Figure 6 - uploaded by Ender Bugday
Content may be subject to copyright.
Source publication
Currently, it is very important to identify and use the most appropriate methods in the management of limited resources and to reach a conclusion in a short time period by using the technology in an effective manner to fastly obtain information in high quality. Remote sensing (RS) techniques are used as a very effective tool for this purpose. Obtai...
Similar publications
![Fig. 1. Change detection LULC map for 1996 and 2016 [11]](profile/Wynnezel-Wayne-Naoto-Akeboshi/publication/367431407/figure/fig1/AS:11431281114872656@1674700808907/Change-detection-LULC-map-for-1996-and-2016-11_Q320.jpg)
![Fig. 3. Fields under AI [9]](profile/Wynnezel-Wayne-Naoto-Akeboshi/publication/367431407/figure/fig2/AS:11431281114892054@1674700808984/Fields-under-AI-9_Q320.jpg)
![Fig. 4. Deep Learning Process in Image Classification [9]](profile/Wynnezel-Wayne-Naoto-Akeboshi/publication/367431407/figure/fig3/AS:11431281114872657@1674700809019/Deep-Learning-Process-in-Image-Classification-9_Q320.jpg)
![Fig. 5. CNN Architecture [15]](profile/Wynnezel-Wayne-Naoto-Akeboshi/publication/367431407/figure/fig4/AS:11431281114866071@1674700809050/CNN-Architecture-15_Q320.jpg)
A prompt change over an area can be observed in Land Use and Land Cover maps (LULC) brought by several factors, thus classifying LULC can give vital information such as the scale of commercialized and residential lands, and the availability of forest and barren land. In this study, CNN using pre-trained models such as ResNet50, VGG16, and Efficient...
Citations
... Accurate LULC maps are important for monitoring, planning, and managing the Earth and remote sensing studies [11][12][13]. Besides, the Geographic Information System (GIS) is one of the important tools for documenting, storing, evaluation and presenting sites for making accurate decisions [14][15][16]. Remote sensing is capturing the information of an object or land t through sensing either own or artificially emitted electromagnetic radiation [17]. ...
Land Use and Land Cover Analysis are important in detecting the changes in urban areas, rural areas, and focused lands like university campuses. The availability of high-quality satellite images from diverse time sequences makes evaluations for changes by time possible. The analysis methods include insights from remote sensing fields to Artificial intelligence (AI) tools. AI has been significantly developed in the last decades in various fields, and applications of AI on satellite imagery analysis are being influenced. This study explores the capability of Chat GPT, which is one of the leading Language Models and can generate prompts and analysis due to inputs for Land Cover and Use Analysis. Firstly, an unstructured conversation with Chat GPT was held, and then, considering this experience, a land cover change analysis was executed for two university campuses. Besides, the analysis was also re-executed in Colab with codes generated by Chat GPT to seek differences. Two university campuses, Erzurumm Technical University and Adıyaman University, founded in the last two decades, were utilized as case studies. Chat GPT explained the steps and procedure of the analysis in detail generated codes in a defined framework. The analysis results have problems in classifying the land cover; however, the imperviousness change analysis shows most of the construction improvement. The experiment and findings have important implications for future research in Land Cover analysis implementing AI tools.
... En yüksek doğruluğu SVM yöntemi sağlamıştır. Literatürde SVM yönteminin ANN yönteminden daha yüksek doğruluk sağladığı[3][4][5][6][7][8] çalışmalar yer almaktadır. Khurana ve Gupta, Hindistan'da Sentinel-2A görüntülerine SVM, ANN ve MLC piksel tabanlı sınıflama yöntemleri uygulayarak arazi örtüsünü araştırmıştır. ...
