Ana Paula Marques Ramos
@fct.unesp.br
Ph.D. Assistant Professor in the Department of Cartography
São Paulo State University (Unesp)
Ph.D. in Cartographic Sciences from the São Paulo State University (Unesp). Currently, she is an Assistant Professor at Unesp in the Department of Cartography. Her experience is regarding Geosciences, with an emphasis on Remote Sensing of Vegetation, and spatial analysis. Recently she started to develop applied research by integrating Geomatics (mainly Remote Sensing of Vegetation, and spatial analysis) and Machine Learning (shallow and deep learning) areas into environmental and precision agriculture issues studies. She is a CNPq Research Productivity Scholarship (2021-2024) in the area of Geosciences (PQ level -two).
EDUCATION
2011-04-01 to 2015-04-30 | Ph.D. (Cartographic Sciences Post-Graduation Program);
2009-03-01 to 2011-03-31 | Master (Cartographic Sciences Post-Graduation Program);
2004-03-01 to 2008-12-31 | Graduated Cartographic Engineer (Cartographic Engineering)
RESEARCH INTERESTS
Develops research in the Geomatics area, focusing on Remote Sensing of Vegetation and Cartography. Has been involved in research focused on the application of Machine Learning (shallow and deep algorithms) in Remote Sensing data.
Scopus Publications
Scholar Citations
Scholar h-index
Scholar i10-index
Scopus Publications
Mauricio De Souza, Henrique Lopes Siqueira, Márcio Santos Araujo, Lucas Oliveira, Wesley Nunes Gonçalves, Ana Paula Marques Ramos, and José Marcato Junior
Universidade Federal do Rio de Janeiro
The use of Unmanned aerial vehicles (UAVs) as a tool for image acquisition has been applied in several fields, some applications require cartographic products with high accuracy. With this comes the need for planning the acquisition of images and distribution of control points (GCP) so that digital products meet the required level of accuracy. The aim of this work was to investigate whether the quantity of control points as well as their distribution in different altitude planes in elevated ground can improve the accuracy of the generated cartographic products. RGB images captured by an onboard camera with a resolution of 20 MP were used. Images were captured by a multirotor UAV with an overlap of 80% (front and side) and estimated GSD of 0.017 m. The surveyed area of 5.5 ha overflown area had 31 targets surveyed with GNSS RTK, 21 defined as checkpoints (CP) and 12 as ground control points (GCP), which were used in image processing to generate orthomosaic. We evaluated the accuracy of the generated products based on the PEC-PCD. The results showed that when using only 2 GCPs the altimetric errors are high, being the single configuration that did not fit the PEC-PCD scale 1: 1,000 class A. With 5 GCPs we obtained the best RMSE in altimetry (0.026 m). With 6 GCPs we obtained the best RMSE in planimetry (0.046 m). Altimetry is the most sensitive aspect in generating cartographic products, and the use of GCPs in elevation improves altimetric accuracy.
Diogo Nunes Gonçalves, Jos'e Marcato Junior, Mauro dos Santos de Arruda, Vanessa Jordão Marcato Fernandes, Ana Paula Marques Ramos, Danielle Elis Garcia Furuya, Lucas Prado Osco, Hongjie He, Lucio Andr'e de Castro Jorge, Jonathan Li,et al.
Elsevier BV
José Augusto Correa Martins, Alberto Yoshiriki Hisano Higuti, Aiesca Oliveira Pellegrin, Raquel Soares Juliano, Adriana Mello de Araújo, Luiz Alberto Pellegrin, Veraldo Liesenberg, Ana Paula Marques Ramos, Wesley Nunes Gonçalves, Diego André Sant’Ana,et al.
MDPI AG
Crop segmentation, the process of identifying and delineating agricultural fields or specific crops within an image, plays a crucial role in precision agriculture, enabling farmers and public managers to make informed decisions regarding crop health, yield estimation, and resource allocation in Midwest Brazil. The crops (corn) in this region are being damaged by wild pigs and other diseases. For the quantification of corn fields, this paper applies novel computer-vision techniques and a new dataset of corn imagery composed of 1416 256 × 256 images and corresponding labels. We flew nine drone missions and classified wild pig damage in ten orthomosaics in different stages of growth using semi-automatic digitizing and deep-learning techniques. The period of crop-development analysis will range from early sprouting to the start of the drying phase. The objective of segmentation is to transform or simplify the representation of an image, making it more meaningful and easier to interpret. For the objective class, corn achieved an IoU of 77.92%, and for background 83.25%, using DeepLabV3+ architecture, 78.81% for corn, and 83.73% for background using SegFormer architecture. For the objective class, the accuracy metrics were achieved at 86.88% and for background 91.41% using DeepLabV3+, 88.14% for the objective, and 91.15% for background using SegFormer.
