This research aimed to map land-use in Riau Province using a landscape ecological approach and to assess the level of landscape fragmentation using fragmentation index, edge density, and landscape metrics at observation window sizes of 3×3, 5×5, and 7×7 pixels. Sentinel-2A imagery was the primary input for land-cover mapping, while SRTM DEM was used as the input for terrain unit mapping. The land-cover and terrain unit maps were overlaid to create a land-use map. This map was then used as an input for spatial analysis of fragmentation by computing the fragmentation index and edge density. Another analysis that has been carried out was landscape metrics calculation. The results of the land-use mapping revealed that the study area is predominantly occupied by plantations and forests, with an overall accuracy of 81.88%. The landscape fragmentation analysis showed that areas with high fragmentation level are scattered in the central part of the region, characterized by dense human activity in heterogeneous land-use types. Meanwhile, low fragmentation levels found in homogeneous land-use areas such as natural and semi-natural forests.
Remote sensing data has been proven capable and efficient as a powerful resource for large-scale land cover mapping. However, a map is considered acceptable with the required accuracy value. The problem related to sampling is how the sample amount and sample technique affect the accuracy of the land cover mapping. Furthermore, the accuracy assessment for mapping usually only utilizes accuracy measurement standards, which are commonly used. This research was conducted to measure the effect of the different sampling sizes and sampling methods on the accuracy value of largescale land cover mapping using area based assessment approach. A visual interpretation was used as a reference while multispectral classification was carried out independently as an object to be tested for accuracy assessment. The number of classes interpreted was 25 and 9. We demonstrated the sampling methods applied were random sampling, stratified random sampling, and systematic grid sampling. A confusion matrix method was used to gain the overall accuracy. The result of this study showed that the number of 200 samples for land cover with 25 classes and 36 sample for nine classes could start the regularity against the actual accuracy. While the sample number below 200 and 36 for both land cover classes showed irregular fluctuations in the accuracy value. Using stratified random sampling was satisfactory for modeling the accuracy compared to random and systematic grid sampling. Thus, those results could be used to indicate accuracy value against different scenarios and gain a recommendation for assessing the accuracy of land cover on a large scale.
Gully erosion is the most destructive type of soil erosion, induced by soil detachment. As a result, modest to massive incisions are made in the field. The process can degrade the quantity and quality of soil and potentially cause structural damage. Field studies are used to map the position of gullies, but they are inefficient in terms of time and cost, especially on a regional scale. Therefore, another approach is applied to visualize the probability of gully erosion development using geoenvironmental factors. Remote sensing data can be used to examine the condition of the land, leading to an accurate representation of the earth's surface. This research's primary goal is to predict the location of gully erosion using remote sensing data in the upper section of the Sapi Watershed, Banjarnegara, Indonesia. This location primarily consists of mountainous areas used for massive cultivation. Parameters comprising land use and vegetation area derived from SENTINEL 2A, and topographic and hydrological data from DEMNAS. The mapping process considers the actual location of the gully and other geographical characteristics using Random Forest. A total of 85 gully location records were collected and verified using Google Earth and field surveys. Nongully data were obtained using median filters to distinguish between river and mountain top. 70% of the data is used for modelling and the rest for validation of model results. RF-generated prediction maps could provide an essential instrument for planning and land conservation in the early phases of gully formation.
This study aimed to evaluate the effects of atmospheric and topographic corrections on the vegetation density estimates based on vegetation index transformation. The research was conducted in Arjuno-Welirang volcanic complex, East Java, using Landsat 8 OLI imagery at 30 m spatial resolution. The image was corrected at two levels, i.e., atmospheric correction to at-surface reflectance using FLAASH method, and topographic correction using SCS-C method. The topographic correction referred to ALOS PALSAR DSM data, which was resampled at 30 m pixel size. The vegetation indices used includes NDVI, SAVI, ARVI, EVI and MSARVI. Fieldwork for measuring vegetation density was carried out by vertical bottom-up photography of the canopy on each sample, supported by observations of vegetation density using high-spatial resolution Google Earth imagery. The results showed that—in comparison with the atmospheric correction—the topographic correction was able to increase the correlation coefficients between the spectral information and the measured vegetation density in the field, especially for SAVI, EVI and MSARVI transformations. On the other hand, the NDVI and ARVI showed slight decreases. Based on the vegetation density maps generated using regression equations, the SAVI, EVI and MSARVI showed slight increases from atmospheric to topographic corrections, while the NDVI and ARVI showed declines. The rugged terrain condition affected the accuracies of the models due to the difficulty of vegetation density measurement in the field and even distribution of the samples.