Uydu teknolojilerinin gelişmesiyle birlikte, arazi kullanımı ve arazi örtüsü belirlemek amacıyla uzaktan algılama teknikleri daha sık kullanılmaktadır. Nergizlik barajı önemli bir su kaynağı olup, çevresinde yerleşim alanları, tarım arazileri ve orman alanları bir arada bulunduğu için arazi örtüsünün düzenli olarak gözlemlenmesi gereklidir. Bu çalışmanın temel amacı Nergizlik Baraj Gölü çevresinde piksel tabanlı görüntü sınıflandırma yöntemleri ile sınıflama işlemi uygulayarak ve bant oranlama teknikleri kullanarak arazi kullanım ve arazi örtüsünü belirlemek, bu tekniklerin genel doğruluklarını karşılaştırmak ve çalışma alanına en uygun yöntemi tespit etmektir. Bu çalışmada Nergizlik Baraj Gölü’nün 3 Ocak 2024 tarihinde alınan Sentinel-2B uydu görüntüsü kullanılmıştır. Görüntü, kırmızı, mavi ve yeşil bantları kullanılarak, Support Vector Machines (SVM), Artificial Neural Networks (ANN) yöntemiyle piksel tabanlı sınıflandırma yöntemiyle sınıflandırılmıştır. Görüntüye Normalized Difference Vegetation Index (NDVI), Normalized Difference Red Edge Index (NDRE) bant oranlama teknikleri de uygulanmıştır. Arazi kullanımı ve arazi örtüsü haritaları oluşturulmuştur. Bu çalışmada SVM %89, ANN %88, NDVI %81, NDRE %84 doğruluk sağlamıştır. Tüm yöntemler beklenen doğruluğu sağlamış olup, piksel tabanlı SVM sınıflama yönteminin diğer yöntemlere göre daha yüksek doğruluk sağladığı tespit edilmiştir.
... (1) distance from roads, (2) distance from water bodies, (3) distance from urban centers, (4) digital elevation model (DEM), and (5) population density. These biophysical variables are widely used to analyze patterns of change and extract effective information about the impact of human activities on land use change (DOĞAN and BUĞDAY, 2018), and were used with the same pixel size and coordinate system maps. These spatial variables were provided by the Sergipe State Water ...
The forest ecosystems of the Atlantic Forest biome in Brazil are experiencing intense deforestation, contributing to climate change. Valuing the ecosystem service of CO2 sequestration can help in environmental policies to reduce deforestation and promote ecological restoration in this region. In this study, we sought to understand the dynamics of land use and land cover in the Atlantic Forest region of the state of Sergipe between 1985 and 2021 and to simulate land use and land cover for 2031 and 2051, assuming the current business as usual (BAU) trend was used as a reference scenario. The results showed that the agriculture-pasture mosaic class obtained the greatest percentage (5.27%) and net (around 514.00 km²) increase among the land use and cover classes evaluated in the study area between 1985 and 2021. In the BAU scenario, total emissions 2 reach 1,137,942.07 Mg CO2 in 2031 and 692,815.16 Mg CO2 in 2051, imposing an opportunity cost for society of US 4.35 million, respectively.
... Dogan et. al [24] modelled an artificial neural network to map the land use and land cover in Kastamonu area. Babu et al [25] suggested a method which used pixel based classification with LSTM classifier applying swarm optimization. ...
Agroforestry is indeed a traditional practice followed in tropical countries like India. About 28.43 million hectare area is used for agroforest cultivation. By 2050 India has the mission of increasing the area under agroforestry to 53 million hectares. In this study, we have made an effort to map the agroforest areas using the geospatial tools and hybrid deep learning techniques. The land utilized for cultivation and various agroforestry activities such as rubber, tea, coconut, and banana plantation were classified as forest canopy by the existing classifiers taking the tree canopy density as a parameter. In light of proposing a solution to the issue, we have put forth a multi temporal hybrid deep learning framework which is a fusion of convolutional neural network, a deep neural net and long short term memory network to classify agroforestry distinguishing it from the forest canopy using remote sensing data. The experimentation was carried out in the southern districts of India, and Landsat 8 imagery was used to classify the agroforestry of the study area that includes tea, banana, rubber, coconut, and crop lands. An efficient multi temporal hybrid deep learning framework was designed to classify the agroforest plantation distinguishing it from crop lands and forest clusters. The experimental results of multi temporal hybrid CNN+LSTM outperformed CNN, LSTM, BiLSTM model reducing the error rate with respective accuracy and kappa score of 98.23% and 0.88. The proposed method provides a benchmark to accurately classify and estimate the LULC, particularly mapping the agroforest plantation for other regions across the country.
