When extracting land-use information from remote sensing imagery using image segmentation, obtaining fine edges for extracted objects is a key problem that is yet to be solved. In this study, we developed a new weight feature value convolutional neural network (WFCNN) to perform fine remote sensing image segmentation and extract improved land-use information from remote sensing imagery. The WFCNN includes one encoder and one classifier. The encoder obtains a set of spectral features and five levels of semantic features. It uses the linear fusion method to hierarchically fuse the semantic features, employs an adjustment layer to optimize every level of fused features to ensure the stability of the pixel features, and combines the fused semantic and spectral features to form a feature graph. The classifier then uses a Softmax model to perform pixel-by-pixel classification. The WFCNN was trained using a stochastic gradient descent algorithm; the former and two variants were subject to experimental testing based on Gaofen 6 images and aerial images that compared them with the commonly used SegNet, U-NET, and RefineNet models. The accuracy, precision, recall, and F1-Score of the WFCNN were higher than those of the other models, indicating certain advantages in pixel-by-pixel segmentation. The results clearly show that the WFCNN can improve the accuracy and automation level of large-scale land-use mapping and the extraction of other information using remote sensing imagery.

A new algorithm for solving large scale bound constrained minimization problems is proposed. The algorithm is based on an accurate identification technique of the active set proposed by Facchinei, Fischer and Kanzow in 1998. A further division of the active set yields the global convergence of the new algorithm. In particular, the convergence rate is superlinear without requiring the strict complementarity assumption. Numerical tests demonstrate the efficiency and performance of the present strategy and its comparison with some existing active set strategies.

We employ the active set strategy which was proposed by Facchinei for solving large scale bound constrained optimization problems. As the special structure of the bound constrained problem, a simple rule is used for updating the multipliers. Numerical results show that the active set identification strategy is practical and efficient.

Hyperspectral unmixing is a powerful method of the remote sensing image mining that identifies the constituent materials and estimates the corresponding fractions from the mixture. We consider the application of nonnegative matrix factorization (NMF) for the mining and analysis of spectral data. In this paper, we develop two effective active set type NMF algorithms for hyperspectral unmixing. Because the factor matrices used in unmixing have sparse features, the active set strategy helps reduce the computational cost. These active set type algorithms for NMF is based on an alternating nonnegative constrained least squares (ANLS) and achieve a quadratic convergence rate under the reasonable assumptions. Finally, numerical tests demonstrate that these algorithms work well and that the function values decrease faster than those obtained with other algorithms.

Nonnegative matrix factorization (NMF) is a powerful tool for hyperspectral unmixing (HU). This method factorizes a hyperspectral cube into constituent endmembers and their fractional abundances. In this paper, we propose a two-stage nonnegative matrix factorization algorithm. During the first stage, k-means clustering is first employed to obtain the estimated endmember matrix. This matrix serves as the initial matrix for NMF during the second stage, where we design a new cost function for the purpose of refining the solutions of NMF. The two-stage NMF model is solved with multiplicative update rules, and the monotonic convergence of this algorithm is proven with an auxiliary function. Numerical tests demonstrate that our two-stage NMF algorithm can achieve accurate and stable solutions.

A non-interior point algorithm based on projection for second-order cone programming problems is proposed and analyzed. The main idea of the algorithm is that we cast the complementary equation in the primal-dual optimality conditions as a projection equation. By using this reformulation, we only need to solve a system of linear equations with the same coefficient matrix and compute two simple projections at each iteration, without performing any line search. This algorithm can start from an arbitrary point, and does not require the row vectors of A to be linearly independent. We prove that our algorithm is globally convergent under weak conditions. Preliminary numerical results demonstrate the effectiveness of our algorithm.