Pixel-based classification is considered as a classic method of extracting land-cover related information from remotely sensed imagery, and has been used in various applications, including vegetation mapping. However, several recent studies also mentioned the weakness of the pixel-based approach, including the vegetation index transformation, in mapping the structural composition of vegetation. This study aimed to test several pixel-based classification algorithms for mapping the structural composition of vegetation using Sentinel-2A (10 meters) imagery in Salatiga and its surrounding, Central Java. In this study three classification algorithms, namely Maximum Likelihood, Minimum Distance to Mean, and Support Vector Machine were compared with respect to their accuracy results in mapping the vegetation structural composition. The authors evaluated the effects of additional data in the classification process by comparing two different datasets, i.e. (i) the one using original bands only, and (ii) the one containing original bands and additional data in the form of several vegetation indices and Leaf Area Index (LAI). We collected field samples using stratified random strategy, which were separated into two sub-datasets, as a basis for structural composition classification reference and accuracy assessment. In addition, comparison was also carried out using the original results and the one which was majority filtered. The results showed that the Maximum Likelihood algorithm performed the highest accuracies at a range of 74-86% using a combination of original bands and RVI (Ratio Vegetation Index). The result that was processed using a 5x5 majority filter showed the highest accuracy 86.29%. These results demonstrated that the pixel-based classification of Sentinel 2A imagery using the Maximum Likelihood algorithm could be used to map the structural composition of vegetation in the study area.
Development of satellite sensor systems capable of producing high spatial resolution digital images has led to the emergence of various alternative methods beside the more established per-pixel multispectral classifications. One alternative method is object-based image analysis (OBIA). At the beginning of its development, OBIA was primarily used for high spatial resolution images. However, the OBIA is now widely applied to images with medium- and even low-spatial resolutions. This study aimed to compare the effects of the OBIA and per-pixel classifications using using Landsat-8 OLI medium-spatial resolution image. Since the per-pixel classification relies solely on spectral aspects on various spectral bands, while the OBIA classification made use of spatial aspects as the main criteria, this study also made use of two land-cover/land-use classification schemes as references, i.e. spectral-oriented and spatial-oriented classification systems. The spectral-oriented classification scheme specifies categories from spectral perspective, i.e. pixel values in n-dimensional feature space; while the spatial-oriented one specifies categories with respect to their spatial characteristics. By using Kulon Progo region as a test area, the results showed that the OBIA classification was able to provide higher accuracy than that of per-pixel classification, both by referring to the spectral and spatial dimension classification schemes. The increase in accuracy provided by the OBIA classification proved to be greater when applied with a spatial dimension classification scheme, which was more than 10%, as compared to the improvement obtained by the spectral dimension classification scheme, i.e. 7%. This study also recommends the need for comparison studies using higher-spatial resolution imagery.
This study developed a method of satellite imagery-based land-cover/land-use mapping for Indonesia at 1:50,000 scale, but with a very detailed level of categorization. The method was developed by taking into account: (a) categorization target specified in the reference classification scheme, (b) the ecological characteristics of the Indonesian region, (c) the type of data used, and (d) the main approach that can be applied to all regions in Indonesia. A landscape ecological approach was selected, by combining digital and visual interpretation. The main data source was Landsat-8 OLI recorded in various dates of recording between 2016 and 2018, supported with SPOT-7 and Sentinel 2A imagery. Digital analysis includes geometric and radiometric corrections, pan-sharpening, and vegetation index transformation. Visual interpretation was carried out with an interpretive overlays strategy and/or land unit approach. Field work was carried out for collecting information on terrain characteristics that are relevant to the land-cover/land-use variation, to be used as a basis for re-interpretation process. Based on the developed methods, a set of land-cover/land-use maps on a scale of 1: 50,000 of the southern part of Sumatera, except Lampung Province, was delivered. It covers 247 map sheets. The interpretation accuracies have been assessed statistically, and they reached 79.54% for Bengkulu, 80.75% for Jambi, 79.2% for Riau Islands, and 81.02% for Bangka-Belitung. With a large number of classes has been mapped, i.e. over 70 categories, the accuracy levels achieved in this study were considerably high. Some notes on the results of the mapping were also included in this report.
