Pub Date : 2021-08-18DOI: 10.1590/s1982-21702021000s00020
Rovane Marcos de França, I. Klein, L. Veiga
The densification of geodetic surveys using classical positioning techniques such as total stations may be necessary due to the quality of Global Navigation Satellite System (GNSS) positioning in urban canyons. However, the correction of distances and angles due to the deflection of the vertical (DV) is usually neglected in commercial softwares and internal software of total stations. Given that context, this research seeks to estimate the influence of DV on the horizontal geodetic positioning with total station in the Brazilian territory. Secondarily, it seeks to demonstrate the practical application of DV in the densification of geodetic networks. It is important to note that land surveys in Brazil must be connected to a geodetic network; therefore, the neglect of DV may degrade the positional quality of geodetic surveys. Results obtained indicate differences in horizontal geodetic positions of up to 45 ppm. Considering the desired positional quality of the geodetic network, such values demonstrate the importance of a proper correction for the DV.
{"title":"THE INFLUENCE OF THE DEFLECTION OF THE VERTICAL ON GEODETIC SURVEYS IN BRAZIL","authors":"Rovane Marcos de França, I. Klein, L. Veiga","doi":"10.1590/s1982-21702021000s00020","DOIUrl":"https://doi.org/10.1590/s1982-21702021000s00020","url":null,"abstract":"The densification of geodetic surveys using classical positioning techniques such as total stations may be necessary due to the quality of Global Navigation Satellite System (GNSS) positioning in urban canyons. However, the correction of distances and angles due to the deflection of the vertical (DV) is usually neglected in commercial softwares and internal software of total stations. Given that context, this research seeks to estimate the influence of DV on the horizontal geodetic positioning with total station in the Brazilian territory. Secondarily, it seeks to demonstrate the practical application of DV in the densification of geodetic networks. It is important to note that land surveys in Brazil must be connected to a geodetic network; therefore, the neglect of DV may degrade the positional quality of geodetic surveys. Results obtained indicate differences in horizontal geodetic positions of up to 45 ppm. Considering the desired positional quality of the geodetic network, such values demonstrate the importance of a proper correction for the DV.","PeriodicalId":55347,"journal":{"name":"Boletim De Ciencias Geodesicas","volume":null,"pages":null},"PeriodicalIF":0.5,"publicationDate":"2021-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43207791","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-08-17DOI: 10.1590/s1982-21702021000300016
Danielle Silva de Paula, J. O. Ortiz, S. Rosim, L. Namikawa
The Permanent Protection Areas (PPA) are relevant to ensure vegetation around the drainage network. This paper presents an automated methodology for the extraction of drainage from the river and automated generation of PPA, and analysis of environmental adequacy. The methodology is based on geoprocessing and remote sensing techniques applied to RapidEye satellite images. The analyzed area covers a portion of the Paraiba do Sul river basin, located in the city of Sao Jose dos Campos (Southern Brazil). Land use and land cover were determined using a digital classifier and estimated within the APP of four rural properties bordering the river. The digital classification of the RapidEye images was evaluated based on the visual interpretation of high spatial resolution airborne orthophotos, as well as through random points that enabled the generation of the Kappa index and global accuracy, showing high agreement. The analysis shows the inadequate land use practice in some properties analyzed, indicating changes in the areas of PPA over the years analyzed. The results of this research show that the proposed methodology can be used for supervision purposes in properties declared in the Rural Environmental Registry (CAR), thus assisting in the decision-making process.