... The neighbourhood value defines the number of neighbouring pixels around the current pixel, while the number of hidden layers enables the network to learn more complex problems [12]. The remaining learning parameters (i.e., learning rate, number of iterations, and momentum) define how quick or slow the learning process should be [22]. The resulting transition matrix derived from our input variables (i.e., LULC maps and spatial variables) provides information on temporal changes between LULC classes by indicating the likelihood of a particular class transforming to another class. ...
Green spaces such as forests, grasslands, and croplands (including gardens) can be found in urban environments.
Although they benefit human and animal well-being, they have become threatened due to rapid urban growth
and unplanned development. Yet, little attention has been given to studying the dynamics of urban green spaces
in sub-Saharan Africa. In this study, we examined the dynamics of land use and land cover (LULC) change and
fragmentation (especially, green spaces) within the second fastest urbanising city in Ghana, Tamale. In particular,
we focused our analyses on its urban core (~5 km radius around the city centre) due to its relevance to
urban economy and society. Landsat data was used to estimate fragmentation metrics of past and future LULC
changes in the study area from 1990 to 2052. We found clear patterns of green space decline and fragmentation
within the urban core: i.e., green spaces became patchy over time and the pattern was expected to continue in the
future. Additionally, we found the built-up class benefited from the decline of green spaces with the latter being
significantly negatively correlated with human population size. Our investigation reveals that protected forests
and tree plantations contributed to a significant proportion of available green spaces in the urban core. However,
these areas were becoming increasingly threatened by forest reserve downsizing, indiscriminate activities (e.g.,
logging and encroachment), and sale of public lands to private developers, practices commonly associated with
population growth. Hence, the enforcement of relevant local legislations (e.g., the 2016 Land Use and Spatial
Planning Act [Act 925]) coupled with the integration of urban initiatives and policies that encourage green
spaces are needed to ensure the sustainability of urban ecosystems for the well-being of humans and the
environment.
... A simulação do UOS futuro possibilita uma série de análises espaciais, entre as quais podem ser destacados os estudos que visam compreender a evolução da cobertura florestal e sua substituição por agricultura (Tornquist & Silva 2019), a evolução das áreas urbanas visando dar suporte aos planejadores e tomadores de decisão na construção de um ambiente mais sustentável (Wang et al. 2018) e a interferência nos processos hidrológicos e consequências nos processos biofísicos e na precipitação regional (Quesada et al. 2017). A simulação para os cenários dos UOS futuros vem sendo amplamente realizada com a utilização de Redes Neurais artificiais (Doğan & Buğday 2018;Qiang & Lam 2015;Silva et al. 2020), por permitir a resolução de problemas como complexos, uma vez que são capazes de aprender a partir de exemplos e com isto generalizar a informação, sendo fundamental o processo de treinamento e validação (Fleck et al. 2016) Dentre os diversos relatos de modelagem com a utilização de redes neurais, destaca-se o uso do Perceptron Multicamadas (MLP) que consiste em um modelo com três camadas que pode identificar relações que são não lineares na natureza e tomar decisões sobre qual parâmetro usar na modelagem e que modificações realizar de forma a obter o melhor resultado (Dzieszko 2014). Para as predições se utiliza da Cadeia de Markov, que é um processo que se caracteriza pela determinação do estado futuro em função apenas do estado atual, onde os estados pretéritos não exercem influência sobre o estado futuro (Levin & Peres 2017). ...
