RESUMEN
Precise identification and spatial analysis of land salinity in China's Yellow River Delta are essential for the rational utilization and sustainable development of land resources. However, the accurate retrieval model construction for monitoring land salinity remains challenging. This study constructed a land salinity retrieval framework using a harmonized UAV and Landsat-9 multi-spectral dataset. The Kenli district of the Yellow River Delta was selected as the case study area, and a land salinity monitoring index (LSMI) was proposed based on field survey data and UAV multi-spectral image and applied to the reflectance-corrected Landsat-9 OLI image. The land salinity distribution patterns were then mapped and spatially analyzed using Moran's I and Getis-Ord GI* analysis. The results demonstrated the following: (1) The LSMI-based method can accurately retrieve land salinity content with a validation determination coefficient (R2), root mean square error (RMSE), and residual predictive deviation (RPD) of 0.75, 1.89, and 2.11, respectively. (2) Land salinization affected 93.12% of the cultivated land in the study area, and the severely saline soil grade (with a salinity content of 6-8 g/kg) covered 38.41% of the total cultivated land area and was widely distributed throughout the study area. (3) Saline land exhibited a positive spatial autocorrelation with a value of 0.311 at the p = 0.000 level; high-high cluster types occurred mainly in the Kendong and Huanghekou towns (80%), while low-low cluster types were mainly located in the Dongji, Haojia, Kenli, and Shengtuo towns (88.46%). The spatial characteristics of various salinity grades exhibit significant variations, and conducting separate spatial analyses is recommended for future studies.
RESUMEN
In pattern classification, we may have a few labeled data points in the target domain, but a number of labeled samples are available in another related domain (called the source domain). Transfer learning can solve such classification problems via the knowledge transfer from source to target domains. The source and target domains can be represented by heterogeneous features. There may exist uncertainty in domain transformation, and such uncertainty is not good for classification. The effective management of uncertainty is important for improving classification accuracy. So, a new belief-based bidirectional transfer classification (BDTC) method is proposed. In BDTC, the intraclass transformation matrix is estimated at first for mapping the patterns from source to target domains, and this matrix can be learned using the labeled patterns of the same class represented by heterogeneous domains (features). The labeled patterns in the source domain are transferred to the target domain by the corresponding transformation matrix. Then, we learn a classifier using all the labeled patterns in the target domain to classify the objects. In order to take full advantage of the complementary knowledge of different domains, we transfer the query patterns from target to source domains using the K-NN technique and do the classification task in the source domain. Thus, two pieces of classification results can be obtained for each query pattern in the source and target domains, but the classification results may have different reliabilities/weights. A weighted combination rule is developed to combine the two classification results based on the belief functions theory, which is an expert at dealing with uncertain information. We can efficiently reduce the uncertainty of transfer classification via the combination strategy. Experiments on some domain adaptation benchmarks show that our method can effectively improve classification accuracy compared with other related methods.
Asunto(s)
Aprendizaje , Aprendizaje AutomáticoRESUMEN
Fluorescence molecular tomography (FMT) is an imaging technique that can localize and quantify fluorescent markers to resolve biological processes at molecular and cellular levels. Owing to a limited number of measurements and a large number of unknowns as well as the diffusive transport of photons in biological tissues, the inverse problem in FMT is usually highly ill-posed. In this work, a sparsity-constrained preconditioned Kaczmarz (SCP-Kaczmarz) method is proposed to reconstruct the fluorescent target for FMT. The SCP-Kaczmarz method uses the preconditioning strategy to minimize the correlation between the rows of the forward matrix and constrains the Kaczmarz iteration results to be sparse. Numerical simulation and phantom and in vivo experiments were performed to test the efficiency of the proposed method. The results demonstrate that both the convergence and accuracy of the proposed method are improved compared with the classical memory-efficient low-cost Kaczmarz method.