Remote Sensing

Appraisal of Urban Heat Island Detection of Peshawar Using Land Surface Temperature and Its Impacts on Environment Abstract Last couple of decades witnessed a rapid escalation in urban temperature of Peshawar city and its neighboring localities. This alarming condition gave birth to climatic term urban heat island-created drastic alteration in surface temperatures. In this study, thermal infrared remote sensing data have been employed to map out and monitor such micro-climatic variation in temperatures in land use/land cover exposed surface to the environment. To assess these outcomes resulting from human activities, Landsat TM data band 6 was subjected through ERDAS Imagine 2013. For further processing, ARC GIS helped a lot in making maps to pinpoint the heat island in and around the city. Moreover, a relationship of land surface temperature with urban sprawl, environmental and industrialization was established. This study has shown a substantial upsurge in temperature about to 1°–3°. Urban sprawl and industrialization at the edges are accounting for these conditions. Urban and industrial data have also reinforced the fact being drawn from remotely sensed data. Hence, evaluation of land surface temperature data captured through remote satellite has proven to be effective tool not only for monitoring and analyzing temperature but also for assessing its adverse impacts on the environment and climate.

The Effect of Rapid Population Growth on Urban Expansion and Destruction of Green Space in Tehran from 1972 to 2017 Abstract Time series analysis of satellite data is a very powerful tool for change detection applications especially for monitoring urban expansion. Urban expansion results in an accelerating proportion of the population living in cities and caused considerable effects on green space structures and areas. Many methods and models have been tested to determine urban expansion. In this paper, after computing the Tehran area and its green space coverage, the relationship between the population with urban expansion and green space coverage is estimated by regression analysis. As a consequence of Tehran texture and dataset for processing, raster to vector conversion was the most effective method for computing urban area, and also green space coverage was determined with normalized difference vegetation index. Machine learning algorithms such as support vector regression (SVR) and random forest (nonparametric method) were utilized to estimate green space coverage and urban area, respectively. Although RF model improves the accuracy of estimated population (R2 = 0.97, RMSE = 38.18, and MAE = 83.44), the use of SVR model considerably improves the estimation of green space coverage (R2 = 0.93, RMSE = 47.12, and MAE = 100.18) as well.

Object-Oriented Method Combined with Deep Convolutional Neural Networks for Land-Use-Type Classification of Remote Sensing Images Abstract Land-use information provides a direct representation of the effect of human activities on the environment, and an accurate and efficient land-use classification of remote sensing images is an important element of land-use and land-cover change research. To solve the problems associated with traditional land-use classification methods (e.g., rapid increase in dimensionality of data, inadequate feature extraction, and low running efficiency), a method that combines object-oriented approach with deep convolutional neural network (COCNN) is presented. First, a multi-scale segmentation algorithm is used to segment images to generate image segmentation regions with high homogeneity. Second, a typical rule set of feature objects is constructed on the basis of the object-oriented segmentation results, and the segmentation objects are classified and extracted to form a training sample set. Third, a convolutional neural network (CNN) model structure is modified to improve classification performance, and the training algorithm is optimized to avoid the overfitting phenomenon that occurs during training using small datasets. Ten land-use types are classified by using the remote sensing images covering the area around Fuxian Lake as an example. By comparing the COCNN method with the method based solely on CNN, precision and kappa index were selected to evaluate the classification accuracy of the two methods. For the COCNN method, on the basis of the classification statistics, precision and kappa index coefficients are 96.2% and 0.96, respectively, which are 8.98% and 0.1 higher than those of the method based solely on CNN. Experimental results show that the COCNN method reasonably and efficiently combines object-oriented and deep learning approaches, thereby effectively solving the problem of the inaccurate classification of typical features with better classification accuracy than the simple use of CNN.

Hybrid Band Combination for Discriminating Lithology of Dunite in Ultramafic Rocks Abstract In recent years, satellite image and other data provided by the terrestrial component of remote sensing technology have been used for detecting hydrothermal alteration minerals and regional geology mappings. Existing rock detection methods and algorithms in remote sensing can only be used to determine main rock groups such as ultramafic, granitoid, metamorphic and sedimentary. The use of new remote sensing methods in the determination of subgroups of these rocks (i.e. ultramafic rocks consist of harzburgite, dunite and lherzolite) is important for geological mapping. In particular, the mapping of dunite is necessary for the exploration of chromite deposits. In this study, it was aimed to determine the spectral behaviour of dunite in a new band ratio and to adopt a new hybrid band combination for geological mapping and exploration of new chromite mineralization areas. Due to the presence of the complex units it holds, it is composed mainly of ultramafic rocks, and Eskikarahisar Region, Sivas City (Turkey), where it is located on the South of Sivas, was selected as study area. Ultramafic rock samples were collected in the study area. To determine the types and mineral contents of the samples, petrographic investigations of the samples were conducted for identifying the samples that contain dunite. Spectral measurements of the representative dunite samples were taken to combine the averages of spectral signatures. After merging spectral signatures and the ASTER SWIR band regions, ASTER SWIR bands that could be used for rationing were determined for adopting a new ASTER band rationed image and band with enhanced images data. This article focuses on the adopting a hybrid combination with gamma transform applied, which contains the rationed image and enhanced ASTER SWIR components, leading to effective mapping of dunite rocks exposed on the study area.