Monitoring of rice field, as a place for producing rice is very important to realize one aspect of food security, namely food availability. Modern agriculture has been widely utilize remote sensing data, especially optical images for monitoring agricultural land in various aspects of land management. However, the use of optical images is hampered by cloud cover when monitoring rice fields because most of them located in tropical countries, so there is an alternative to using SAR imagery that has ability to penetrate clouds. One of the SAR image products is Sentinel-1A with band C on its sensors which was launched in 2014 and the data can be utilized by the wider community for free. The purpose of this study was to determine the ability of multitemporal Sentinel-1A SAR imagery in identifying paddy and non-paddy in Bantul Regency’s agriculture field which was measured through its mapping accuracy. Sentinel-1A multi-temporal images with ten recording dates from February to May 2018 were used as the main data for this study. The method used is a digital classification with two approaches i.e. parametric with MLC algorithm and non-parametric with k-NN algorithm. In addition, the Sentinel-1A, which consists of VV and VH polarization, performed in three classification schemes (VV multi-temporal, VH multi-temporal, and VV and VH multi-temporal). The classification results show that multi-temporal Sentinel-1A can be used to identify paddy and non-paddy fields with an accuracy of 77.69% (VV multitemporal-MLC), 82.15% (VH multi-temporal-MLC), 88.45% (VV and VH multi-temporal-MLC), 76.64% (VV multitemporal-kNN), 78.47% (VH multi-temporal-kNN) and 79.52% (VV and VH multi-temporal-kNN).
Topographic feature is one of the several factors affecting the distortion of the real reflectance value of objects. Digital processing used the surface reflectance values of satellite imagery needs the corrected images with the most minimized disturbances, hence several topographic correction methods using digital elevation data have been developed. This study examined the different result of topographic correction from several available elevation data in Indonesia, including SRTM DEM, topographic map (RBI), and DEMNAS. Sun-Canopy-Sensor+C (SCS+C) correction was applied on Landsat-8 data over Menoreh Mountains, Indonesia. The results obtained showed that DEMNAS produced the most topographically normalized images based on statistical and visual analysis. The availability of DEMNAS throughout Indonesia is the advantage to be used as an input of this pre-processing method. However, it needs to be examined first since the quality is not surely similar to our study site.
Development of spatial databases for systematic thematic mapping is a relatively complex activity, as compared to mapping a small area with arbitrary boundary. This research was conducted in the provinces of Riau Islands, Bangka Belitung Islands, Jambi, and Bengkulu, southern Sumatera. The stages of spatial databases development involved remote sensing, GIS, and cartographic activities. A synoptic overview of the landscape was carried out prior to the spatial database development. The landscape complexity was assessed using landscape-ecological approach, which was implemented in the delineation and classification. The remote sensing process started from spatial data collection of various images with various spatial resolutions, image pre-processing, followed by image analysis and interpretation. Pan-sharpened Landsat-8 images (15 meters) were used as main data, supported by SPOT 6/7 imagery (6 meters), Sentinel-2A imagery (10 meters) and DEMNAS digital elevation model (8 meters) for particular areas. This stage gave consequences to the multi-scale analysis in the process of land-cover delineation. The GIS process comprised the stages of compiling the interpretation results to form a seamless mosaic, topology construction, coding into Indonesian Geographic Element Catalog (KUGI) standard, metadata development, followed by topology checking. Those processes aimed to achieve a single map of Sumatera at 1: 50,000 scale. The cartographic layout design was the final stage of the spatial database development, where the land-cover classes symbol was also carried out in accordance with the established standards. Some problems and solutions in the whole processes were also discussed in this paper.
Geometric correction is an important step in image pre-processing, because it determines the the positional accuracy of the data. However, the geometric correction also includes pixel values interpolation in their new position, so that it may change original values. This study objectives were (a) to provide information on the effect of geometric correction models on the accuracy of land-cover classification, especially using per-pixel classification with maximum likelihood algorithm; and (b) to assess the effect of image resampling methods on the accuracy of the multispectral classification results. This study made use of Landsat 8 OLI Level 1G imagery covering Kulon Progo Area, Yogyakarta, so that several ground control points (GCPs) were needed to suppress geometric errors. Non-systematic geometric correction was undertaken using first, second and third order polynomial transformations. After that, several resampling processes were applied to the geometrically corrected image, i.e. Nearest Neighbour, Bilinear and Cubic Convolution interpolations. It was found that the affine transformation using six GCPs distributed over the edges of the image, delivered an RMSE value of 0.355539. In addition, the second order polynomial with 10 GCPs scattered around the edges of the image gave an RMSE value of 0.178053. While the third order polynomial transformation with 17 GCPs that were evenly distributed in the image produced an RMSE value of 0.100343. The resampling process produced new images with new pixel values, which were then tested with respect to their classification accuracies based on maximum likelihood algorithm. Samples for accuracy assessment were taken using stratified random sampling strategy. Samples were taken in terms of polygons whose size was determined by considering the pixels’ displacement as the results of geometric corrections. This study also found that resampling with nearest neighbour interpolation using third order polynomial equation produced the best overall accuracy of 75.46%, with a Kappa of 0.7032.
