ISODATA Method Research Articles

Remote methods of research assessment of the impact of agricultural economic processes on the environmental destruction of the territory

The research is about solving the problem of assessing the impact of agricultural processes on the environmental safety of territories. The research was carried out on the selection, uploading and processing of data from the aerial survey of the area. High-quality multispectral images of the study areas were selected in accordance with the time-related thematic tasks of aerospace research of agricultural plots. With the help of Google Earth Engine platform, JavaScript programming language, ENVI and ArcMap software complexes, segmentation of the image of space images was carried out on these images. Each of them has its own normalized color scale, and highlights certain objects with different colors. To improve the visualization of differences in images, the classification of index images using the Isodata method was studied.An Agrocontrol module for automated deciphering of agricultural was developed. The module allows to select agricultural land with further vectorization and the possibility of creating a geodatabase automatically. Based on results, preliminary cartographic models of the studied territories were created with the indication of land plots that are dangerous for the ecological state of the region and require more careful consideration.At the stage of practical research, an analysis of the received data, based on the agricultural holding "Mriya", was carried out on the scheme of crop rotation, hybrids of grain crops grown on the territory of the plots, the types and amount of fertilizers used, as well as the volume of the harvest for the last 5 years. This made it possible to prove the correlation between the chemical composition of the used fertilizers and the corresponding exceedances of the MPC norms of the soil samples chemical composition.

An assessment of seasonal changes in forests using the segmentation and cluster analysis of Landsat space images

For the purposes of monitoring the state of forest ecosystems, it is most effective to use the capabilities of remote methods. On multispectral Landsat satellite images (a time series of summer and autumn images for 2000, 2001, 2013, 2014) of the territory of the Olekminsky State Nature Reserve, fragments of forests dominated by Gmelin larch (Larix gmelinii Rupr.) with an area of 250 km2 (scale 1:5000). Then the polygons were saved at three levels of segmentation (scaling) – 4, 16, 64 with scales of 1:2500, 1:1250, 1:625. During decryption, uncontrolled classification of polygons using the ISODATA (Iterative Self-Organizing Data Analysis Technigue) method into 2,4,10 classes was carried out. Classification into two classes was used to calculate the forest cover index. Seasonal changes were determined by the difference in the forest cover index values of the polygons in summer and autumn. It has been shown that the greater the difference, the greater the proportion of larch in mixed forest stands. Distribution curves of forest cover index values were constructed for polygons of the 3rd scaling level. Based on the results of classification into 4, 10 classes, statistical processing was carried out with the calculation of indicators of difference and similarity of polygons - dispersion of the general population and Fisher's test (F-test). The results of changes in population dispersion and F-test at different levels of segmentation and in different years are considered.

An assessment of forest restoration after fie using segmentation and cluster analysis of Landsat images

For the purposes of monitoring the state of forest ecosystems, it is most effctive to use the capabilities of remote methods. Fragments with an area of 250 km2 (scale 1:5000) were identifid on Landsat multispectral satellite images (time series of summer images for 1995, 2000, 2004, 2008, 2013, 2016) of the Olyokminsky State Nature Reserve. Then polygons were saved at three levels of segmentation: 4, 16, 64 with the scales of 1:2500, 1:1250, 1:625. When deciphering, an unmanaged classifiation of polygons was carried out using the ISODATA method (Iterative Self-Organizing Data Analysis Technigue) into 2, 4, 10 classes. Distribution curves for the values of the forest cover index for polygons of the 3rd level of segmentation were constructed. The results of classifiation into 4 classes were used to calculate the thematic pixel diffrence. According to the classification results for 4.10 classes, statistical processing was carried out with the calculation of the diffrence and similarity indicators of polygons, the dispersion of the general population and the Fisher test (F-test). A method is described for determining the disturbance of ecosystems and their restoration from the distribution curves of the forest cover index. The results of the change in the dispersion of the general population and the F-test at diffrent levels of segmentation and at diffrent stages of forest restoration are considered. The features of the transition between three levels of self-similarity (scaling) of multifractal structures as forests are restored are determined.