Jeremie Garnier, Rejane E. Cicerelli, Tati de Almeida, Julia C. R. Belo, Julia Curto, Ana Paula M. Ramos, Larissa V. Valadão, Frederic Satge, and Marie-Paule Bonnet
MDPI AG
In the western Andes, climate changes have led to drastic ecological changes during the Pleistocene and Holocene. Given the debate surrounding precipitation pattern changes and the lack of research on lakes in the Chilean Altiplano, this study aims to assess recent climate changes. The paper presents an innovative methodology based on Google Earth Engine (GEE), utilizing fluctuations in water levels in endorheic lakes as natural precipitation indicators. Three lakes (Chungará, Miscanti, and Miniques) in isolated drainage systems were studied, where changes in water levels directly reflect rainfall variations. Data from Landsat-OLI 8, Landsat-ETM+, Landsat-TM 5, and MODIS spanning 31 years were processed using the Google Earth Engine platform. The shapes of the water bodies were extracted using hue saturation value (HSV) composites. The surface areas of the lakes were compared with precipitation data from national meteorological stations and the Tropical Rainfall Measuring Mission (TRMM) using linear regression analyses. Both lake area and rainfall volume showed a decrease over time, with varying trends depending on environmental conditions. However, the analysis consistently indicates a reduction in the area and volume of Chilean lakes corresponding to observed rainfall patterns over the past three decades.
Diogo Nunes Gonçalves, José Marcato Junior, Mauro dos Santos de Arruda, Vanessa Jordão Marcato Fernandes, Ana Paula Marques Ramos, Danielle Elis Garcia Furuya, Lucas Prado Osco, Hongjie He, Lucio André de Castro Jorge, Jonathan Li,et al.
Elsevier BV
Keiller Nogueira, Mayara Maezano Faita-Pinheiro, Ana Paula Marques Ramos, Wesley Nunes Gonçalves, José Marcato Junior, and Jefersson A. Dos Santos
IEEE
Binary segmentation is the main task underpinning several remote sensing applications, which are particularly interested in identifying and monitoring a specific category/object. Although extremely important, such a task has several challenges, including huge intra-class variance for the background and data imbalance. Furthermore, most works tackling this task partially or completely ignore one or both of these challenges and their developments. In this paper, we propose a novel method to perform imbalanced binary segmentation of remote sensing images based on deep networks, prototypes, and contrastive loss. The proposed approach allows the model to focus on learning the foreground class while alleviating the class imbalance problem by allowing it to concentrate on the most difficult background examples. The results demonstrate that the proposed method outperforms state-of-the-art techniques for imbalanced binary segmentation of remote sensing images while taking much less training time.
Lucas Prado Osco, Qiusheng Wu, Eduardo Lopes de Lemos, Wesley Nunes Gonçalves, Ana Paula Marques Ramos, Jonathan Li, and José Marcato
Elsevier BV
Lucas Prado Osco, Eduardo Lopes de Lemos, Wesley Nunes Gonçalves, Ana Paula Marques Ramos, and José Marcato Junior
MDPI AG
Recent advancements in Natural Language Processing (NLP), particularly in Large Language Models (LLMs), associated with deep learning-based computer vision techniques, have shown substantial potential for automating a variety of tasks. These are known as Visual LLMs and one notable model is Visual ChatGPT, which combines ChatGPT’s LLM capabilities with visual computation to enable effective image analysis. These models’ abilities to process images based on textual inputs can revolutionize diverse fields, and while their application in the remote sensing domain remains unexplored, it is important to acknowledge that novel implementations are to be expected. Thus, this is the first paper to examine the potential of Visual ChatGPT, a cutting-edge LLM founded on the GPT architecture, to tackle the aspects of image processing related to the remote sensing domain. Among its current capabilities, Visual ChatGPT can generate textual descriptions of images, perform canny edge and straight line detection, and conduct image segmentation. These offer valuable insights into image content and facilitate the interpretation and extraction of information. By exploring the applicability of these techniques within publicly available datasets of satellite images, we demonstrate the current model’s limitations in dealing with remote sensing images, highlighting its challenges and future prospects. Although still in early development, we believe that the combination of LLMs and visual models holds a significant potential to transform remote sensing image processing, creating accessible and practical application opportunities in the field.