{"title":"A METHODOLOGY FOR DETERMINING AND ANALYZING PERMANENT PROTECTION AREAS OF PROPERTIES DECLARED IN THE RURAL ENVIRONMENTAL REGISTER- CAR","authors":"Danielle Silva de Paula, J. O. Ortiz, S. Rosim, L. Namikawa","doi":"10.1590/s1982-21702021000300016","DOIUrl":"https://doi.org/10.1590/s1982-21702021000300016","url":null,"abstract":"The Permanent Protection Areas (PPA) are relevant to ensure vegetation around the drainage network. This paper presents an automated methodology for the extraction of drainage from the river and automated generation of PPA, and analysis of environmental adequacy. The methodology is based on geoprocessing and remote sensing techniques applied to RapidEye satellite images. The analyzed area covers a portion of the Paraiba do Sul river basin, located in the city of Sao Jose dos Campos (Southern Brazil). Land use and land cover were determined using a digital classifier and estimated within the APP of four rural properties bordering the river. The digital classification of the RapidEye images was evaluated based on the visual interpretation of high spatial resolution airborne orthophotos, as well as through random points that enabled the generation of the Kappa index and global accuracy, showing high agreement. The analysis shows the inadequate land use practice in some properties analyzed, indicating changes in the areas of PPA over the years analyzed. The results of this research show that the proposed methodology can be used for supervision purposes in properties declared in the Rural Environmental Registry (CAR), thus assisting in the decision-making process.","PeriodicalId":55347,"journal":{"name":"Boletim De Ciencias Geodesicas","volume":null,"pages":null},"PeriodicalIF":0.5,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46470722","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-08-17DOI: 10.1590/1982-2170-2020-0061
Valéria Cristina Silva, Flavio Guilherme Vaz de Almeida Filho, D. Blitzkow, A. Matos
Abstract The combination of physical and geometric heights, required for geodetic purposes, uses Global Geopotential Models (GGMs), local geoid, or quasigeoid models. The geoid height and the height anomaly, provided by GGMs, are not accurate enough for most engineering applications. Considering the normal height system of Brazil and the physical concepts of the involved reference surfaces, a quasigeoid model is more appropriate than the current Brazilian geoid model MAPGEO2015. This paper shows the determination of the geoid and the quasigeoid models for São Paulo state using the updated gravimetric data and the new system of the normal height of the 2018 Brazilian Vertical Reference Frame (BVRF). The computation of the quasigeoid model was performed by numerical integration through the Fast Fourier Transform (FFT). The Molodensky gravity anomaly was determined in a 5’ grid and reduced and restored using the Residual Terrain Model (RTM) technique and the XGM2019e GGM truncated at degree and order 250 and 720. The geoid model was derived from the Bouguer gravity anomalies. The quasigeoid model validation has shown a Root Mean Square (RMS) difference of 18 cm compared with the Global Positioning System (GPS) measurements in the levelling network.
{"title":"THE GEOID AND QUASIGEOID OF SÃO PAULO STATE USING THE UPDATED GRAVIMETRIC DATA AND THE 2018 BVRF","authors":"Valéria Cristina Silva, Flavio Guilherme Vaz de Almeida Filho, D. Blitzkow, A. Matos","doi":"10.1590/1982-2170-2020-0061","DOIUrl":"https://doi.org/10.1590/1982-2170-2020-0061","url":null,"abstract":"Abstract The combination of physical and geometric heights, required for geodetic purposes, uses Global Geopotential Models (GGMs), local geoid, or quasigeoid models. The geoid height and the height anomaly, provided by GGMs, are not accurate enough for most engineering applications. Considering the normal height system of Brazil and the physical concepts of the involved reference surfaces, a quasigeoid model is more appropriate than the current Brazilian geoid model MAPGEO2015. This paper shows the determination of the geoid and the quasigeoid models for São Paulo state using the updated gravimetric data and the new system of the normal height of the 2018 Brazilian Vertical Reference Frame (BVRF). The computation of the quasigeoid model was performed by numerical integration through the Fast Fourier Transform (FFT). The Molodensky gravity anomaly was determined in a 5’ grid and reduced and restored using the Residual Terrain Model (RTM) technique and the XGM2019e GGM truncated at degree and order 250 and 720. The geoid model was derived from the Bouguer gravity anomalies. The quasigeoid model validation has shown a Root Mean Square (RMS) difference of 18 cm compared with the Global Positioning System (GPS) measurements in the levelling network.","PeriodicalId":55347,"journal":{"name":"Boletim De Ciencias Geodesicas","volume":null,"pages":null},"PeriodicalIF":0.5,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45553298","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-08-17DOI: 10.1590/1982-2170-2020-0057
L. L. S. França, A. Seixas, Luciene Ferreira Gama, João Naves de Moraes
