Visual Behaviour and Cognitive Preferences of Users for Constituent Elements in Forest Landscape Spaces

Background: Elements of forest landscape spaces are important media through which landscape information is conveyed. Therefore, it is very important for designers and managers of forests to explore the relationship among visual behaviour, landscape preferences, and element characteristics. Purpose: This study took forest landscape spaces as the subject, discussed the characteristics of visual behaviour and cognitive preferences for landscape elements, and analysed the relationship among element characteristics, visual behaviour, and cognitive preferences in forest landscape spaces. The findings will help designers better plan the spatial composition of forest landscapes. Methods: We collected data from 53 graduate and undergraduate students and then used Spearman’s rho correlation analysis and multiple linear regressions to analyse the experimental data. Main results: 1. As the composition of forest landscape spaces varies and landscape elements are combined in different ways, visual behaviour towards landscape elements also differs. 2. People are easily attracted by highly fascinating landscape elements, but they will spend more time on low fascinating landscape elements. 3. Element characteristics significantly affect visual behaviour and cognitive preferences. Elements with high complexity or a large proportion of elements take more time for the participants to recognize, which reduces the evaluation of satisfaction.

Color Enhancement of Low Illumination Garden Landscape Images

The unfavorable shooting environment severely hinders the acquisition of actual landscape information in garden landscape design. Low quality, low illumination garden landscape images (GLIs) can be enhanced through advanced digital image processing. However, the current color enhancement models have poor applicability. When the environment changes, these models are easy to lose image details, and perform with a low robustness. Therefore, this paper tries to enhance the color of low illumination GLIs. Specifically, the color restoration of GLIs was realized based on modified dynamic threshold. After color correction, the low illumination GLI were restored and enhanced by a self-designed convolutional neural network (CNN). In this way, the authors achieved ideal effects of color restoration and clarity enhancement, while solving the difficulty of manual feature design in landscape design renderings. Finally, experiments were carried out to verify the feasibility and effectiveness of the proposed image color enhancement approach.

Herbs and Spices in the Florida Garden

This 10-page publication of the UF/IFAS Horticultural Sciences Department discusses culinary herbs and spices that can be grown in a Florida home garden or landscape. Information on the general cultural requirements, propagation, harvesting, and use of herbs is included as well as detailed descriptions of common culinary herbs. Major revision by Mary Salinas, Sydney Park Brown, and James M. Stephens.https://edis.ifas.ufl.edu/vh020

Direct Aerial Visual Geolocalization Using Deep Neural Networks

Unmanned aerial vehicles (UAVs) must keep track of their location in order to maintain flight plans. Currently, this task is almost entirely performed by a combination of Inertial Measurement Units (IMUs) and reference to GNSS (Global Navigation Satellite System). Navigation by GNSS, however, is not always reliable, due to various causes both natural (reflection and blockage from objects, technical fault, inclement weather) and artificial (GPS spoofing and denial). In such GPS-denied situations, it is desirable to have additional methods for aerial geolocalization. One such method is visual geolocalization, where aircraft use their ground facing cameras to localize and navigate. The state of the art in many ground-level image processing tasks involve the use of Convolutional Neural Networks (CNNs). We present here a study of how effectively a modern CNN designed for visual classification can be applied to the problem of Absolute Visual Geolocalization (AVL, localization without a prior location estimate). An Xception based architecture is trained from scratch over a >1000 km2 section of Washington County, Arkansas to directly regress latitude and longitude from images from different orthorectified high-altitude survey flights. It achieves average localization accuracy on unseen image sets over the same region from different years and seasons with as low as 115 meters average error, which localizes to 0.004% of the training area, or about 8% of the width of the 1.5 × 1.5km input image. This demonstrates that CNNs are expressive enough to encode robust landscape information for geolocalization over large geographic areas. Furthermore, discussed are methods of providing uncertainty for CNN regression outputs, and future areas of potential improvement for use of deep neural networks in visual geolocalization.

Quantitative Evaluation of Plant and Modern Urban Landscape Spatial Scale Based on Multiscale Convolutional Neural Network

Modern urban landscape is a simple ecosystem, which is of great significance to the sustainable development of the city. This study proposes a landscape information extraction model based on deep convolutional neural network, studies the multiscale landscape convolutional neural network classification method, constructs a landscape information extraction model based on multiscale CNN, and finally analyzes the quantitative effect of deep convolutional neural network. The results show that the overall kappa coefficient is 0.91 and the classification accuracy is 93% by calculating the confusion matrix, production accuracy, and user accuracy. The method proposed in this study can identify more than 90% of water targets, the user accuracy and production accuracy are 99.78% and 91.94%, respectively, and the overall accuracy is 93.33%. The method proposed in this study is obviously better than other methods, and the kappa coefficient and overall accuracy are the best. This study provides a certain reference value for the quantitative evaluation of modern urban landscape spatial scale.

