Non-contact heart rate measurement methods based on remote PhotoPlethysmoGraphy （rPPG） technology face two main challenges： the accurate selection of facial Regions of Interest （RoI）， and the precise interpretation of periodic patterns in rPPG from long video sequences. To address the inaccurate localization of facial RoI in the existing works and the neglect of temporal lag among different signal sources， an end-to-end rPPG signal extraction model， named RppgMamba， was proposed for non-contact heart rate measurement. In the model， the hierarchical spatial-temporal Mamba module with cross-scan mechanism was designed to capture spatial perception relationships of weak signals from multi-scale spatial-temporal receptive fields， so as to localize facial RoI adaptively， thereby avoiding interference from background and motion noise. And the frequency-domain refinement module was designed to further learn the quasi-periodic patterns of rPPG signals in the frequency domain， which enhanced output signal quality and corrected phase shifts of the output signals combining with phase compensator， thereby reducing the impact of inherent temporal lag. Experimental results demonstrate that the proposed model achieves the best performance on three public datasets， confirming its high accuracy and applicability in heart rate measurement.
Estimating ambiguous Three-Dimensional （3D） human poses from monocular videos accurately is a current research challenge. Though the existing methods can estimate 3D joint coordinates using deep learning models， most of them fail to consider the multi-solution nature of this inverse problem adequately. Some multi-hypothesis estimation methods address multi-solution problems， but they suffer from insufficient cross-level feature fusion. To address these issues， a 3D human pose estimation model based on Temporal-SPatial Feature Pyramid Network （TSP-FPN） and multi-hypothesis interaction mechanism， called TSP-FPN-MHFormer （Temporal-SPatial Feature Pyramid Network-Multi-Hypothesis Transformer）， was proposed. Firstly， based on Transformer encoder， the multi-possibility distribution of human poses was captured by using multi-head self-attention mechanism， thereby generating multiple initial hypothesis features. Then， a TSP-FPN was designed， and a gated adaptive fusion strategy was employed to achieve dynamic weighted integration of multi-level skeleton sequence features， thereby balancing the fusion of local details and global temporal information effectively. Finally， based on Multi-Hypothesis Transformer （MHFormer）， a multi-hypothesis optimization module that combined joint Relative Position Bias （RPB） with a cross-attention mechanism was implemented， thereby facilitating cross-hypothesis communication and feature aggregation to enhance the model’s long-range reasoning capability to human topology for high-precision 3D joint coordinate estimation. Experimental results on the Human3.6M dataset demonstrate that the proposed model achieves a Mean Per Joint Position Error （MPJPE） of 42.3 mm， and reduces the estimation error by 1.6% compared to the state-of-the-art method MHFormer， indicating substantial progress obtained by the proposed model in addressing multi-solution challenge of monocular 3D pose estimation.
In recent years， anchor-based 3D hand pose estimation methods are becoming popular， and Anchor-to-Joint （A2J） is one of the more representative methods. In A2J， anchor points are densely set on depth map， and neural network is used to predict offsets between anchor points and key points together with weights of anchor points； predicted offsets and weights are used to calculate the coordinates of key points in a weighted summation mode to reduce noise in network regression results. A2J methods are simple and effective， but they are sensitive to ill-suited network structure and prone to inaccurate regression due to loss function. Therefore， an improved network HigherA2J was proposed. Firstly， a single branch jointly predicted X， Y and Z offsets between anchors and key points to better utilize 3D characteristics of depth map； secondly， network branch structure was simplified to reduce network parameters； finally， the loss function for key point estimation was designed， combined with offset estimation loss， which improved the overall estimation accuracy effectively. Experimental results show the reductions in average hand pose estimation error of 0.32 mm， 0.35 mm and 0.10 mm compared to conventional A2J on three datasets NYU， ICVL and HANDS 2017 respectively.