O processo de mudança do uso e ocupação do solo (UOS), com destaque para o desmatamento da Mata Atlântica, é um dos principais fatores que influenciam o fenômeno do aquecimento global, visto que significativas quantidades de carbono deixam de ser retidas e são liberadas na atmosfera. Esse processo vem sendo observado em quase toda a Zona Costeira brasileira, na qual está inserido o município de Lucena, no estado da Paraíba. Diante desta problemática, o presente estudo objetivou avaliar quantitativamente e espacialmente o estoque de carbono decorrente das mudanças do UOS, para os períodos pretéritos de 1994 a 2018 e futuros de 2030 e 2040, considerando as suas tendências de alteração em Lucena. Para isso foram integradas duas metodologias o LCM ( Land Change Modeler ) para simulação do UOS futuro e o CSS ( Carbon Storage and Sequestration) para a quantificação do estoque de carbono. A análise preditiva realizada para os anos de 2030 e 2040 indicou uma tendência de crescimento da área urbana, destacando-se a ocupação ao longo da planície litorânea e na porção mais oriental do Tabuleiro Costeiro. Em sentido oposto, a Mata Atlântica é que apresenta a maior redução, estimando-se que até o ano de 2040 ocorra uma redução de 3% por ano, de forma que a cobertura florestal do munícipio corresponda a apenas 9,82% do seu território. Em relação ao estoque total de carbono a tendência é a redução da capacidade de retenção por hectare, estimando-se que até o ano de 2040 haverá uma redução de 48,04% do armazenamento observado no ano de 1994. Assim, pôde ser concluído que a diminuição do estoque de carbono em Lucena é fortemente influenciada pelo UOS e pelo desmatamento da Mata Atlântica.
... In [93], the authors detect LULC changes using cellular automata (CA) and ANN architectures, where they rely on extracting spatial and explicit representations of LULC dynamics at cross-border region scales. Similarly, in [94], the LULC changes between a period of 18 years (1999 to 2016) in the central district of Kastamonu (Turkey) are detected using ANN. Hence, the authors use satellite images of the study region to classify the LULC changes and model the probable land area. ...
Analyzing satellite images and remote sensing (RS) data using artificial intelligence (AI) tools and data fusion strategies has recently opened new perspectives for environmental monitoring and assessment. This is mainly due to the advancement of machine learning (ML) and data mining approaches, which facilitate extracting meaningful information at a large scale from geo-referenced and heterogeneous sources. This paper presents the first review of AI-based methodologies and data fusion strategies used for environmental monitoring, to the best of the authors’ knowledge. The first part of the article discusses the main challenges of geographical image analysis. Thereafter, a well-designed taxonomy is introduced to overview the existing frameworks, which have been focused on: (i) detecting different environmental impacts, e.g. land cover land use (LULC) change, gully erosion susceptibility (GES), waterlogging susceptibility (WLS), and land salinity and infertility (LSI); (ii) analyzing AI models deployed for extracting the pertinent features from RS images in addition to data fusion techniques used for combining images and/or features from heterogeneous sources; (iii) describing existing publicly-shared and open-access datasets; (iv) highlighting most frequent evaluation metrics; and (v) describing the most significant applications of ML and data fusion for RS image analysis. This is followed by an overview of existing works and discussions highlighting some of the challenges, limitations and shortcomings. To provide the reader with insight into real-world applications, two case studies illustrate the use of AI for classifying LULC changes and monitoring the environmental impacts due to dams’ construction, where classification accuracies of 98.57% and 97.05% have been reached, respectively. Lastly, recommendations and future directions are drawn.
... The spatial variables including elevation, slope, and population density, as well as the distance to town and roads, were considered. These biophysical and socio-economic variables are widely used to analyze the change patterns and to extract effective information about the effects of human activities on land-use change (Dogan and Bugday, 2018). ArcGIS 10.1 software was used to prepare a raster spatial map. ...
