This study examined the relationship between the leaf reflectance of different seasons and the concentration of heavy metal elements in leaves, such as Co, Cu, Mo, and Ni in a post-mining area. The reflectance spectra and leaf samples of three typical plants were measured and collected in a whole growth cycle (June, July, August, and September). The Red Edge Position (REP), Readjustment Normalized Difference Vegetation Index (RE-NDVI), and Photochemical Reflectance Index (PRI) were extracted and used to explore its relation with the heavy metals concentrations in leaves between different seasons. The results show that all three Vegetation Indices (VIs) were insensitive indicators for monitoring the metal effects of vegetation in different seasons, which showed similar trends. Based on this, the Continuum Removal Indices (CRIs) were proposed from the continuum removed approach and extended for detecting the effects of heavy metal pollution over a full growth cycle. The relationship between the metal concentrations and CRIs of different plants was respectively analyzed by Stepwise Multiple Linear Regression (SMLR) and Partial Least Squares Regression (PLSR). It is found that a significant correlation exists between the band depth and the concentration of Cu and Ni based on the White birch data sets using the PLSR, resulting in a small deviation from the established relationships. Compared with VIs, the approach of coupling CRIs and multiple regressions was effective for improving the estimation accuracy. The presented study provides a detection model of leaf heavy metals that can be adapted to different growing cycles, even an arbitrary growing cycle.

Under the background of global warming, rising sea level, extreme weather and other global climate changes, vegetation has played a targeted and irreplaceable role. The characteristics of individual plant, community landscape and vegetation succession in response to the major driving factor (mainly includes habitat relative elevation, net loss of coastal habitat, salinity, etc.) were analyzed. An obvious development of vegetation landscape fragmentation has results from the competitive advantages of salt-tolerant species or invasive species, which eventually results in the regressive succession and unreasonable secondary succession of vegetation. Compared with the botanical community statistics method, the method of combined of GIS-mapping and remote sensing data provide a more effective way to extract the individual plant stress information, vegetation community structure and dynamic change of vegetation landscape pattern, which can reflect the spatial differentiation of the vegetation at a macro-scale. In addition, in view of the high-efficiency carbon sequestration capability of coastal wetland vegetation, the spatial distribution, temporal dynamic and extraction method of vegetation and soil sequestration were discussed. Synthesize above analysis result, further studies in vegetation response to global climate change were proposed, which need to be improved or expanded.

In this paper, we present the detection of ore‐forming elements migrated from soil and rock to vegetation leaves using hyperspectral data. The rock, soil (C, B, and A layer), and vegetation (root, stem, and leaf) samples with the measurements of spectra and element concentrations in the vertical section of the abandoned pits were collected. The wavelet approach is applied to analyze the correlations between spectral features and element concentrations. The results show that the significant correlations are found at a lower order mother wavelets such as Haar. Moreover, the significant correlations are negative between Cu and Mo elements and the wavelet energy vector of the soil and rock to vegetation leaves spectrum. The results also show that ore‐forming elements migrated from rock and soil to vegetation leaves can be detected using hyperspectral data in the study area.

In order to determine the space-time characteristics of vegetation growing, aiming at the problems of poor efficiency in remote sensing information extraction of traditional methods, based on the Google Earth Engine (GEE) platform, NDVI data was extracted using Landsat image in the growing season from 1990 to 2019, and the space-time characteristics of vegetation were discussed with unitary linearity regression. The results showed that, 1) Taking 2010 as the turning point, NDVI of surface vegetation in Liaodong Bay showed a trend of first increase, then decrease and then increase during the past 30 years. The total area of vegetation cover have less change, while the proportion of medium-high coverage (0.6<NDVI<0.8) and high coverage (0.8<NDVI<1) increased significantly, 2) The spatial distribution of NDVI in Liaodong Bay have obvious differences, which shows a trend of gradual decrease from the Liaohe estuary delta to both sides. Affected by the development of coastal resources, NDVI increased firstly and then decreased from land to sea. In conclusion, the space-time change of NDVI in Liaodong Bay is affected by human development activities and ecological restoration. As an important quantitative factor, NDVI can provide basic data for the evaluation of environmental carrying capacity and ecological restoration.

Video anomaly detection plays a vital role in intelligent video monitoring systems. It has found extensive applications in the fields of public safety and social security. Nevertheless, the area of video anomaly detection continues to be a formidable task because of the intricate nature of actual data and the challenge of precisely identifying anomalies. Current anomaly detection approaches suffer from overpowered generalization ability, weak spatio-temporal feature extraction capability, and insufficient global information aggregation ability. Therefore, an unsupervised Hyperbolic Graph-based Normalizing Flows (HGNF) model is proposed in this paper, which is constructed with a Spatio-temporal Encoder (STEncoder) and stacked normalizing flows to reduce the overgeneralization of auto-encoder-based anomaly detection models. STEncoder consists of spatio-temporal attention and inter-frame feature aggregation. In normalizing flows, a Poincaré ball graph extractor is developed to improve the representation ability of the dynamic changes of the input data, and a masked affine coupling block is established to improve the performance of this model in global information aggregation. According to the experimental results obtained on four public datasets, HGNF achieves excellent performance and the best AUC (Area Under the Curve) score of 74.5% on the UBnormal dataset.