Michelle Taís Garcia Furuya, Danielle Elis Garcia Furuya, Lucas Yuri Dutra de Oliveira, Paulo Antonio da Silva, Rejane Ennes Cicerelli, Wesley Nunes Gonçalves, José Marcato Junior, Lucas Prado Osco, and Ana Paula Marques Ramos
Springer Science and Business Media LLC
Samuel Fernando Schwaida, Rejane Ennes Cicerelli, Tati de Almeida, Edson Eyji Sano, Carlos Henrique Pires, and Ana Paula Marques Ramos
Springer Science and Business Media LLC
Diogo Nunes Gonçalves, José Marcato, André Caceres Carrilho, Plabiany Rodrigo Acosta, Ana Paula Marques Ramos, Felipe David Georges Gomes, Lucas Prado Osco, Maxwell da Rosa Oliveira, José Augusto Correa Martins, Geraldo Alves Damasceno,et al.
Elsevier BV
Bruna Maria Casachi Bernardes de Melo Carapeba, Sérgio Marques Costa, Rogério Giuffrida, Ana Paula Alves Favareto, Ana Paula Marques Ramos, Fabíola de Azevedo Mello, Marcus Vinicius Pimenta Rodrigues, and Renata Calciolari Rossi
Universidade de São Paulo. Agência de Bibliotecas e Coleções Digitais
O objetivo deste trabalho foi analisar a distribuição espaço-temporal dos pacientes com cardiopatias congênitas atendidos no Ambulatório de Cardiologia Pediátrica do Hospital de referência do Oeste Paulista. Realizamos um estudo retrospectivo com análise de dados de base eletrônica e prontuários dos pacientes diagnosticados com cardiopatiacongênita entre os períodos de julho de 2013 a julho de 2018. Foram selecionados 298 prontuários para análise das variáveis de CID-10, gênero, distribuição espacial e série temporal. Foi possível observar que os defeitos septais foram as cardiopatias mais prevalentes, não houve diferença entre os gêneros. Notou-se aumento do diagnóstico a partir de 2014, com implementação do teste do coraçãozinho e 51% dos casos eram da cidade de Presidente Prudente,com maior concentração de casos na região do parque industrial. Há uma relação na incidência das malformações cardíacas com o meio ambiente desfavorável. Os resultados encontrados podem guiar políticas de saúde pública, visando reduzir a exposição da população mais vulnerável, na busca da melhora nos índices de saúde.
Mário de Araújo Carvalho, José Marcato, José Augusto Correa Martins, Pedro Zamboni, Celso Soares Costa, Henrique Lopes Siqueira, Márcio Santos Araújo, Diogo Nunes Gonçalves, Danielle Elis Garcia Furuya, Lucas Prado Osco,et al.
Elsevier BV
Diogo Nunes Gonçalves, Plabiany Rodrigo Acosta, Ana Paula Marques Ramos, Lucas Prado Osco, Danielle Elis Garcia Furuya, Michelle Taís Garcia Furuya, Jonathan Li, José Marcato Junior, Hemerson Pistori, and Wesley Nunes Gonçalves
Elsevier BV
Lucas Prado Osco, Danielle Elis Garcia Furuya, Michelle Taís Garcia Furuya, Daniel Veras Corrêa, Wesley Nunes Gonçalvez, José Marcato Junior, Miguel Borges, Maria Carolina Blassioli-Moraes, Mirian Fernandes Furtado Michereff, Michely Ferreira Santos Aquino,et al.
Elsevier BV
Mauro dos Santos de Arruda, Lucas Prado Osco, Plabiany Rodrigo Acosta, Diogo Nunes Gonçalves, José Marcato Junior, Ana Paula Marques Ramos, Edson Takashi Matsubara, Zhipeng Luo, Jonathan Li, Jonathan de Andrade Silva,et al.
