Explore the vast range of titles available.

Modeling forest fire spread is a very complex problem, and the existing models usually need some input parameters which are hard to get. How to predict the time series of forest fire spread rate based on passed series may be a key problem to break through the current technical bottleneck. In the process of forest fire spreading, spread rate and wind speed would affect each other. In this paper, three kinds of network models based on Long Short-Term Memory (LSTM) are designed to predict fire spread rate, exploring the interaction between fire and wind. In order to train these LSTM-based models and validate their effectiveness of prediction, several outdoor combustion experiments are designed and carried out. Process data sets of forest fire spreading are collected with an infrared camera mounted on a UAV, and wind data sets are recorded using a anemometer simultaneously. According to the close relationship between wind and fire, three progressive LSTM based models are constructed, which are called CSG-LSTM, MDG-LSTM and FNU-LSTM, respectively. A Cross-Entropy Loss equation is employed to measure the model training quality, and then prediction accuracy is computed and analyzed by comparing with the true fire spread rate and wind speed. According to the performance of training and prediction stage, FNU-LSTM is determined as the best model for the general case. The advantage of FNU-LSTM is further demonstrated by doing comparison experiments with the normal LSTM and other LSTM based models which predict both fire spread rate and wind speed separately. The experiment has also demonstrated the ability of the model to the real fire prediction on the basis of two historical wildland fires.

This paper is focused on designing a cost function of selecting a foothold for a physical quadruped robot walking on rough terrain. The quadruped robot is modeled with Denavit–Hartenberg (DH) parameters, and then a default foothold is defined based on the model. Time of Flight (TOF) camera is used to perceive terrain information and construct a 2.5D elevation map, on which the terrain features are detected. The cost function is defined as the weighted sum of several elements including terrain features and some features on the relative pose between the default foothold and other candidates. It is nearly impossible to hand-code the weight vector of the function, so the weights are learned using Supporting Vector Machine (SVM) techniques, and the training data set is generated from the 2.5D elevation map of a real terrain under the guidance of experts. Four candidate footholds around the default foothold are randomly sampled, and the expert gives the order of such four candidates by rotating and scaling the view for seeing clearly. Lastly, the learned cost function is used to select a suitable foothold and drive the quadruped robot to walk autonomously across the rough terrain with wooden steps. Comparing to the approach with the original standard static gait, the proposed cost function shows better performance.

Quadruped robot has great potential to walk on rough terrain, in which static gait is preferred. A hierarchical structure based controlling algorithm is proposed in this paper, in which trajectory of robot center is searched, and then static gaits are generated along such trajectory. Firstly, cost map is constructed by computing terrain features under robot body and cost of selecting footholds at default positions, and then the trajectory of robot center in 2D space is searched using heuristic A⁎ algorithm. Secondly, robot state is defined from foothold and robot pose, and then state series are searched recursively along the trajectory of robot center to generate static gaits, where a tree-like structure is used to store such states. Lastly, a classical model for quadruped robot is designed, and then the controlling algorithm proposed in the paper is demonstrated on such robot model for both structured terrain and complex unstructured terrain in a simulation environment.

In the past few decades, a large number of time series classification (TSC) algorithms have been published based on different class pattern hypotheses, among which a vital component is the feature extraction of time series. Currently, the TSC algorithm with the highest classification accuracy is the heterogeneous ensemble, which significantly improves the classification accuracy but greatly increases the algorithm complexity. Therefore, a novel feature extraction algorithm called pipeline transform is proposed to achieve feature diversity enhancement by three fast time series transformations and feature vector concatenation. The time series is transformed to Hilbert, first-order differential, and second-order differential spaces to generate corresponding transformed series. A random convolutional kernel transform is performed on each series to generate the corresponding feature vector. The raw feature vector is concatenated with the three transformed feature vectors respectively to obtain three feature vectors, namely the three-view. In the proposed TSC algorithm FT-FVC, pipeline transform and hard voting are combined, which has excellent classification accuracy and speed. Furthermore, the semi-supervised TSC algorithm, semi-FT-FVC, combines pipeline transform with Tri-Training, improving classification accuracy and reducing classification volatility. The proposed algorithms are compared with other state-of-the-art algorithms on 85 datasets in the UCR archive, whose excellent classification performance is demonstrated by statistical tests.

ObjectivesMicrocalcifications are an important feature in the diagnosis of breast cancer, especially in the early stages. In this paper, a CT-based method is proposed to potentially distinguish benign and malignant breast diseases based on the distributions of microcalcifications using grating-based phase-contrast imaging on a conventional X-ray tube.MethodsThe method presented based on the ratio of dark-field signals to attenuation signals in CT images is compared with the existing method based on the ratio in projections, and the threshold for the classification of microcalcifications in the two types of breast diseases is obtained using our approach. The experiment was operated on paraffin-fixed specimens that originated from 20 female patients ranging from 27–65 years old.ResultsCompared with the method based on projection images (AUC = 0.87), the proposed method is more effective (AUC = 0.95) to distinguish the two types of diseases. The discrimination threshold of microcalcifications for the classification of diseases in CT images is found to be 3.78 based on the Youden index.ConclusionsThe proposed method can be further developed to improve the early diagnosis and diagnostic accuracy and reduce the clinical misdiagnosis rate of breast cancer.Key Points• Microcalcifications are of special importance to indicate early breast cancer.• Grating-based phase-contrast imaging can improve the diagnosis of breast cancers.• The method described here can better classify benign and malignant breast diseases.