Isodata-Based Method for Clustering Surveys Responses with Mixed Data: The 2021 StackOverflow Developer Survey

Massive amounts of data are generated every day from all kinds of sources, from numerical data generated by sensors to veiled messages on social networks. Transforming these data into properly organized pieces of information and transforming it into resources for decision-making is complicated, not only because of the speed and volume at which it is produced, but due to the fact the high complexity of the context in which it is generated. Often, the first step in analyzing the data is to separate it into categories that correspond to segments of interest in that context. However, in many real cases, the limits of these segments and even the number of existing segments is unknown. Clustering techniques allow defining the classes of entities in a data set with sufficient relevance. However, those techniques usually work only with numerical data. Surveys are a very useful tool for collecting data in ill-defined contexts, but these data usually contain values that are not only numerical but of a very diverse nature. This paper presents a modification to the Isodata method to process data with mixed numerical and categorical values. The resulting algorithm is tested by analyzing the results of the 2021 Stack Overflow developer survey. The results obtained in the clustering of such data are sound and show that the Isodata method, with the proposed adaptations, can be successfully employed to discover patterns in complex mixed data.

An evolutionary algorithm based on approximation method and related techniques for solving bilevel programming problems.

In the engineering and economic management fields, optimisation models frequently involve different decision-making levels. These are known as multi-level optimisation problems. Because the decision-making process of such problems are hierarchical, they are also called a hierarchical optimisation problems. When the problem involves only two-level decision-making, the corresponding optimisation model is referred to as a bilevel programming problem(BLPP). To address the complex nonlinear bilevel programming problem, in this study, we design an evolutionary algorithm embedded with a surrogate model-that it is a approximation method and correlation coefficients. First, the isodata method is used to group the initial population, and the correlation coefficients of the individuals in each group are determined based on the rank of the leader and follower objective functions. Second, for the offspring individuals produced by the evolutionary operator, the surrogate model is used to approximate the solution of the follower's programming problem, during which the points in the population are screened by combining the correlation coefficients. Finally, a new crossover operator is designed by the spherical search method, which diversifies the generated offspring. The simulation experimental results demonstrate that the proposed algorithm can effectively obtain an optimal solution.

Application of satellite imagery to update depth-area-volume relationships in reservoirs in the semiarid region of Northeast Brazil

ABSTRACT Reservoirs are the primary source of water supply in the semiarid region of Pernambuco state, Brazil, because of the constant water scarcity affecting this region. Knowledge of the amount of water available is essential for the effective management of water resources. The volume of water stored in the reservoirs is calculated using the depth-area-volume relationship. However, in most reservoirs in the semiarid region, this relationship is currently out of date. Therefore, the objective of this study was to explore the potential and limitations of the application of the ISODATA unsupervised classification method to calculate the depth-area-volume relationships of reservoirs in the semiarid region of Pernambuco, Brazil. The application of the ISODATA method was evaluated in three reservoirs in the state of Pernambuco, i.e., Poço da Cruz, Barra do Juá, and Jucazinho. The results were compared with the updated curves of reservoirs obtained from bathymetry and recent LiDAR surveys. The ISODATA method presented satisfactory results for the three reservoirs analyzed. The mean absolute error of the volume in Poço da Cruz and Barra do Juá was lower than 1% of the maximum capacity. The use of the ISODATA method meant that the surface area underestimation tendency in the Poço da Cruz reservoir was less than when spectral indices were used.

Mapping Urban Impervious Surface With an Unsupervised Approach Using Interferometric Coherence of SAR Images

Impervious surface is significant in hydrology, urban management, ecology, and other research areas. Therefore, extracting impervious surface is crucial to understanding the change of environment and ecosystem. However, previous supervised classification methods usually rely on comprehensive training samples and human experiences. The article on automatic and efficient impervious surface extraction is still underexplored. In this study, we investigated the potential of using interferometric synthetic aperture radar technology for unsupervised urban impervious surface (UIS) mapping. A total 136 coherence maps of Hong Kong with different perpendicular and temporal baselines were used to classify UIS and non-UIS through setting different coherence thresholds. We proposed a new method, entitled interferometric coherence thresholding method, which can achieve high classification accuracy using coherence map. The result illustrates: first, using a threshold of 0.4, nearly 90% of images achieve an overall accuracy of over 80%. The highest one reaches 88.25%, which is much higher than using K-means and ISODATA method; second, small temporal baselines (12 and 24 days) are likely to reduce the classification accuracy; third, using optimal classification threshold, the coherence of two SAR images would not bring large impact on classification results.