Elsevier BV
Caroline Souza Senkiio, Ana Paula Marques Ramos, Silvio Jorge Coelho Simões, and Tatiana Sussel Gonçalves Mendes
Springer Science and Business Media LLC
Patrik Olã Bressan, José Marcato Junior, José Augusto Correa Martins, Maximilian Jaderson de Melo, Diogo Nunes Gonçalves, Daniel Matte Freitas, Ana Paula Marques Ramos, Michelle Taís Garcia Furuya, Lucas Prado Osco, Jonathan de Andrade Silva,et al.
Elsevier BV
Maximilian Jaderson de Melo, Diogo Nunes Gonçalves, Marina de Nadai Bonin Gomes, Gedson Faria, Jonathan de Andrade Silva, Ana Paula Marques Ramos, Lucas Prado Osco, Michelle Taís Garcia Furuya, José Marcato Junior, and Wesley Nunes Gonçalves
Elsevier BV
Ana Paula Marques Ramos, Felipe David Georges Gomes, Mayara Maezano Faita Pinheiro, Danielle Elis Garcia Furuya, Wesley Nunes Gonçalvez, José Marcato Junior, Mirian Fernandes Furtado Michereff, Maria Carolina Blassioli-Moraes, Miguel Borges, Raúl Alberto Alaumann,et al.
Springer Science and Business Media LLC
Samara Peruzzo Gusman, Ana Paula Marques Ramos, and Alba Regina Azevedo Arana
Universidade Federal de Goias
A produção desordenada do espaço urbano gera consequências negativas no bem-estar e na qualidade de vida dos habitantes, promovendo, danos ambientais que podem der irreversíveis. O artigo objetiva analisar a percepção ambiental no ambiente universitário e entender como a legibilidade do espaço ajuda na percepção ambiental. Trata-se de uma pesquisa aplicada e exploratória, utilizando trabalho de campo de abordagem qualitativa, os dados foram obtidos através das entrevistas e observação participante. Os entrevistados foram divididos em três classes: aluno, professor e colaborador. Eles foram questionados sobre o significado do Campus, elementos distintivos do local, indicações do percurso mais comum feito pelos entrevistados e justificativas das indicações dos elementos distintivos. Predominaram definições positivas nas classificações sobre o significado do Campus, sendo: amplo e arborização/árvores, citadas por 25% dos entrevistados. Os resultados obtidos indicam que os mapas mentais foram instrumentos importantes e eficazes, para identificar a construção do conhecimento espacial e legibilidade por parte dos estudantes, professores e servidores do campus. Palavras-chave: Paisagem. Espaço urbano. Universidade. Areas verdes.
Danielle Elis Garcia Furuya, Lingfei Ma, Mayara Maezano Faita Pinheiro, Felipe David Georges Gomes, Wesley Nunes Gonçalvez, José Marcato Junior, Diego de Castro Rodrigues, Maria Carolina Blassioli-Moraes, Mirian Fernandes Furtado Michereff, Miguel Borges,et al.
Elsevier BV
Luciene Sales Dagher Arce, Lucas Prado Osco, Mauro dos Santos de Arruda, Danielle Elis Garcia Furuya, Ana Paula Marques Ramos, Camila Aoki, Arnildo Pott, Sarah Fatholahi, Jonathan Li, Fábio Fernando de Araújo,et al.
Springer Science and Business Media LLC
AbstractAccurately mapping individual tree species in densely forested environments is crucial to forest inventory. When considering only RGB images, this is a challenging task for many automatic photogrammetry processes. The main reason for that is the spectral similarity between species in RGB scenes, which can be a hindrance for most automatic methods. This paper presents a deep learning-based approach to detect an important multi-use species of palm trees (Mauritia flexuosa; i.e., Buriti) on aerial RGB imagery. In South-America, this palm tree is essential for many indigenous and local communities because of its characteristics. The species is also a valuable indicator of water resources, which comes as a benefit for mapping its location. The method is based on a Convolutional Neural Network (CNN) to identify and geolocate singular tree species in a high-complexity forest environment. The results returned a mean absolute error (MAE) of 0.75 trees and an F1-measure of 86.9%. These results are better than Faster R-CNN and RetinaNet methods considering equal experiment conditions. In conclusion, the method presented is efficient to deal with a high-density forest scenario and can accurately map the location of single species like the M. flexuosa palm tree and may be useful for future frameworks.