Compressed sensing mainly contains random projection and reconstruction. Because of lower convergence speed of iterative shrinkage algorithm and the lacking of direction of traditional 2-dimensional wavelet transform， random projection was implemented by using Permute Discrete Cosine Transform （PDCT）， and the gradient projection was used for reconstruction. Based on the simplification of computation complexity， the transformation coefficients in the dual-tree complex wavelet domain were improved by iteration. Finally， the reconstructed image was obtained by the inverse transform. In the experiments， the reconstruction results of DT CWT （Dual-Tree Complex Wavelet Transform） and bi-orthogonal wavelet were compared with the same reconstruction algorithm， and the former is better than the latter in image detail and smoothness with higher Peak Signal-to-Noise Ratio （PSNR） of 1.5 dB. In the same sparse domain， gradient projection converges faster than iterative shrinkage algorithm. And in the same sparse domain and random projection， PDCT has a slightly higher PSNR than the structural random matrix.
Focusing on the underdeveloped robustness when the existing extended rough set model encounters the noise for the incomplete information system, the necessity of adjusting the size of basic knowledge granule as well as introducing the relative degree of misclassification was analyzed. Then the Variable Precision Rough Set model based on Variable-Precision Tolerance Relation (VPRS-VPTR) was established on the basis of the object connection weight matrix, which was proposed according to the lack probability of system attribute value. Moreover, the properties of the VPRS-VPTR model were discussed, the classification accuracy under the basic knowledge granule size and the relative degree of misclassification was analyzed, the corresponding algorithm was depicted and the time complexity analysis was given afterwards. The experimental results show that the VPRS-VPTR model has higher classification accuracy compared with some other research about the expanded rough set, and the change trend of the classification accuracy is similar for the train set and the test set of several groups of incomplete data sets in UCI database. It proves that the proposed model is more precise and flexible, and the algorithm is feasible and effective.
Similarity measurement of cloud model is a method that is used to measure the correlation between cloud models, which have same concept but different languages. Both similar cloud and its measurement analysis method are the extension of cloud model theory. To overcome the disadvantages of high consumption and low precision of calculation, a similarity measure algorithm based on overlap degree was proposed. Firstly, the position and logical relationships between these two clouds were defined according to three digital features: expected value, entropy and hyper entropy; secondly, the overlap degree of two clouds was calculated by using their location and shape features; finally, combined with overlap degree and similarity, the similarity measurement was converted to quantitative description of the overlapping part. In the time series classification experiments with compared Likeness comparing method based on Cloud Model (LICM), the computing consumption of the proposed measurement algorithm is reduced by 50% on the premise of ensuring the stability and accuracy. It is proved to be feasible and effective by the application.
According to the problem of the traditional Gravitational Search Algorithm (GSA) such as falling into the local minimum point easily, a hybrid algorithm based on Estimation of Distribution (ED) and gravitational search (GSEDA) was proposed. By characterizing the distribution of current solutions found by GSA, ED was used to generate promising solutions based on the constructed probability matrix, thus guiding the search to new solution areas. The proposed GSEDA was able to balance the exploration and exploitation of the search, therefore possessing a better local optima jumping capacity. The experimental results based on the traveling salesman problem indicate that GSEDA performs better than traditional algorithms in terms of solution quality and robustness.
It is difficult for users to find the needed items from a large-scale project resource repository because the project resources in it are disordered, so a parallel fuzzy partition algorithm based on MapReduce was proposed. The algorithm firstly abstracted and standardized characteristic attributes of original project resource. Then a similarity matrix was established based on the standardized characteristic attributes of the project, and it was segmented by using block matrix. MapReduce was used to process the block matrix and merge the results. Finally, the algorithm obtained the partition results according to the threshold. The contrast experiment among the proposed algorithm, K-means algorithm and genetic algorithm shows that the proposed algorithm has higher accuracy and recall, it can achieve better speedup in large-scale data calculation and divide project resources effectively and accurately.