Abstract: The forward intersection method is already widely used in the geodetic survey of coordinates of inaccessible points, especially when only angle measurements are available, in this case, also called the triangulation method. However, the mathematical solution of the 3D forward intersection with the analytical definition of spatial lines, resolved by the Minimum Distances Method, is still not widespread in the academic and professional environment. This mathematical modeling determines the 3D coordinates of a point located in the middle of the minimum distance between two or more spatial lines, which spatially "intersect" towards the observation point. This solution is more accurate than others presented in the literature because it simultaneously solves the problem of 3D determination of a point by the method of least squares, in addition to providing an estimate of the coordinate precision, which are inherent to the adjustment. This work, therefore, has the objective of explaining the Minimum Distances Method for the spatial intersection of targeted measurements with a Total Station from two or more known observation points for the 3D determination of inaccessible points located in corners of buildings. For the analysis of the method, a Python tool was developed for QGIS that calculates the 3D coordinates and generates the adjustment processing report, being applied with real observations of the Geodetic survey of the SUDENE building, in Recife-PE. The methodology developed in this work proved to be suitable for measurements of large structures, achieving spherical precision better than ±1.0 cm, following the Brazilian standards for urban cadastre.
{"title":"OPTIMIZED DETERMINATION OF 3D COORDINATES IN THE SURVEY OF INACCESSIBLE POINTS OF BUILDINGS - EXAMPLE OF APPLICATION IMPLEMENTED IN FREE SOFTWARE","authors":"L. L. S. França, A. Seixas, Luciene Ferreira Gama, João Naves de Moraes","doi":"10.1590/1982-2170-2020-0057","DOIUrl":"https://doi.org/10.1590/1982-2170-2020-0057","url":null,"abstract":"Abstract: The forward intersection method is already widely used in the geodetic survey of coordinates of inaccessible points, especially when only angle measurements are available, in this case, also called the triangulation method. However, the mathematical solution of the 3D forward intersection with the analytical definition of spatial lines, resolved by the Minimum Distances Method, is still not widespread in the academic and professional environment. This mathematical modeling determines the 3D coordinates of a point located in the middle of the minimum distance between two or more spatial lines, which spatially \"intersect\" towards the observation point. This solution is more accurate than others presented in the literature because it simultaneously solves the problem of 3D determination of a point by the method of least squares, in addition to providing an estimate of the coordinate precision, which are inherent to the adjustment. This work, therefore, has the objective of explaining the Minimum Distances Method for the spatial intersection of targeted measurements with a Total Station from two or more known observation points for the 3D determination of inaccessible points located in corners of buildings. For the analysis of the method, a Python tool was developed for QGIS that calculates the 3D coordinates and generates the adjustment processing report, being applied with real observations of the Geodetic survey of the SUDENE building, in Recife-PE. The methodology developed in this work proved to be suitable for measurements of large structures, achieving spherical precision better than ±1.0 cm, following the Brazilian standards for urban cadastre.","PeriodicalId":55347,"journal":{"name":"Boletim De Ciencias Geodesicas","volume":null,"pages":null},"PeriodicalIF":0.5,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49099964","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-08-17DOI: 10.1590/1982-2170-2019-0060
João Victor Pacheco Gomes, L. Delazari, M. Schmidt
The words "environment" and "space" demonstrate distinct spatial units. It must be questioned whether the internal space, seen as an analytical subcategory of space, adds specificities of this type of designation. Therefore, if indoor is a subcategory of space, then its characteristics and types of representation must be observed and analyzed considering aspects of space. The purpose of this article is to present the characteristics of the indoor space unit as a subcategory of space. The “space” terminology applied to specify the indoor spatial unit has some features of spatial analysis that allow a broader and deeper spectrum as an object of study. Compared to space, the "environment" proves to be limited to represent the characteristics of the indoor. The intern must be understood as a space within a space, inserting a subcategory of the urban space, however, it is never seen as in its entirety. The totality does not observe space as it is, but everything within it. Space, as a creation of man, allows the creation of subspaces with no connection to the outside, in the category called indoor contributing to the analysis procedures based on the understanding of their relationships.