Paulo Eduardo Teodoro, Larissa Pereira Ribeiro Teodoro, Fábio Henrique Rojo Baio, Carlos Antonio da Silva Junior, Regimar Garcia dos Santos, Ana Paula Marques Ramos, Mayara Maezano Faita Pinheiro, Lucas Prado Osco, Wesley Nunes Gonçalves, Alexsandro Monteiro Carneiro,et al.
MDPI AG
In soybean, there is a lack of research aiming to compare the performance of machine learning (ML) and deep learning (DL) methods to predict more than one agronomic variable, such as days to maturity (DM), plant height (PH), and grain yield (GY). As these variables are important to developing an overall precision farming model, we propose a machine learning approach to predict DM, PH, and GY for soybean cultivars based on multispectral bands. The field experiment considered 524 genotypes of soybeans in the 2017/2018 and 2018/2019 growing seasons and a multitemporal–multispectral dataset collected by embedded sensor in an unmanned aerial vehicle (UAV). We proposed a multilayer deep learning regression network, trained during 2000 epochs using an adaptive subgradient method, a random Gaussian initialization, and a 50% dropout in the first hidden layer for regularization. Three different scenarios, including only spectral bands, only vegetation indices, and spectral bands plus vegetation indices, were adopted to infer each variable (PH, DM, and GY). The DL model performance was compared against shallow learning methods such as random forest (RF), support vector machine (SVM), and linear regression (LR). The results indicate that our approach has the potential to predict soybean-related variables using multispectral bands only. Both DL and RF models presented a strong (r surpassing 0.77) prediction capacity for the PH variable, regardless of the adopted input variables group. Our results demonstrated that the DL model (r = 0.66) was superior to predict DM when the input variable was the spectral bands. For GY, all machine learning models evaluated presented similar performance (r ranging from 0.42 to 0.44) for each tested scenario. In conclusion, this study demonstrated an efficient approach to a computational solution capable of predicting multiple important soybean crop variables based on remote sensing data. Future research could benefit from the information presented here and be implemented in subsequent processes related to soybean cultivars or other types of agronomic crops.
Diego Bedin Marin, Gabriel Araújo e Silva Ferraz, Lucas Santos Santana, Brenon Diennevan Souza Barbosa, Rafael Alexandre Pena Barata, Lucas Prado Osco, Ana Paula Marques Ramos, and Paulo Henrique Sales Guimarães
Elsevier BV
Abstract Coffee leaf rust (CLR) is one of the most devastating leaf diseases in coffee plantations. By knowing the symptoms, severity, and spatial distribution of CLR, farmers can improve disease management procedures and reduce losses associated with it. Recently, Unmanned Aerial Vehicles (UAVs)-based images, in conjunction with machine learning (ML) techniques, helped solve multiple agriculture-related problems. In this sense, vegetation indices processed with ML algorithms are a promising strategy. It is still a challenge to map severity levels of CLR using remote sensing data and an ML approach. Here we propose a framework to detect CLR severity with only vegetation indices extracted from UAV imagery. For that, we based our approach on decision tree models, as they demonstrated important results in related works. We evaluated a coffee field with different infestation classes of CLR: class 1 (from 2% to 5% rust); class 2 (from 5% to 10% rust); class 3 (from 10% to 20% rust), and; class 4 (from 20% to 40% rust). We acquired data with a Sequoia camera, producing images with a spatial resolution of 10.6 cm, in four spectral bands: green (530–570 nm), red (640–680 nm), red-edge (730–740 nm), and near-infrared (770–810 nm). A total of 63 vegetation indices was extracted from the images, and the following learners were evaluated in a cross-validation method with 10 folders: Logistic Model Tree (LMT); J48; ExtraTree; REPTree; Functional Trees (FT); Random Tree (RT), and; Random Forest (RF). The results indicated that the LMT method contributed the most to the accurate prediction of early and several infestation classes. For these classes, LMT returned F-measure values of 0.915 and 0.875, thus being a good indicator of early CLR (2 to 5% of rust) and later stages of CLR (20 to 40% of rust). We demonstrated a valid approach to model rust in coffee plants using only vegetation indices and ML algorithms, specifically for the disease's early and later stages. We concluded that the proposed framework allows inferring the predicted classes in remaining plants within the sampled area, thus helping the identification of potential CLR in non-sampled plants. We corroborate that the decision tree-based model may assist in precision agriculture practices, including mapping rust in coffee plantations, providing both an efficient non-invasive and spatially continuous monitoring of the disease.