The spread of forest fire is an extremely complex and harmful natural phenomenon. At present, the forest fire spread model has some shortcomings, such as complex formula, inaccurate simulation value and so on. In this paper, the Ensemble Kalman Filter(ENKF) algorithm is applied to the field of forest fire spread so that it can better predict the spread of forest fire. Firstly, the Rothermel forest fire speed formula is simplified, and the simplified Rothermel speed value is modified by the actual measured forest fire spread speed value, so that the optimal model simulation value is obtained. Then the optimal speed is input into Cellular Automata(CA) to simulate the spread of forest fire. Secondly, the experiment is carried out by changing the slope, bed thickness, moisture content, load and wind speed. And the actual measured speed value, the simplified Rothermel model value and the optimized value after ENKF are compared in the process of fire spread. Finally, The experimental results show that the error of fire spread speed corrected by ENKF is smaller, the forest fire spread contour obtained from the optimal speed value by ENKF is closer to the actual fire spread contour, and the highest similarity index is 0.854. The model proposed in this paper has the ability to predict the spread of forest fire indoors.

Rothermel model is a common method for predicting forest fire spread rate, but Its application is limited, due to complexity of the formula and too many parameters. In this paper, the Rothermel model is optimized to a simple format, which contains 4 independent variables as input, 1 dependent variable as output and 8 parameters to be estimated. In order to validate the effectiveness of the optimized model, the indoor ignition experiment is designed and carried out, and then the fire spreading data is collected and processed in advance for training the parameters of the model. By analyzing the effectiveness of 3 nonlinear optimizing methods , the Levenberg-Marquardt(LM) method is chosen to estimated the parameters of the model. At last, by comparing to the actual measured value, precision of the optimized model is validated on the verification data, and with the ability to predict the speed of fire spreading in the indoor laboratory.

Forecasting of forest fire area is of great significance to effectively control the spread of forest fire. In this paper, the forest fire spreading velocity model and the forest fire spreading simulation technology based on huygens principle are used to estimate the forest fire area. Firstly, binocular camera is used to collect the firing state data of wild forest fire, and segment the firing image, extract the firing line, locate the firing line and calculate the three-dimensional coordinates of the firing line pixels according to perspective projection model;. Secondly, the forest fire spreading velocity model based on Wang Zhengfei’s model is redesigned. The model parameters of forest fire area were optimized by gradient method. The prediction accuracy is much higher than that of the model before optimization.

With advanced X-ray source and detector technologies being continuously developed, non-traditional CT geometries have been widely explored. Generalized-Equiangular Geometry CT (GEGCT) architecture, in which an X-ray source might be positioned radially far away from the focus of arced detector array that is equiangularly spaced, is of importance in many novel CT systems and designs. GEGCT, unfortunately, has no theoretically exact and shift-invariant analytical image reconstruction algorithm in general. In this study, to obtain fast and accurate reconstruction from GEGCT and to promote its system design and optimization, an in-depth investigation on a group of approximate Filtered BackProjection (FBP) algorithms with a variety of weighting strategies has been conducted. The architecture of GEGCT is first presented and characterized by using a normalized-radial-offset distance (NROD). Next, shift-invariant weighted FBP-type algorithms are derived in a unified framework, with pre-filtering, filtering, and post-filtering weights. Three viable weighting strategies are then presented including a classic one developed by Besson in the literature and two new ones generated from a curvature fitting and from an empirical formula, where all of the three weights can be expressed as certain functions of NROD. After that, an analysis of reconstruction accuracy is conducted with a wide range of NROD. We further stretch the weighted FBP-type algorithms to GEGCT with dynamic NROD. Finally, the weighted FBP algorithm for GEGCT is extended to a three-dimensional form in the case of cone-beam scan with a cylindrical detector array.

Purpose: With the development of spectral CT, several novel spectral filters have been introduced to modulate the spectra, such as split filters and spectral modulators. However, due to the finite size of the focal spot of X-ray source, these filters cause spectral mixing in the penumbra region. Traditional spectrum estimation methods fail to account for it, resulting in reduced spectral accuracy. Methods: To address this challenge, we develop a multi-ray spectrum estimation model and propose an Adaptive Subsampled WeIghting of Filter Thickness (A-SWIFT) method. First, we estimate the unfiltered spectrum using traditional methods. Next, we model the final spectra as a weighted summation of spectra attenuated by multiple filters. The weights and equivalent lengths are obtained by X-ray transmission measurements taken with altered spectra using different kVp or flat filters. Finally, the spectra are approximated by using the multi-ray model. To mimic the penumbra effect, we used a spectral modulator (0.2 mm Mo, 0.6 mm Mo) and a split filter (0.07 mm Au, 0.7 mm Sn) in simulations, and used a copper modulator and a molybdenum modulator (0.2 mm, 0.6 mm) in experiments. Results: Simulation results show that the mean energy bias in the penumbra region decreased from 7.43 keV using the previous SCFM method (Spectral Compensation for Modulator) to 0.72 keV using the A-SWIFT method for the split filter, and from 1.98 keV to 0.61 keV for the spectral modulator. In experiments, the root mean square error of the selected ROIs was decreased from 77 to 7 Hounsfield units (HU) for the pure water phantom with a molybdenum modulator, and from 85 to 21 HU with a copper modulator. Conclusion: Based on a multi-ray spectrum estimation model, the A-SWIFT method provides an accurate and robust approach for spectrum estimation in penumbra region of CT systems utilizing spectral filters.