{"title":"THE INDOOR SPACE AS A DISTINCT ENVIRONMENTAL CATEGORY FOR SPATIAL ANALYSIS","authors":"João Victor Pacheco Gomes, L. Delazari, M. Schmidt","doi":"10.1590/1982-2170-2019-0060","DOIUrl":"https://doi.org/10.1590/1982-2170-2019-0060","url":null,"abstract":"The words \"environment\" and \"space\" demonstrate distinct spatial units. It must be questioned whether the internal space, seen as an analytical subcategory of space, adds specificities of this type of designation. Therefore, if indoor is a subcategory of space, then its characteristics and types of representation must be observed and analyzed considering aspects of space. The purpose of this article is to present the characteristics of the indoor space unit as a subcategory of space. The “space” terminology applied to specify the indoor spatial unit has some features of spatial analysis that allow a broader and deeper spectrum as an object of study. Compared to space, the \"environment\" proves to be limited to represent the characteristics of the indoor. The intern must be understood as a space within a space, inserting a subcategory of the urban space, however, it is never seen as in its entirety. The totality does not observe space as it is, but everything within it. Space, as a creation of man, allows the creation of subspaces with no connection to the outside, in the category called indoor contributing to the analysis procedures based on the understanding of their relationships.","PeriodicalId":55347,"journal":{"name":"Boletim De Ciencias Geodesicas","volume":null,"pages":null},"PeriodicalIF":0.5,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44163466","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-08-17DOI: 10.1590/s1982-21702021000200015
M. V. Y. Garcia, H. C. Oliveira
Abstract: Technological improvement in sensors and the use of computer vision algorithms made possible the generation of high accuracy mapping products (cm level) using data acquired by low-cost Unmanned Aerial Vehicles (UAV). However, the procedure to optimally set the aerial block configuration is not well understood for some users mainly due to the popularization of the UAV and its use by non-specialists. This study aims to contribute to this aspect, investigating and highlighting the influence of flight parameters, camera calibration and number of Ground Control Points (GCP) on generating digital terrain models and orthomosaic. To address this issue, several field experiments and data processing were carried out. The quality was assessed by calculating the Root Mean Square Error (RMSE) together with a bias evaluation (t-Student test at 90% confidence level). The results suggest that an optimum block configuration for accurate and unbiased products is achieved by surveying at rates of 80%/60% (forward and sidelap, respectively), with an average Ground Sample Distance (GSD) of around 1 cm at a flight height of 31 m, using a pre-calibrated camera and 5 GCP at least.
