Image fusion aims at exploiting complementary information in multimodal images to create a single composite image with extended content. An framework is proposed for different types of fast filtering the spatial domain. First, gradient magnitude used detect contrast and sharpness. Second, morphological closing operation performed on bridge gaps fill holes. Third, weight map obtained from filtered by structure-preserving filter. Finally, fused composed using weighed-sum rule. Experimental...

The advancing of glaciers is a manifestation dynamic glacial instability. Glaciers in the Tien Shan region, especially Central Shan, show instability, and have been recently detected. In this study, we used Landsat TM/ETM+/OLI remote sensing images to identify region from 1990 2019 found that 48 advanced. Among them, thirty-four exhibited terminal advances, 14 experienced advances on tributary or trunk. Ten experiencing identified as surging glaciers. These are distributed western part Halik...

DBSCAN is one of the most popular algorithms for cluster analysis. It can discover all clusters with arbitrary shape and separate noises. But this algorithm canpsilat choose parameter according to distributing dataset. simply uses global MinPts parameter, so that clustering result multi-density database inaccurate. In addition, when it used large databases, will cost too much time. For these problems, we propose GMDBSCAN which based on spatial index grid technique. An experimental evaluation...

Plants are ubiquitous in human life. Recognizing an unknown plant by its leaf image quickly is a very interesting and challenging research. With the development of processing pattern recognition, recognition based on has become possible. Bag features (BOF) one most powerful models for classification, which been used many projects studies. Dual-output pulse-coupled neural network (DPCNN) shown good ability texture such as segmentation. In this paper, method BOF DPCNN (BOF_DP) proposed...

Image intensity value is determined by both the albedo component and shading component. The describes physical nature of different objects at surface earth, land-cover classes are from each other because their intrinsic materials. We, therefore, recover feature hyperspectral image to exploit spatial semantic information. Then, we use support vector machine (SVM) classify recovered image. SVM tries maximize minimum margin achieve good generalization performance. Experimental results show that...

Xinjiang is located in an arid and semi-arid climate region China, but Ili river valley more humid, with higher precipitation intensity precipitation, which closely related to the role of Tianshan Mountains. In this paper, through NCRP 1° × reanalysis data conventional observation River Valley Xinjiang, terrain sensitivity experiment conducted by WRF model used analyze short-term extreme event from 18–19 May 2017, reveal influence Mountains on Valley. The results show that: (1) reduction or...

The complexity of clouds, particularly in terms texture detail at high resolutions, has not been well explored by most existing cloud detection networks. We introduce the high-resolution network (HR-cloud-Net), which utilizes a hierarchical integration approach. HR-cloud-Net integrates representation module, layer-wise cascaded feature fusion and multiresolution pyramid pooling module to effectively capture complex features. This architecture preserves detailed information while facilitating...

The complexity of clouds, particularly in terms texture detail at high resolutions, has not been well explored by most existing cloud detection networks. This paper introduces the High-Resolution Cloud Detection Network (HR-cloud-Net), which utilizes a hierarchical high-resolution integration approach. HR-cloud-Net integrates representation module, layer-wise cascaded feature fusion and multi-resolution pyramid pooling module to effectively capture complex features. architecture preserves...

Comprehensive integrated research on ecological and hydrological processes the simulation of river basin environments are critical foundations for decision making by governments river-basin managers. The demand a holistic understanding environmental systems such as basins is increasing. Eco-hydrological needs two types monitoring platforms to access collect data from basins: modeling platform support access, select, run models online, build new with collected data, manipulation generate...

We propose a fast document image thresholding method (FADIT) and evaluations of the two classic methods for demonstrating effectiveness FADIT. put forward assumptions: (1) probability occurrence grayscale text background is ideally constants, (2) pixel with low has high being classified as background. With assumptions, new criterion function applied to in Bayesian framework. The been borne quantitative metric well qualitative comparisons state-of-the-art methods.

In recent years, the scientific research model of Data - Model Simulating has become one main methods to support surface process in high-cold environment (alpine cold area and high latitude area). This kind mode needs e-Geoscience based on data, models, performance computing, visualization collaborative support. this paper, a highly efficient platform named High-cold joint Observation Research cloud China (HeorCloud) is established for Geoscience regions china computing technologies....

The performance of graph-based clustering methods highly depends on the quality data affinity graph as a good can approximate well pairwise similarity between samples. To large extent, existing construct based fixed distance metric, which is often not an accurate representation underlying structure. Also, they require postprocessing to obtain results. Thus, results are sensitive particular construction methods. address these two drawbacks, we propose <mml:math...