Arazi Kullanımı/Arazi Örtüsü Değişikliklerinin Ekosistem Hizmetleri Üzerindeki Etkisinin Sayısallaştırılması: Adana-Karaisalı Örneği

Global scale changes in land use affect biodiversity, global climate, and ecosystem services. Integrating a local and regional ecosystem service approach into plans and policies levels is essential for sustainable land management. In this study, land use/land cover differences in Adana-Karaisali district of Turkey were investigated using Normalized Difference Vegetation Index (NDVI). Landsat satellite images of the area from 2000 and 2016 were analyzed using ISODATA method and 16 land use were identified based on the third level of the CORINE classification framework. The results showed that the biggest changes occurred in agricultural areas, 2.57% decrease in non-irrigated arable land while a 7.14% increase was observed in permanently irrigated lands. In spite of the decrease in the number of inhabitants in the area, continuous urban fabric increased at the rate of 0.07% and 7.62% increase was observed in forest lands. The positive or negative effects of the identified changes on ecosystem services based on the Common International Classification of Ecosystem Services (CICES) framework were evaluated. In the case of Karaisalı, in spite of the decrease in water bodies and natural landscapes, it has been observed that increases in agricultural areas, forests, and artificial surfaces positively have affected ecosystem services valuation in the district.

An adaptive learning method of anchor shape priors for biological cells detection and segmentation

Background and objective: Owing to the variable shapes, large size difference, uneven grayscale and dense distribution among biological cells in an image, it is still a challenging task for the standard Mask R-CNN to accurately detect and segment cells. Especially, the state-of-the-art anchor-based methods fail to generate the anchors of sufficient scales effectively according to the various sizes and shapes of cells, thereby hardly covering all scales of cells.Methods: We propose an adaptive approach to learn the anchor shape priors from data samples, and the aspect ratios and the number of anchor boxes can be dynamically adjusted by using ISODATA clustering algorithm instead of human prior knowledge in this work. To solve the identification difficulties for small objects owing to the multiple down-samplings in a deep learning-based method, a densification strategy of candidate anchors is presented to enhance the effects of identifying tinny size cells. Finally, a series of comparative experiments are conducted on various datasets to select appropriate a network structure and verify the effectiveness of the proposed methods.Results: The results show that the ResNet-50-FPN combining the ISODATA method and densification strategy can obtain better performance than other methods in multiple metrics (including AP, Precision, Recall, Dice and PQ) for various biological cell datasets, such as U373, GoTW1, SIM+ and T24.Conclusions: Our adaptive algorithm could effectively learn the anchor shape priors from the various sizes and shapes of cells. It is promising and encouraging for a real-world anchor-based detection and segmentation application of biomedical engineering in the future.

Improving Statistics Learning Literacy in Estimating the Grouping of the Number of Health Workers Using the Fuzzy ISODATA Method

Improving Statistics Learning Literacy in Estimating the Grouping of the Number of Health Workers Using the Fuzzy ISODATA Method

COMPARISON OF SEVERAL REMOTE SENSING IMAGE CLASSIFICATION METHODS BASED ON ENVI

Abstract. With the development of remote sensing technology and the increasing accuracy of remote sensing images, research on the accuracy of remote sensing classification is becoming more and more important. However, the classification accuracy obtained by different classification algorithms is also different. To this end, this paper selects the maximum likelihood method and the minimum distance method in the traditional supervised classification, the ISODATA method and the k-means algorithm in the unsupervised classification, and uses these four algorithms to classify the Landsat images in the research area of Heze City. The classification results are obtained and the results are evaluated. Then the four algorithms are compared separately, and the advantages and disadvantages of each algorithm are analyzed. The results show that the classification accuracy of the maximum likelihood method in the supervised classification is relatively high, and the classification accuracy is 82.3281%. The ISODATA algorithm in the supervised classification is superior to the K-means algorithm in clustering effect.

