A deep inverse convolutional neural network-based semantic classification method for land cover remote sensing images.

The imbalance of land cover categories is a common problem. Some categories appear less frequently in the image, while others may occupy the vast majority of the proportion. This imbalance can lead the classifier to tend to predict categories with higher frequency of occurrence, while the recognition effect on minority categories is poor. In view of the difficulty of land cover remote sensing image multi-target semantic classification, a semantic classification method of land cover remote sensing image based on depth deconvolution neural network is proposed. In this method, the land cover remote sensing image semantic segmentation algorithm based on depth deconvolution neural network is used to segment the land cover remote sensing image with multi-target semantic segmentation; Four semantic features of color, texture, shape and size in land cover remote sensing image are extracted by using the semantic feature extraction method of remote sensing image based on improved sequential clustering algorithm; The classification and recognition method of remote sensing image semantic features based on random forest algorithm is adopted to classify and identify four semantic feature types of land cover remote sensing image, and realize the semantic classification of land cover remote sensing image. The experimental results show that after this method classifies the multi-target semantic types of land cover remote sensing images, the average values of Dice similarity coefficient and Hausdorff distance are 0.9877 and 0.9911 respectively, which can accurately classify the multi-target semantic types of land cover remote sensing images.

Atomistic liquid crystalline structures of discotic bent-core-like mesogens formed by hydrogen bonding and interchain interactions.

Integrated atomistic and molecular dynamic simulations are used to characterize the role hydrogen bonding and interchain interactions on structures and phase transitions of novel bent-core-like mesogenic materials that exhibit new self-assembly features, attractive to the development of functional materials. Multi-step simulations were carried out to model phase transitions and various spectra of two complex mesogenic materials formed from acid functionalized azo compounds, 4-[2,3,4-tri(octyloxy)phenylazo] benzoic acid and 4-[2,3,4- tri(heptyloxy)phenylazo] benzoic acid. The simulations contain three consecutive steps, involving molecular quantum chemistry, molecular crystal packing, and super cell molecular dynamics calculations. These two mesogens are supposed to form phasmidic molecular conformers. However, simulations point to the formation of complex discotic bent-core-like liquid crystals with tetramer mesogenic assemblies, in very good agreement with experimental observations. The wide range agreements between simulations and experimental results include transitions of crystal structures to columnar and uniaxial nematic phases, x-ray diffraction patterns of columnar phases, the structure of the two-dimensional complex bent-core-like tetramers, molecular Raman spectra, Raman depolarization spectra, and order parameters of nematic phases. The multi-step simulation methodology and its results shed light on this unique behaviour of plasmids with flexible side chains for simulation design of novel bent-core-like mesogenic materials.

Temperature influence on the structure and interfacial properties of surfactin micelle: a molecular dynamics simulation study.

Surfactin is an efficient biosurfactant excreted by different strains of Bacillus subtilis. Our study provides a molecular view of the temperature dependence of the structure and the interfacial properties of un-ionized surfactin micelles. The overall size and shape, the surface area, the radial density distribution of the micelles, the conformation of the hydrocarbon chain, and the intramolecular/intermolecular hydrogen bonds formed in surfactin molecules were investigated. The micelles were mostly in sphere shapes, and the radii of surfactin micelle were estimated to be around 2.2 nm. The peptide rings occupied most of the surface of the micelles. Small amounts of ß-turn and γ-turn structures were found in the conformations of the peptide rings. When the temperature increased, the shape of the peptide rings became planar; the solvent accessible surface area decreased as temperature dehydration occurred. At 343 K some hydrocarbon chains reversed their orientation (flip-flopped). In addition, the stability of the hydrogen bond interactions in the micelles decreases with the increasing temperature.