{"title":"THE INFLUENCE OF FLIGHT CONFIGURATION, CAMERA CALIBRATION, AND GROUND CONTROL POINTS FOR DIGITAL TERRAIN MODEL AND ORTHOMOSAIC GENERATION USING UNMANNED AERIAL VEHICLES IMAGERY","authors":"M. V. Y. Garcia, H. C. Oliveira","doi":"10.1590/s1982-21702021000200015","DOIUrl":"https://doi.org/10.1590/s1982-21702021000200015","url":null,"abstract":"Abstract: Technological improvement in sensors and the use of computer vision algorithms made possible the generation of high accuracy mapping products (cm level) using data acquired by low-cost Unmanned Aerial Vehicles (UAV). However, the procedure to optimally set the aerial block configuration is not well understood for some users mainly due to the popularization of the UAV and its use by non-specialists. This study aims to contribute to this aspect, investigating and highlighting the influence of flight parameters, camera calibration and number of Ground Control Points (GCP) on generating digital terrain models and orthomosaic. To address this issue, several field experiments and data processing were carried out. The quality was assessed by calculating the Root Mean Square Error (RMSE) together with a bias evaluation (t-Student test at 90% confidence level). The results suggest that an optimum block configuration for accurate and unbiased products is achieved by surveying at rates of 80%/60% (forward and sidelap, respectively), with an average Ground Sample Distance (GSD) of around 1 cm at a flight height of 31 m, using a pre-calibrated camera and 5 GCP at least.","PeriodicalId":55347,"journal":{"name":"Boletim De Ciencias Geodesicas","volume":null,"pages":null},"PeriodicalIF":0.5,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43480971","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-08-17DOI: 10.1590/s1982-21702021000300017
C. Amisse, M. E. Jijón-Palma, J. Centeno
Abstract: The wide use of cameras enables the availability of a large amount of image frames that can be used for people counting or to monitor crowds or single individuals for security purposes. These applications require both, object detection and tracking. This task has shown to be challenging due to problems such as occlusion, deformation, motion blur, and scale variation. One alternative to perform tracking is based on the comparison of features extracted for the individual objects from the image. For this purpose, it is necessary to identify the object of interest, a human image, from the rest of the scene. This paper introduces a method to perform the separation of human bodies from images with changing backgrounds. The method is based on image segmentation, the analysis of the possible pose, and a final refinement step based on probabilistic relaxation. It is the first work we are aware that probabilistic fields computed from human pose figures are combined with an improvement step of relaxation for pedestrian segmentation. The proposed method is evaluated using different image series and the results show that it can work efficiently, but it is dependent on some parameters to be set according to the image contrast and scale. Tests show accuracies above 71%. The method performs well in other datasets, where it achieves results comparable to state-of-the-art approaches.
{"title":"PEDESTRIAN SEGMENTATION FROM COMPLEX BACKGROUND BASED ON PREDEFINED POSE FIELDS AND PROBABILISTIC RELAXATION","authors":"C. Amisse, M. E. Jijón-Palma, J. Centeno","doi":"10.1590/s1982-21702021000300017","DOIUrl":"https://doi.org/10.1590/s1982-21702021000300017","url":null,"abstract":"Abstract: The wide use of cameras enables the availability of a large amount of image frames that can be used for people counting or to monitor crowds or single individuals for security purposes. These applications require both, object detection and tracking. This task has shown to be challenging due to problems such as occlusion, deformation, motion blur, and scale variation. One alternative to perform tracking is based on the comparison of features extracted for the individual objects from the image. For this purpose, it is necessary to identify the object of interest, a human image, from the rest of the scene. This paper introduces a method to perform the separation of human bodies from images with changing backgrounds. The method is based on image segmentation, the analysis of the possible pose, and a final refinement step based on probabilistic relaxation. It is the first work we are aware that probabilistic fields computed from human pose figures are combined with an improvement step of relaxation for pedestrian segmentation. The proposed method is evaluated using different image series and the results show that it can work efficiently, but it is dependent on some parameters to be set according to the image contrast and scale. Tests show accuracies above 71%. The method performs well in other datasets, where it achieves results comparable to state-of-the-art approaches.","PeriodicalId":55347,"journal":{"name":"Boletim De Ciencias Geodesicas","volume":null,"pages":null},"PeriodicalIF":0.5,"publicationDate":"2021-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44307307","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-08-13DOI: 10.1590/s1982-21702021000300022
M. Martins, E. Mitishita
Abstract: The characteristics of data points obtained by laser scanning (LiDAR) and images have been considered complementary in the field of photogrammetric applications, and research to improve their integrated use have recently intensified. This study aim to verify the performance of determining punctual entities in a LiDAR point cloud using linear regression and intersecting lines obtained from buildings with square rooftop containing four planes (hip roof), as well as compare punctual entities three-dimensional coordinates determined by planes intersection. Our results show that the proposed method was more accurate in determining three-dimensional coordinates than plan intersection method. The obtained coordinates were evaluated and framed into the map accuracy standard for digital cartographic products (PEC-PCD), besides being analyzed for trend and precision. Accuracy analysis results frame punctual entities three-dimensional coordinates into the 1/2,000 or lower scale for Class A of PEC-PCD.