Геоинформационное картографирование ландшафтов на примере Приморского хребта (Прибайкалье)

The authors present the results of geoinformation mapping the Primorskiy Ridge landscapes using Landsat 8 satellite images, the digital elevation model SRTM and the factor-dynamic classification of geosystems. At the first stage, the remote sensing data for different seasons were classified using the ISODATA method. Then, using the digital elevation model, the landforms were classified basing upon the topographic position index. According to combining the classification parameters of one of the space images and digital elevation model, each polygon is automatically assigned to a certain preliminary type of landscapes using boolean expressions. Legend adjustments were made basing upon the fieldwork materials. As a result, a digital landscape map of the southern part of the Primorsky Ridge was created; it reflects the landscape structure at the level of facies groups and contains attributive information about the landform, altitude, slope and aspect, topographic wetness index. The analysis of the landscape pattern showed a high fragmentation of landscape polygons, formed due to overlay operations, which indicates the need for generalization of landscape contours.

ACCURACY OF UNSUPERVISED CLASSIFICATION TO DETERMINE CORAL HEALTH USING SPOT-6 AND SENTINEL-2A

Abstract. Characteristics of corals spectral from different species are expected to have optically different characters. The aims of this research are to compare unsupervised classification between IsoData and K-Means methods with Lyzenga application, and to analyze the precision of SPOT-6 and Sentinel-2A satellite imagery in classsifying shallow water habitat. The image processing are atmosferic correction, cropping, masking, Depth Invariant Index, Unsupervised classification, ground truthing, reclassify, accuracy assessment, and shallow water habitat spectral reflectance analysis. Rubble and dead coral with algae were indicating as coral death due to either damaging human activity or natural death such as bleaching. The accuracy of unsupervised classification IsoData and K-Means method have the same accuracy 62.50%. The IsoData method is better detected live coral and algae. Rubble were dominant detected in K-Means method.

CLASSIFICATION OF LAND COVER FOR NPP "PRIPYAT STOKHID" ACCORDING TO SENTINEL-2 DATA

The methodology for researching of natural protected territories using GIS QGIS according to remote sensing data by Sentinel-2 on the territory of the National Nature Park «Pripyat-Stokhid» is proposed. Based on the results of the unsupervised classification using the ISODATA method and interpretation were chosen 7 classes that clearly correspond to natural and anthropogenic objects. There are open reservoirs; territories covered with forests; territories covered with young forest and continuous bushes; territories with natural herbaceous vegetation; territories with swamps and characteristic marsh vegetation; built-up territories and open ground; and unclassified objects. On the basis of the classes received by the ISODATA method, an unsupervised classification was performed by the K-Means method. Difference in pixels’ distributions of classified images by classes between the two methods is an average of 1.4%. The results of classification by both methods are correlates with the project of National Nature Park «Pripyat-Stokhid» natural protected territories. Due to the lack of reliable information about natural objects, the Using of the proposed methodology, it is possible for any natural protected territories to obtain freely available Sentinel-2 remote sensing data, using free open source GIS software QGIS make analyses it and obtain classification signatures of the main classes natural and anthropogenic objects of the researched territory.

Global classification of lunar reflectance spectra obtained by Kaguya (SELENE): Implication for hidden basaltic materials