Urban Landscape Information Atlas and Model System Based on Remote Sensing Images

City landscape is an element of city image, and city landscape logo should also belong to the entire system of the city image logo. Since the reform and opening up, China’s economic construction industry has attracted worldwide attention, and the improvement of people’s living standards has led to the continuous development of urban construction. The original urban style is gradually replaced by high-rise buildings, the differences between cities are getting smaller and smaller, and the people living in them have gradually abandoned the inheritance of various cultures and customs. This paper aims to study the urban landscape information map and its model system based on remote sensing images. With the support of remote sensing image technology and geographic information system platform, the urban landscape information map model system is developed, which can vividly reflect the changes in the urban landscape pattern. The urban landscape information map is used to display and reveal the spatial evolution of the urban landscape in the process of urban development. Result of empirical analysis: it summarizes the methods of geoscience information map and urban landscape information map and establishes the goal and content of researching urban landscape information map. And it provides a basis for solving the problems of urban development and urban management. In 2021, Shanghai’s urban areas have occupied more than 80%. It is conceivable that the area of arable land must be very small, but the area of green land still accounts for 40%; compared with last year, it is still an increase of 6%.

EKF-based self-attitude estimation with DNN learning landscape information

This paper presents an EKF-based self-attitude estimation with a DNN (deep neural network) learning landscape information. The method integrates gyroscopic angular velocity and DNN inference in the EKF. The DNN predicts a gravity vector in a camera frame. The input of the network is a camera image, the outputs are a mean vector and a covariance matrix of the gravity. It is trained and validated with a dataset of images and corresponded gravity vectors. The dataset is collected in a flight simulator because we can easily obtain various gravity vectors, although the method is not only for UAVs. Using a simulator breaks the limitation of amount of collecting data with ground truth. The validation shows the network can predict the gravity vector from only a single shot image. It also shows that the covariance matrix expresses the uncertainty of the inference. The covariance matrix is used for integrating the inference in the EKF. Flight data of a drone is also recorded in the simulator, and the EKF-based method is tested with it. It shows the method suppresses accumulative error by integrating the network outputs.

Landscape perspectives in hydrological understanding and modelling for water management

<p>The Darcy medal acknowledges water-resources research, engineering and management. In my medal lecture I will embrace these aspects by telling the story of how my team merges numerical models and observations with landscape information to learn about hydrological processes and provide decision-support to society. We predict spatial and temporal variability of water fluxes and resources at local, regional and global scales to estimate hydrological variables in the past, present and future. We also explore “what if” scenarios for societal planning. Such predictions provide useful knowledge to maintain water resources at suitable quantities and qualities, despite on-going global warming, urbanization and environmental change. Water is the basis for all life and most societal sectors; hence, it must be managed properly for sustainable development. I will demonstrate how our scientific findings from the model applications have influenced water resources engineering and management policy.</p><p>Water management is always local but wider landscape information, such as knowledge about upstream/downstream conditions and residence-time, is needed when designing management measures. Water resources are normally shared by many stakeholders often with opposing objectives. Here, we found that models can have added value for science communication, participatory processes and conflict resolution to reach environmental goals.</p><p>It is well known that numerical models are more or less wrong and linked with uncertainties, but nevertheless, models combined with multiple sources of observations can be very helpful to aggregate information, quantify influence from various processes and describe outcome of complex phenomena. From modelling experiments, I will show how we reached deeper understanding of hydrological process when using the landscape perspective and large-sample empirical data across different physiographical conditions. Linking the model to landscape characteristics also gave us the possibility to make water predictions with some confidence even in data sparse regions and for ungauged catchments.</p><p>Large-scale modelling of water resources should be accompanied with site-specific data and local knowledge to be applicable for water resources engineering and management. Therefore, we share our model and I will exemplify how we reach a better understanding and make use of new science in collaborative efforts across the globe. Recently, the modelled data was also aggregated into societal-relevant indicators and provided through web-based climate and water services. During co-development of such on-line tools with practitioners, however, we encountered a large knowledge gap between data producers and data users, which calls for mutual engagement to reach understanding.</p><p>To sum up, my team uses and provides open data, open science and community building world-wide to accelerate water research by sharing local insights and collective intelligence in addressing multiple landscapes. Yet, scientific knowledge is always preliminary and needs to be challenged by peers and explored by users to be practically beneficial. I therefore advocate for science communication as an emerging field to engage more with. Hydrological scientists have a lot to contribute and learn in dialogues to find hope and solutions under global change, which will help in sustaining the water resources and the Planet as we know it.</p>

Modelagem da Informação da Paisagem

Este artigo apresenta a plataforma LIM vista como uma ferramenta para integrar e operacionalizar os projetos complementares de paisagem. Para tanto propõe o desenvolvimento do Landscape Information Modeling (LIM) que consiste em uma plataforma integrada de projeto capaz de simular aspectos dos espaços projetados, compatibilizando elementos construídos e processos naturais. Acredita-se que o projeto da paisagem, por ser uma dimensão integradora das camadas de intervenção no espaço, permite maior transparência e democratização do processo de decisão e execução, podendo ser orientado para o atendimento das demandas sociais, ecológicas, econômicas e do respeito as suas particularidades históricas. Para estruturar o LIM serão definidos um conjunto de parâmetros de projeto caracterizados por sua diversidade, adaptação e responsividade. Ao incluir diferentes parâmetros pode-se propor inúmeras variações para o modelo, de forma a objetivar diferentes soluções e avaliar seu grau de resposta e adaptação e, portanto, da sua inteligência, aqui entendida como a capacidade de adaptação do sistema as novas condições oferecidas pelo projeto.