A Guided Multi-Scale Categorization of Plant Species in Natural Images

In natural images, the scales (thickness) of object skeletons may dramatically vary among objects and object parts. Thus, robust skeleton detection requires powerful multi-scale feature integration ability. To address this issue, we present a new convolutional neural network (CNN) architecture by introducing a novel hierarchical feature integration mechanism, named Hi-Fi, to address the object skeleton detection problem. The proposed CNN-based approach intrinsically captures high-level semantics from deeper layers, as well as low-level details from shallower layers. By hierarchically integrating different CNN feature levels with bidirectional guidance, our approach (1) enables mutual refinement across features of different levels, and (2) possesses the strong ability to capture both rich object context and high-resolution details. Experimental results show that our method significantly outperforms the state-of-the-art methods in terms of effectively fusing features from very different scales, as evidenced by a considerable performance improvement on several benchmarks.

Download Full-text

Importance of Soil Organic Matter and the Species Pool for Local Species Richness in Montane Ecosystems

Sustainability ◽

10.3390/su131910634 ◽

2021 ◽

Vol 13 (19) ◽

pp. 10634

Author(s):

Xiang Li ◽

Wenhao Hu ◽

Zhenrong Yu

Keyword(s):

Species Richness ◽

Plant Species ◽

Plant Species Richness ◽

Species Pool ◽

Multi Scale ◽

Local Species Richness ◽

Local Plant ◽

Regional Species Pool ◽

Urban Rural ◽

Local Species

Understanding the response of plant species richness to environmental filters is critical for conservation management as there is an increasing emphasis on plant restoration in urban/rural planning. However, empirical studies on the effects that the regional species pool has on plant species richness often overlook small spatial scales, therefore requiring more comprehensive approaches. As mountains can act as barriers to plant dispersal, the impact on the species pool, particularly, should be a priority. This study aimed to investigate how the regional species pool affects the local plant species richness in a multivariate context. We sampled vascular plant communities along three transects located in three valleys across the Chongli District, China, where four common habitat types were selected for sampling: grassland, shrubbery, pure forest, and mixed forest. We compared the differences in the multi-scale species richness and species composition between habitats and regions and used piecewise structural equation modeling to analyze the relative importance of the regional species pool, habitat species pool, soil resource availability, and exposure for local plant richness. The β-diversity had the highest contribution to the total species richness between valleys and habitats. The species composition between regions and habitats showed a significant difference and the local species richness was most strongly affected by the soil characteristics, but effects from the regional species pool still played an important role. Conservation efforts and urban/rural planning should use a multi-level and multi-scale approach based on a detailed structural investigation.

Download Full-text

An implicit Markov random field model for the multi-scale oriented representations of natural images

2009 IEEE Conference on Computer Vision and Pattern Recognition ◽

10.1109/cvprw.2009.5206797 ◽

2009 ◽

Author(s):

Siwei Lyu

Keyword(s):

Random Field ◽

Markov Random Field ◽

Field Model ◽

Natural Images ◽

Random Field Model ◽

Markov Random Field Model ◽

Multi Scale ◽

Markov Random

Download Full-text

Seed dispersal as a search strategy: dynamic and fragmented landscapes select for multi-scale movement strategies in plants

Movement Ecology ◽

10.1186/s40462-020-00239-1 ◽

2021 ◽

Vol 9 (1) ◽

Author(s):

Jelle Treep ◽

Monique de Jager ◽

Frederic Bartumeus ◽

Merel B. Soons

Keyword(s):

Seed Dispersal ◽

Plant Species ◽

Suitable Habitat ◽

Dispersal Strategy ◽

Plant Dispersal ◽

Multi Scale ◽

Reference Framework ◽

Trade Offs ◽

Movement Strategies ◽

Wide Range

Abstract Background Plant dispersal is a critical factor driving ecological responses to global changes. Knowledge on the mechanisms of dispersal is rapidly advancing, but selective pressures responsible for the evolution of dispersal strategies remain elusive. Recent advances in animal movement ecology identified general strategies that may optimize efficiency in animal searches for food or habitat. Here we explore the potential for evolution of similar general movement strategies for plants. Methods We propose that seed dispersal in plants can be viewed as a strategic search for suitable habitat, where the probability of finding such locations has been optimized through evolution of appropriate dispersal kernels. Using model simulations, we demonstrate how dispersal strategies can optimize key dispersal trade-offs between finding habitat, avoiding kin competition, and colonizing new patches. These trade-offs depend strongly on the landscape, resulting in a tight link between optimal dispersal strategy and spatiotemporal habitat distribution. Results Our findings reveal that multi-scale seed dispersal strategies that combine a broad range of dispersal scales, including Lévy-like dispersal, are optimal across a wide range of dynamic and patchy landscapes. At the extremes, static and patchy landscapes select for dispersal strategies dominated by short distances, while uniform and highly unpredictable landscapes both select for dispersal strategies dominated by long distances. Conclusions By viewing plant seed dispersal as a strategic search for suitable habitat, we provide a reference framework for the analysis of plant dispersal data. Consideration of the entire dispersal kernel, including distances across the full range of scales, is key. This reference framework helps identify plant species’ dispersal strategies, the evolutionary forces determining these strategies and their ecological consequences, such as a potential mismatch between plant dispersal strategy and altered spatiotemporal habitat dynamics due to land use change. Our perspective opens up directions for future studies, including exploration of composite search behaviour and ‘informed searches’ in plant species with directed dispersal.

Download Full-text