... Artificial Neural Networks (ANNs) are widely used in the estimation-based processes of several fields of engineering, such as aircraft, automobiles, electronics, production, robotics, communications, and civil engineering. ANNs can be a very useful tool in engineering practices and can be a strong tool in data modeling (Esteban et al. 2009;Atkinson and Tatnall 1997;Ashraf et al. 2013;Buğday 2018;Doğan and Buğday 2018 Binoti et al., (2014), Bhering et al. (2015), Miguel et al. (2016) and Sanquetta et al. (2018) found the best predictive volume by using ANN with respect to other prediction methods. In addition to these studies, it is particularly necessary to conduct more studies on artificial neural networks that can be defined as a member of artificial intelligence and ANN as the new technique may probably provide an opportunity to obtain more accurate and predictive volume predictions in the field of forestry beyond classical regression models. ...
Cilj ovog rada je usporediti modele umjetne neuralne mreže (ANN) za predviđanje pojedinih drvnih volumena krimskih borova u šumama Çankirija. Jednoulazne i dvoulazne jednadžbe i kompatibilna volumna jednadžba Fang et al. (2000) temeljena na klasičnim i tradicionalnim metodama primijenjena je na 360 krimskih borova u cilju dobivanja ovih drvnih volumena. Kako bi se odredila najbolja alternativna metoda za predviđanje ANN modela, ukupno je obučeno 320 treniranih mreža u višeslojnom perceptronu (MLP) i ukupno 20 treniranih mreža u arhitekturi Radial Basis Function (RBF). Na temelju statistike goodness-of-fit , ANN u smislu MLP 1-9-1 uključujući dbh kao input varijablu za jednoulazna volumna predviđanja pokazao je bolju fitting sposobnost sa SSE (2.7763), Radj2 (0.9339), MSE (0.00910), RMSE (0.0954), AIC (-823.25) i SBC (-1421.81) nego onaj u ostalim proučavanim volumnim metodama koje uključuju dbh kao eksplanatornu varijablu. Za dvoulazna volumna predviđanja, što uključuju dbh i ukupnu visinu kao input varijable, ANN temeljen na MLP 2-15-1 rezultirao je boljom fitting statistikom sa SSE (0.8354), Radj2 (0.9801), MSE (0.00274), RMSE (0.0523), AIC (-579.55) and SBC (-1788.11).
Uydu tekniklerinin gelişmesiyle birlikte uzaktan algılama çalışmalarında, LULC (arazi kullanımı ve arazi örtüsü) belirlemek amacıyla görüntü sınıflandırma algoritmaları daha sık kullanılmaktadır. Bu çalışmanın temel amacı Yedigöze Baraj Gölü çevresinde nesne tabanlı ve piksel tabanlı görüntü sınıflandırma yöntemleri ile sınıflama işlemini gerçekleştirmek ve sınıflama tekniklerinin genel doğruluklarını karşılaştırmaktır. Bu çalışmada Yedigöze Baraj Gölü'nün 24 Haziran 2023 tarihinde alınan Sentinel-2B uydu görüntüsü kullanılmıştır. İlk olarak görüntü, SVM (destek vektör makineleri), ANN (yapaya sinir ağları) ve MLC (Maksimum olabilirlik sınıflandırması) yöntemiyle piksel tabanlı sınıflandırma yöntmiyle sınıflandırılmıştır. Daha sonra görüntü, KNN (K-en yakın komşuluk) yöntemiyle nesne tabanlı sınıflandırma yöntemiyle sınıflandırılmıştır. Bu algoritmalar kullanılarak su alanı, tarım alanı, orman alanı ve yerleşim alanı olmak üzere dört sınıf belirlenmiştir. Son olarak SVM, ANN, MLC ve KNN yöntemlerinin genel doğrulukları tespit edilmiştir. Nesne tabanlı KNN yönteminin diğer sınıflandırma yöntemlerine göre daha yüksek doğruluk sağlamıştır.
