Heavy metal concentration is an important index for evaluating soil pollution. It is of great significance to measure the trace element content accurately for green agriculture development. In order to detect the trace element content accurately, a new prediction framework including pre-processing, signal extraction, feature selection and decision-making was proposed. The energy dispersive X-ray fluorescence (ED-XRF) spectra of 57 national standard soil samples were investigated based on the proposed methods. Firstly, an innovative background deduction method called iterative adaptive window empirical wavelet transform (IAWEWT) was introduced to extract effective counts of characteristic peaks, and the proposed approach was validated by the coefficient of determination (R²) of the instrumental calibration curve compared with two other conventional methods. Secondly, principal component analysis (PCA) was combined with the analysis of variance (ANOVA) for variable selection optimization of the ED-XRF spectrum. After PCA feature extraction and ANOVA variable selection treatment, the optimum number of principal components for V, Cr, Cu, Zn, Mo, Cd and Pb were determined to be 7, 15, 4, 4, 4, 5 and 12 respectively. Furthermore, the support vector regression (SVR) model was adopted for heavy metal estimation. The evaluation indices included R² and root mean square error (RMSE). It was demonstrated that the predictive capabilities of seven heavy metal elements were improved substantially for elemental analysis by the proposed PCA–ANOVA–SVR model, with excellent results for V, Cr, Cu, Zn, Mo, Cd and Pb estimates, and the R² values were 0.993, 0.996, 0.999, 0.999, 0.997, 0.998 and 0.998 respectively. Therefore, the new framework proposed in this paper can effectively eliminate redundant features and determine the concentration of trace elements in soil. It provides an effective alternative for the quantitative analysis of X-ray fluorescence spectrometry.

Soil is an important source of potentially toxic metal intake for humans through the food chain. Accurate determination of elemental concentrations in soil is of great significance to safeguard human health, thereby requiring reliable detection techniques. This work aimed to propose a new quantitative analysis method that combines pre-processing and concentration prediction. A total of 59 soil samples were collected and subjected to laboratory analyses and energy dispersive X-ray fluorescence (ED-XRF) scans. Firstly, a novel pre-processing method based on iterative adaptive window empirical wavelet transform and Gaussian convolution (IAWEWT-GC) was proposed, which can identify and remove spectral background. Secondly, the instrument calibration curves of peak counts and elemental concentrations were established and the pre-treatment effect was verified by comparison with other common methods. The instrument calibration curves' coefficient of determination (R²) was enhanced to 0.988, 0.962, and 0.980 for Cr, Mn, and Cu, respectively. Finally, to accurately detect the elemental concentration, a hierarchical deep neural network (HDNN) was designed for estimating potentially toxic metals, enabling quantitative regression of multiple elements simultaneously. The proposed method was evaluated and compared with conventional algorithms (DNN, BPNN, and PLSR). The HDNN model achieved the highest R² of 0.943, 0.983, and 0.979 for Cr, Mn, and Cu, respectively, with the lowest RMSEP of 5.74, 84.45, and 29.17, and lowest MAE of 4.51, 78.51, and 16.35, indicating its ability to enable simple, rapid, and effective quantitative analysis of multiple elements. Overall, our proposed method provides a new option for the elemental analysis of samples rich in potentially toxic metals.

In this paper, we introduce a composable and extensible drone with tilting rotors (CEDTR). We aimed for a function that could optimally match the load capacity, degree of freedom (DOF), speed and endurance with diverse mission requirements by changing the quantity and form of combinations. First, we propose a decentralized modular controller to allow a team of physically connected modules to fly cooperatively. Second, we divide all the combinations into three categories according to the different control methods. Three generalized control strategies are proposed to control both position and attitude independently by tilting the directions of the propellers. We carried out experiments to demonstrate the feasibility of this mechanical design and control method. The experiment video is available at https://youtu.be/7RvxiV4FPq4.

Obtaining accurate characteristic spectra and the net peak area is crucial in X-ray fluorescence (XRF) quantitative analysis. To improve the calculation accuracy of the net peak area of the characteristic X-ray spectrum, carrying out the baseline calibration is necessary before resolving the spectrum. First, this article proposes the use of an enhanced generative adversarial network (EGAN) depth network model for baseline calibration of XRF spectra. This method directly takes a clean spectrum after deducting the background as the optimal target for EGAN deep network training, thus eliminating the traditional cumbersome baseline fitting process. It can directly and quickly obtain the XRF spectrum after baseline calibration, and the entire process requires neither fitting the background nor knowing the background data. Further, to improve the generalization ability of the EGAN model, we introduce transfer learning in the model training process. We use the existing alloy sample spectrogram data to pretrain the EGAN model generator and then migrate the pretrained XRF-EGAN generator model to the baseline calibration task of 57 national standard soil sample spectra collected. During the model training process, a cross-validation method is used to train and test the model’s effectiveness. Finally, experiments are conducted on simulated spectra and actual XRF spectra to verify the accuracy and adaptability of the proposed method. Instrument calibration curves for peak counting and element concentration are established to verify the effectiveness of baseline calibration. The coefficient of determination R² of the calibration curves for elements Cu, Zn, Mn and Cr is increased to 0.998, 0.988, 0.922 and 0.991, respectively. The results indicate that our proposed method can effectively estimate the pure spectrum after deducting the background. In addition, this method can also be applied to the baseline correction of other similar spectral signals.

The salp swarm algorithm imitates the swarm behavior of salps during navigation and hunting that has been proven the superiority of search for best solution. However, although it has sufficient global search ability, it is still worth paying attention to problems of falling into local optima and lower convergence accuracy. This paper proposes some improvements to the salp swarm algorithm that are based on a nonlinear dynamic weight and the mapping mutation operation. Firstly, the nonlinear dynamic weight is helpful for further optimizing the transition from exploration to exploitation and alleviating the local optima stagnation phenomena. Secondly, utilizing a mapping mutation operation can increase the diversity of followers in algorithm, to avoid getting trapped into the local optima during the search and provide a better optimal solution. The proposed algorithm is characterized by a stronger global optimization capability and high convergence accuracy. Finally, to confirm the effectiveness of the proposed algorithm, comparative experiments based on other well-known swarm-based algorithms and each improvement for the original algorithm are conducted. The quantitative results and convergence curves among several algorithms demonstrate that the enhanced algorithm with the nonlinear dynamic weight and mapping mutation operation can outperform the original algorithm.

Siamese network trackers based on pre-trained depth features have achieved good performance in recent years. However, the pre-trained depth features are trained in advance on large-scale datasets, which contain feature information of a large number of objects. There may be a pair of interference and redundant information for a single tracking target. To learn a more accurate target feature information, this paper proposes a lightweight target-aware attention learning network to learn the most effective channel features of the target online. The lightweight network uses a designed attention learning loss function to learn a series of channel features with weights online with no complex parameters. Compared with the pre-trained features, the channel features with weights can represent the target more accurately. Finally, the lightweight target-aware attention learning network is unified into a Siamese tracking network framework to implement target tracking effectively. Experiments on several datasets demonstrate that the tracker proposed in this paper has good performance.