A global classification map of lunar absorption spectra is developed under unified classification criteria by applying the K-means and ISODATA unsupervised classification methods. The spectra were obtained by the Spectral Profiler (SP) onboard the Kaguya (SELENE) lunar orbiter spacecraft. The spectra are mostly categorized into seven spectral groups by the K-means method: K1, K2, K3, K4, K5, K6, and K7. Then, each K-means group is divided into 66 total classes by the ISODATA method. The K1 and K3 groups are classified into 21 and 14 classes, respectively, for which the average spectra are high-calcium pyroxene (HCP)-abundant compared to the other groups. Both groups are mainly located on maria and pyroclastic deposits. However, while the K1 group can be found primarily near the center of large maria, the K3 group can be found mainly at small maria and the margins of large maria. The K2 group contains eight classes and shows average spectra, such as low-calcium pyroxene (LCP), in abundance, which are found mainly on the South Pole-Aitken (SPA) basin and on the coast of Mare Frigoris. The K5 (eight classes), K6 (two classes) and K7 (three classes) groups indicate average spectra with shallower absorption (between 0.9 µm and 1.2 µm wavelength) than those of the K1, K2, and K3 groups. These groups are located in highland regions, and the K5 group is located in high-albedo regions, such as fresh craters. The K6 group is found on nearside highland areas and the K7 group is found on farside highland areas. The K4 group with nine classes shows average spectra, such as mixtures of K1/K2/K3 and K5/K6/K7, and these classes are located around boundaries between mare/SPA and highland. In addition, one class is independently defined without using K-means and ISODATA because its spectra are unsuitable for automatic unsupervised classification. However, this class is still geologically meaningful.The average spectra of the ISODATA classes show not only the characteristics of the major host material belonging to a spectral group, but they also show the small influence of materials from other groups in nearby regions on that host material. This influence is considered to be caused by horizontal material mixing of materials in that location with the ejecta from nearby regions. However, occasionally the ISODATA classes of the K4, K5, and K6 groups influenced by the K1/K3 groups are clustered geographically in regions far from the K1/K3 regions. Some locations of the clusters correspond to locations of the well-known cryptomaria and small pyroclastic deposits. On the other hand, some locations of them also contain candidates of hidden subterranean basaltic materials, such as undiscovered cryptomare, dike, and sill estimated based on recent lunar gravitational anomalies. This agreement suggests to be appeared hidden basaltic materials by the material mixing occurring vertically. The total area of the clusters except for known cryptomaria is approximately two times larger than that of the known cryptomaria, and this area reaches approximately 20% of the total area of exposed maria. The total volume of hidden basaltic materials, except for the known cryptomaria, is estimated to be 104 to 106 km3, depending on assumptions of the relative abundance between cryptomare and dike/sill. This volume is comparable to that of the known cryptomaria and approximately 10% of that of the exposed mare.

Monitoring land cover change of a river-floodplain system using high-resolution satellite images

In this study, a method was developed to monitor habitat structure in river-floodplain systems using high-resolution satellite images from 2010 to 2012 across a 30-km longitudinal section of the Tagliamento River, Northeast Italy. Three ortho-corrected RapidEye satellite images at 5-m spatial resolution and cloud cover of 89%) than that of the conventional unsupervised ISODATA method (> 83%) as river water and vegetation could be differentiated more accurately using this new method. The unchanged area was greater for river water than for vegetation and bare alluvium. These results indicated that habitat distribution changed spatially and temporally, especially for fluvial habitats, while the composition of habitat types was preserved in the middle reaches of the Tagliamento River. This method can be used to continuously and accurately monitor the large-scale spatiotemporal dynamics of habitat structures.

Восстановление допожарных характеристик лесных насаждений на гари по данным космической съемки и полевых наблюдений