Tin mining is one of the main sectors of the national economy where the Bangka Regency is the largest tin producer in Indonesia. However, this sector cannot be separated from the pros and cons for a long time. In a way, this sector can increase both national and regional income but on the other side, the adverse effects of it can threaten the survival of humans and the environment. Open tin mining activity has converted previously vegetated land cover become the nonvegetated land cover. Furthermore, the land cover changes to the mining area have a major impact on global warming which has become an international issue in the past few decades. This research aims to map and measuring land cover changes especially from vegetated to non-vegetated land cover related to tin mining activity in Bangka Regency. This research using multitemporal Landsat imagery data acquisition in the year 2004 (Landsat 5 TM) and 2017 (Landsat 8 OLI) through digital image processing using Maximum Likelihood Classifier method. Previously, the image as a classification input through relative radiometric normalization. The result shows that tin mining activity in Bangka regency for thirteen years causes an area reduction in vegetated land cover. These results are expected to be an important input in policymaking for local governments to support the action plan which leads to mitigation of climate change.
Multispectral classification is one of the main methods in the analysis and processing of digital remotely sensed imagery, which until now is still widely used to generate land-cover/ land-use information. Technically, pixel-based classification methods rely on conventional approaches, as compared to GeoBIA, and it can be implemented using either supervised or unsupervised methods. The classification methods are supported by the rapid development of various image processing software, which provide a wide variety of algorithm options, so that the classification process can be carried out easily. Although relatively simple, an appropriate selection of multispectral classification algorithm may provide highly accurate land-cover maps. However, the highly accurate land-cover/land-use maps may also be influenced by image types and classification schemes that are used in the study. This study aimed to compare the results of the multispectral classification using maximum likelihood algorithm, for generating land-cover maps based on Landsat-8 OLI images (30 meters) and Pleiades imagery (2 meters). The classification referred to two different classification schemes relating to spectral and spatial dimensions. The results showed that the multispectral classification with spectral-related classification scheme applied to Pleiades imagery gave higher overall accuracy as compared to that of Landsat-8 OLI. It was also found that the highest overall accuracy achieved in this study was 81.7%, obtained using Pleiades imagery and referring to spectral dimension classification scheme. On the other hand, the lowest overall accuracy was obtained by the same imagery applied using spatial-related dimension. The relatively similar values of low overall accuracy for spatial-related dimension was also gained by Landsat-8 OLI imagery, proving that multispectral classification does not work well for spatial-related land cover classification scheme.
Malaria is one of deadly infectious diseases commonly found in tropical countries, and until now its preventive efforts are still going on. From a spatial-analytical perspective, the preventive efforts can be done by developing malaria vulnerability maps, which can be used as a basis for risk management. Remotely sensed imagery is a powerful source for collecting relevant spatial data for that purpose. Among various models, there are four analysis methods for generating such maps, i.e. scoring, matching, spatial multi-criteria evaluation (SMCE) and geographically weighted regression (GWR), which have been compared according to their effectiveness and accuracies. The authors tested those methods in Purworejo Regency, Central Java, Indonesia, which has been recognized as a malaria endemic area. This study used Landsat-8 OLI imagery as a basis for deriving spatial parameters closely related to malaria vulnerability . Each vulnerability spatial model’s accuracy was then evaluated by calculating the number of cases found in the field, with respect to each vulnerability class, and then compiling all values using cross tabulation. It was found that, among other methods, the SMCE-based malaria vulnerability map statistically delivered the best result.
Mangrove forest is an important ecosystem located in coastal area that provides various important ecological and
economical services. One of the services provided by mangrove forest is the ability to act as carbon sink by sequestering
CO2 from atmosphere through photosynthesis and carbon burial on the sediment. The carbon buried on mangrove
sediment may persist for millennia before return to the atmosphere, and thus act as an effective long-term carbon sink.
Therefore, it is important to understand the distribution of carbon stored within mangrove forest in a spatial and temporal
context. In this paper, an effort to map carbon stocks in mangrove forest is presented using remote sensing technology to
overcome the handicap encountered by field survey. In mangrove carbon stock mapping, the use of medium spatial
resolution Landsat 7 ETM+ is emphasized. Landsat 7 ETM+ images are relatively cheap, widely available and have
large area coverage, and thus provide a cost and time effective way of mapping mangrove carbon stocks. Using field
data, two image processing techniques namely Vegetation Index and Linear Spectral Unmixing (LSU) were evaluated to
find the best method to explain the variation in mangrove carbon stocks using remote sensing data. In addition, we also
tried to estimate mangrove carbon sequestration rate via multitemporal analysis. Finally, the technique which produces
significantly better result was used to produce a map of mangrove forest carbon stocks, which is spatially extensive and
temporally repetitive.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.