{"title":"LINEAR REGRESSION AND LINES INTERSECTING AS A METHOD OF EXTRACTING PUNCTUAL ENTITIES IN A LIDAR POINT CLOUD","authors":"M. Martins, E. Mitishita","doi":"10.1590/s1982-21702021000300022","DOIUrl":"https://doi.org/10.1590/s1982-21702021000300022","url":null,"abstract":"Abstract: The characteristics of data points obtained by laser scanning (LiDAR) and images have been considered complementary in the field of photogrammetric applications, and research to improve their integrated use have recently intensified. This study aim to verify the performance of determining punctual entities in a LiDAR point cloud using linear regression and intersecting lines obtained from buildings with square rooftop containing four planes (hip roof), as well as compare punctual entities three-dimensional coordinates determined by planes intersection. Our results show that the proposed method was more accurate in determining three-dimensional coordinates than plan intersection method. The obtained coordinates were evaluated and framed into the map accuracy standard for digital cartographic products (PEC-PCD), besides being analyzed for trend and precision. Accuracy analysis results frame punctual entities three-dimensional coordinates into the 1/2,000 or lower scale for Class A of PEC-PCD.","PeriodicalId":55347,"journal":{"name":"Boletim De Ciencias Geodesicas","volume":null,"pages":null},"PeriodicalIF":0.5,"publicationDate":"2021-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41501578","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-08-13DOI: 10.1590/s1982-21702021000300018
Murilo Schramm da Silva, A. Vibrans, Adilson Luiz Nicoletti
Abstract: A challenge for the use of medium spatial resolution imagery for land use change detection consists of the reduced availability of ground reference data for previous dates. This study aims to obtain invariant training points using the backdating process for supervised classification of images that have no field data available. The study area comprises 1,353 km² in Santa Catarina, southern Brazil. We compared the accuracy performance of invariant area sets (binary change maps) generated by using three methods (IR-MAD - Iteratively Reweighted Multivariate Alteration Detection, CVA - Change Vector Analysis and SGD - Spectral Gradient Difference) for two periods (2017-2011 and 2011-2006). The classification of the Landsat-5 TM image of 2006 was performed using as training data the sets of points indicated as invariant in the binary maps resulted from the three abovementioned methods. The accuracies for seven land-use classes were computed. The overall accuracy was greater (80,5% and 80,2%) when using training areas achieved by CVA and SGD, respectively than IR-MAD (76%). Were obtained accuracies greater than 80% for the forest class. The results stress that the combination of the IR-MAD and SGD is preferable since the CVA is more time consuming due to the subjective application of thresholds.
{"title":"BACKDATING OF INVARIANT PIXELS: COMPARISON OF ALGORITHMS FOR LAND USE AND LAND COVER CHANGE (LUCC) DETECTION IN THE SUBTROPICAL BRAZILIAN ATLANTIC FOREST","authors":"Murilo Schramm da Silva, A. Vibrans, Adilson Luiz Nicoletti","doi":"10.1590/s1982-21702021000300018","DOIUrl":"https://doi.org/10.1590/s1982-21702021000300018","url":null,"abstract":"Abstract: A challenge for the use of medium spatial resolution imagery for land use change detection consists of the reduced availability of ground reference data for previous dates. This study aims to obtain invariant training points using the backdating process for supervised classification of images that have no field data available. The study area comprises 1,353 km² in Santa Catarina, southern Brazil. We compared the accuracy performance of invariant area sets (binary change maps) generated by using three methods (IR-MAD - Iteratively Reweighted Multivariate Alteration Detection, CVA - Change Vector Analysis and SGD - Spectral Gradient Difference) for two periods (2017-2011 and 2011-2006). The classification of the Landsat-5 TM image of 2006 was performed using as training data the sets of points indicated as invariant in the binary maps resulted from the three abovementioned methods. The accuracies for seven land-use classes were computed. The overall accuracy was greater (80,5% and 80,2%) when using training areas achieved by CVA and SGD, respectively than IR-MAD (76%). Were obtained accuracies greater than 80% for the forest class. The results stress that the combination of the IR-MAD and SGD is preferable since the CVA is more time consuming due to the subjective application of thresholds.","PeriodicalId":55347,"journal":{"name":"Boletim De Ciencias Geodesicas","volume":null,"pages":null},"PeriodicalIF":0.5,"publicationDate":"2021-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48659193","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-06-04DOI: 10.1590/S1982-21702021000200013
