•A necessary condition for geometrical design of orthogonal corrugated sandwich structure is proposed.•The orthogonal corrugated core has better shear properties than the conventional corrugated core by appropriate design.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UILJ, UL, UM, UPCLJ, UPUK, ZAGLJ, ZRSKP
Milling is one of the most popular forms of hard tissue removal process, it is sometimes high-risk because there may be very small distance between the milling trajectory and vital anatomy. This ...paper introduces a vibration signal acquiring and processing method to identify different types of milling states in robot-assisted orthopedics. During milling process, the tissue vibration signal measured by a laser displacement sensor, and the acceleration signal of the operation power device recorded by an accelerometer, are decomposed by the lifting wavelet packet transform, so the harmonic component whose frequency is an integer times of the spindle frequency of the operation power device is extracted. In consideration of that the tissue vibration amplitude varies with the milling force, the relative magnitude of the force is estimated from the acceleration signal, and then the wavelet energy of the tissue vibration signal is divided by that of the acceleration signal to reduce the influence of the force disturbance. The compensated wavelet energy is subsequently used as input vector to a support vector machine for discriminating different milling states. Experimental results on porcine spines demonstrate that the proposed method can successfully discriminate the tissues in the operation field: the success rate achieves 100% for the vertebra being milled and the spinal cord, for the adjacent bony structure and the muscle the success rate is about 90%.
Fiber metal laminated sandwich tubes are made up of alternating fiber-reinforced composite and metal layers. Fiber metal laminated tubes have the advantages of the high strength and high stiffness of ...fiber and the toughness of metal, so they have become an excellent load-bearing and energy-absorbing, lightweight structure. Due to the complexity of the fiber layup, it is difficult to establish an analytical model of the relevant structural properties. In this work, introducing the number and volume fraction of fiber layup, based on the modified rigid–plastic model, an analytical model is established for low-velocity impacts on sandwich tubes with fiber metal laminated tubes, which provided a theoretical basis for the design of fiber–metal composite tubes. In addition, a numerical simulation was conducted for low-velocity impacts on clamped rectangular sandwich tubes with fiber metal laminated (FML) tubes and a foam core. By comparing the results obtained from the theoretical analysis and numerical calculations, it is shown that the analytical results can reasonably agree with the numerical results. The influences of the metal volume fraction (MVF), the strength ratio factor of the FML metal layer to the FML composite layer, and the relative strength of the foam on the dynamic response of the rectangular sandwich tubes with FML tubes and a metal foam core (MFC) are discussed. It is shown that by increasing the fiber content and fiber strength of the FML tubes and the foam strength, the load-carrying and energy-absorbing capacity of the rectangular sandwich tubes can be effectively improved, especially by changing the fiber properties. In addition, present analytical solutions can be applied to make predictions about the dynamic response of the rectangular sandwich tubes with FML tubes and MFC during impacts with low-velocity and reasonably heavy-mass.
Full text
Available for:
IZUM, KILJ, NUK, PILJ, PNG, SAZU, UL, UM, UPUK
Through theoretical analysis and finite element simulation, the low-velocity impact of rectangular foam-filled fiber metal laminate (FML) tubes is studied in this paper. According to the ...rigid-plastic material approximation with modifications, simple analytical solutions are obtained for the dynamic response of rectangular foam-filled FML tubes. The numerical calculations for low-velocity impact of rectangular foam-filled FML tubes are conducted. The accuracy of analytical solutions and numerical results is verified by each other. Finally, the effects of the metal volume fraction of FMLs, the number of the metal layers in FMLs, and the foam strength on the dynamic response of foam-filled tubes are discussed through the analytical model in details. It is shown that the force increases with the increase in the metal volume fraction in FMLs, the number of the metal layers in FML, and the foam strength for the given deflection.
In the pork fat content detection task, traditional physical or chemical methods are strongly destructive, have substantial technical requirements and cannot achieve nondestructive detection without ...slaughtering. To solve these problems, we propose a novel, convenient and economical method for detecting the fat content of pig B-ultrasound images based on hybrid attention and multiscale fusion learning, which extracts and fuses shallow detail information and deep semantic information at multiple scales. First, a deep learning network is constructed to learn the salient features of fat images through a hybrid attention mechanism. Then, the information describing pork fat is extracted at multiple scales, and the detailed information expressed in the shallow layer and the semantic information expressed in the deep layer are fused later. Finally, a deep convolution network is used to predict the fat content compared with the real label. The experimental results show that the determination coefficient is greater than 0.95 on the 130 groups of pork B-ultrasound image data sets, which is 2.90, 6.10 and 5.13 percentage points higher than that of VGGNet, ResNet and DenseNet, respectively. It indicats that the model could effectively identify the B-ultrasound image of pigs and predict the fat content with high accuracy.
A modified Brazilian disk tension test Yu, Yong; Zhang, Jianxun; Zhang, Jichun
International journal of rock mechanics and mining sciences (Oxford, England : 1997),
02/2009, Volume:
46, Issue:
2
Journal Article
Peer reviewed
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UL, UM, UPCLJ, UPUK
•A dynamic analytical model is developed to predict the dynamic response of metal foam core sandwich plates struck by heavy mass with low-velocity.•Large deflection effect is included in analysis by ...considering the interaction between the plastic bending and stretching.•The dynamic response and energy absorption of the sandwich plates depend on the impact location of the striker.
An analytical model is developed to predict the dynamic response of fully clamped square sandwich plate with metal foam core struck transversely by a heavy mass with low-velocity. Large deflection effect is incorporated in analysis by considering the interaction between the plastic bending and stretching. The analytical expressions are obtained for the structural deflection, the structural response time and the impact force. The finite element results validate the accuracy of the analytical model and good agreement is achieved.
Full text
Available for:
GEOZS, IJS, IMTLJ, KILJ, KISLJ, NLZOH, NUK, OILJ, PNG, SAZU, SBCE, SBJE, UL, UM, UPCLJ, UPUK, ZRSKP