В статье представлен метод восстановления допожарных характеристик древостоя путем классификации космического снимка, полученного до пожара, с верификацией и описанием тематических классов. Модельный объект – гарь площадью 9 га в Шатурском районе Московской обл. Предложено при отсутствии таксационных данных устанавливать характеристики насаждений после гибели древостоев (как от пожаров, так и от других причин) на основе данных космической съемки и полевых наблюдений на прилегающей территории. The loss of stem wood after stand destruction was evaluated on a model site - a 9-ha burned area that resulted from a forest and peat fire in 2010 in the Shatura district, Moscow Province. The study was carried out as part of the project "Restoring Peatlands in Russia - for fire prevention and climate change mitigation" financed under the International Climate Initiative (IKI) by the German Federal Ministry for the Environment, Nature Conservation, Building and Nuclear Safety and facilitated by KfW Development Bank. Forest inventory data were unavailable for the model site, so it was proposed to evaluate pre-fire condition of forest stands by analyzing space imagery taken before the fire and by verification and description of thematic classes. Classification of the multispectral Spot 5 image with a spatial resolution of 10 m was carried out using the ISODATA method. Characteristics of thematic classes were obtained using forest-geobotanical releve’s and measurements at sample plots in a non-burned adjoining forest area. Mapping of preserved stumps with identification of tree species, age and diameter was made for additional verification of the defined thematic classes. The loss of stem wood for the forest thematic classes averaged 282.2 m3/ha and for the burned area as a whole (including non-forest classes) 264.7 m3/ha. It has been demonstrated that, even without forest inventory data, original stand characteristics for the area with stand destruction (both from fire and by other reasons) can be identified using space imagery coupled with on-the-ground investigation of an adjoining area.

THE USE OF LANDSAT SATELLITE IMAGES FOR ECONOMIC DAMAGE ASSESSMENT TO FORESTRY OF MORDOVIA DUE TO A FOREST WINDFALL

The damage caused by extreme weather events in some years exceeds the amount of logging even in areas of intensive forest management. So, on the example of the studied object on whose territory a windfall formed as a result of a squally wind, it has been shown as a rapid identification and assessment of damaged forest areas allows one to carry out effective measures for forest protection and/or restoration. In its turn, the untimely information on negative changes in the forests may lead to a decrease in efficiency of sanitary and preventive measures and cause subsequent outbreaks of insect pests and increased fire danger. The aim of our work is the mapping of the spatial distribution of forest windfall sites on the territory of Staroshaigovskoe district forestry of Saransk territorial forestry. As the source of information we used not only remote sensing data but also the results of the ground surveys of forest pest damaged areas. IsoData method was used for the unsupervised classification of images. Only in the study area the damage was nearly 300 ha of stands, totally lost as a result of windfalls, which makes 1.5% of the district forestry area. Analysis of the features of spatial distribution of windfalls in the study area has been made, and based on actual source data a map has been compiled showing the distribution of the windfall formed on lands of the forest Fund as a result of a squally wind on the territory of the Republic of Mordovia on 21 June 2015, and the economic damages to forestry of the studied object calculated. A selective estimation of species composition and age structure of stands affected by forest windfalls has been made.

Fuzzy C-Means Clustering Based on Improved Marked Watershed Transformation

Currently, the fuzzy c-means algorithm plays a certain role in remote sensing image classification. However, it is easy to fall into local optimal solution, which leads to poor classification. In order to improve the accuracy of classification, this paper, based on the improved marked watershed segmentation, puts forward a fuzzy c-means clustering optimization algorithm. Because the watershed segmentation and fuzzy c-means clustering are sensitive to the noise of the image, this paper uses the adaptive median filtering algorithm to eliminate the noise information. During this process, the classification numbers and initial cluster centers of fuzzy c-means are determined by the result of the fuzzy similar relation clustering. Through a series of comparative simulation experiments, the results show that the method proposed in this paper is more accurate than the ISODATA method, and it is a feasible training method.

Remote Monitoring of Heading Rice Growing and Nitrogen Content Based on UAV Images

Rice heading is the critical stage of the growth of rice, rice plants are relatively large, field canopy, poor resistance, and coincided with the high temperature and rainy weather, pests and diseases are more prone period. Therefore, the use of UAV to monitor the growth status of rice, easy to understand rice growth and nutritional status, in order to achieve high-quality, high yield, efficient purpose. This article is the field experiments under different nitrogen levels, using UAV monitoring multispectral images of rice, by reference remote sensing spectral indices, derived green normalized difference vegetation index (GNDVI) relative to other spectral index is more suitable for rice field biomass inversion modeling. At the same time, this study is based on the inversion model, implemented the rice growing and nitrogen content graded by ISODATA methods in ENVI. Achieved classification on rice growth and nitrogen remote sensing thematic map in ArcGIS, to provide timely and accurate information for rice seedling diagnosis and management decision, has reached the purpose of rice production precise management.