C. Amisse, M. E. Jijón-Palma, J. Centeno
Abstract: Object detection in high resolution images is a new challenge that the remote sensing community is facing thanks to introduction of unmanned aerial vehicles and monitoring cameras. One of the interests is to detect and trace persons in the images. Different from general objects, pedestrians can have different poses and are undergoing constant morphological changes while moving, this task needs an intelligent solution. Fine-tuning has woken up great interest among researchers due to its relevance for retraining convolutional networks for many and interesting applications. For object classification, detection, and segmentation fine-tuned models have shown state-of-the-art performance. In the present work, we evaluate the performance of fine-tuned models with a variation of training data by comparing Faster Region-based Convolutional Neural Network (Faster R-CNN) Inception v2, Single Shot MultiBox Detector (SSD) Inception v2, and SSD Mobilenet v2. To achieve the goal, the effect of varying training data on performance metrics such as accuracy, precision, F1-score, and recall are taken into account. After testing the detectors, it was identified that the precision and recall are more sensitive on the variation of the amount of training data. Under five variation of the amount of training data, we observe that the proportion of 60%-80% consistently achieve highly comparable performance, whereas in all variation of training data Faster R-CNN Inception v2 outperforms SSD Inception v2 and SSD Mobilenet v2 in evaluated metrics, but the SSD converges relatively quickly during the training phase. Overall, partitioning 80% of total data for fine-tuning trained models produces efficient detectors even with only 700 data samples.
{"title":"FINE-TUNING DEEP LEARNING MODELS FOR PEDESTRIAN DETECTION","authors":"C. Amisse, M. E. Jijón-Palma, J. Centeno","doi":"10.1590/S1982-21702021000200013","DOIUrl":"https://doi.org/10.1590/S1982-21702021000200013","url":null,"abstract":"Abstract: Object detection in high resolution images is a new challenge that the remote sensing community is facing thanks to introduction of unmanned aerial vehicles and monitoring cameras. One of the interests is to detect and trace persons in the images. Different from general objects, pedestrians can have different poses and are undergoing constant morphological changes while moving, this task needs an intelligent solution. Fine-tuning has woken up great interest among researchers due to its relevance for retraining convolutional networks for many and interesting applications. For object classification, detection, and segmentation fine-tuned models have shown state-of-the-art performance. In the present work, we evaluate the performance of fine-tuned models with a variation of training data by comparing Faster Region-based Convolutional Neural Network (Faster R-CNN) Inception v2, Single Shot MultiBox Detector (SSD) Inception v2, and SSD Mobilenet v2. To achieve the goal, the effect of varying training data on performance metrics such as accuracy, precision, F1-score, and recall are taken into account. After testing the detectors, it was identified that the precision and recall are more sensitive on the variation of the amount of training data. Under five variation of the amount of training data, we observe that the proportion of 60%-80% consistently achieve highly comparable performance, whereas in all variation of training data Faster R-CNN Inception v2 outperforms SSD Inception v2 and SSD Mobilenet v2 in evaluated metrics, but the SSD converges relatively quickly during the training phase. Overall, partitioning 80% of total data for fine-tuning trained models produces efficient detectors even with only 700 data samples.","PeriodicalId":55347,"journal":{"name":"Boletim De Ciencias Geodesicas","volume":null,"pages":null},"PeriodicalIF":0.5,"publicationDate":"2021-06-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"85628914","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：