Tectonic Analysis of Lineaments in the Gara Anticline, Kurdistan, Northern Iraq Abstract Satellite images were used to interpret the tectonic origin of lineaments in the Gara anticline. This anticline is one of the largest structures in the Zagros fold-thrust belt of Iraqi Kurdistan, near the northeastern boundary of the Arabian Plate. Lineaments of the Gara anticline were visually mapped on QuickBird images using the ESRI ArcGIS software and automatically extracted from Landsat 8 OLI data using the lineament extraction (LINE) module of PCI Geomatics software. The linear features obtained by both methods were statistically analyzed according to their orientation, length, frequency and density to provide a useful quantitative framework for interpreting the tectonic origin of these structures. Lineament orientations indicate a good symmetric relationship to the hinge of the Gara anticline. The results obtained from both techniques show five major sets: a hinge-perpendicular “extension” set (N 05°E), a hinge-parallel “extension” set (N 85°W) and three hinge-oblique “shear-hybrid” sets (N 45°E, N 45°W and N 75°E). These dominant trends are consistent with previous field studies. Most of the lineaments in the Gara anticline are tectonic in origin and probably developed during the growth of the Gara anticline by a regional N–S-directed compression in the Mid Miocene time.

Determination of Opium Poppy ( Papaver Somniferum ) Parcels Using High-Resolution Satellite Imagery Abstract Narcotic plants contain substances that cause unusual excitation and subsequent depression of the central nervous system. Many narcotic plants contain substances that have medicinal properties and are used primarily as pain relievers. Opium poppy (Papaver somniferum) is one of the most cultivated narcotic plants. The planning of Opium poppy growing is controlled by United Nations and Drugs and Medicines Control Program. This study was conducted to establish a basis for determining the traceability of poppy cultivating areas using remote sensing in the large plain. A very high-resolution QuickBird-2 satellite image was used in the study. The software of ERDAS Imagine and eCognition Developer was performed for image processing and classification. A variety of classification methods were performed on the satellite image. ArcGIS software was used for accuracy assessment, ground control, and map production of the poppy cultivating area. The accuracy of classification methods was compared. The producer–user accuracy was estimated as 97.99% using spectral difference sub-segmentation process of multiresolution segmentation algorithm in object-based classifications method. This algorithm can be a practical approach for determining of opium poppy parcels in the large plain.

Forest Leaf Area Index Inversion Based on Landsat OLI Data in the Shangri-La City Abstract Leaf Area Index (LAI) is an important index that reflects the growth status of forest vegetation and land surface processes. It is of important practical significance to quantitatively and accurately estimate Leaf Area Index. We used the Landsat-8 operational land imager single-band images, and 15 vegetation indices that were extracted from the multi-band were combined with the LAI data measured from the CI-110 canopy digital imager to establish the LAI estimation model. Through the leave-one-out cross-validation method, the accuracy of various model estimation results was verified and compared, and the optimal estimation model was obtained to generate the LAI distribution map of Shangri-La City. The results show that: (1) the multivariable model method is better than the single-variable model method when estimating LAI, and its determination coefficient is the highest (R2 = 0.7903). (2) The full-sample dataset is divided into Alpine Pine forest, Oak forest, Spruce–fir forest, and Yunnan Pine forest for analysis. The coefficient of determination of the model simulation is improved to varying degrees, and the highest R2 increased by 0.1652, 0.1040, 0.1264, and 0.0079, respectively, over the full-sample. The corresponding best models are LAI–DVI (Difference Vegetation Index), LAI–NNIR (normalized near-infrared), LAI–NMDI (Normalized Multi-band Drought Index), and LAI–RVI (Ratio Vegetation Index). (3) The LAI values in Shangri-La City ranged from 0.9654 to 5.5145 and are mainly concentrated in high vegetation coverage areas; and the higher the vegetation coverage level, the higher the LAI value.

Assessment of Topographical Factor (LS-Factor) Estimation Procedures in a Gently Sloping Terrain Abstract The major uncertainty in soil erosion assessment studies is derived from LS-factor constituting slope length and slope steepness factors. Empirical soil erosion models employing different algorithms for estimation of LS-factor using raster-based digital elevation models (DEMs). Different algorithms have been adopted for LS-factor determination in soil erosion studies without proper justification for their selection according to the terrain characteristics; a few among them addressed suitability of the algorithms on hilly terrains. The present study focused on the performance of LS-factor estimation methods involving specific contributing area (SCA) method and cumulative slope length method for slope length factor and USLE, RUSLE and USPED algorithms for slope steepness factor in a gently sloping terrain. The results showed that SCA method is the best performing method in gently sloping terrain since the effect of contour length exponent get minimized since there are less influence from diagonal flow direction. The pixel-to-pixel-based slope length exponent may result in more appropriate estimation of slope length factor in gently sloping terrains. The SCA-based slope length estimation along with USLE S-factor algorithm was found to perform well under different elevation classes and slope classes in both SRTM DEM and ASTER DEM. The results from the study may be helpful in appropriate prediction of soil erosion in gently sloping terrains.

Introduction of Spatio-Spectral Indices for Using Spatial Data in Multispectral Image Classification Abstract Different methods of using spatial information in image classification are presented. One approach is to quantify image texture to produce features for use in classifiers, and there are various methods with adjustable parameters for texture quantification. The produced features are numerous and are in different discriminating image classes. Therefore, there is a need for selecting their optimum combination, or to alternatively create a set of features that abstract their class discernibility. Inspired by spectral normalized difference indices, the concept of the spatio-spectral index is introduced in this article to produce indices from a series of spatial features created from image spectral bands. In the proposed method, the produced spatio-spectral indices for each class are used as the abstract of spatial features. Along with the image spectral bands, they are used as new feature forms for supervised classification. Features with maximum and minimum values in each class were selected after production of the average vector in the feature space, and the removal of features with a small variation range. Next, non-repetitive band pairs were selected and spatio-spectral indices were produced. Using this method, the number of selected spatial features was at most twice the number of classes and was used to produce spatio-spectral indices. Use of the produced features in classification improves classification accuracy significantly (about 30% and 6% in the two test images used here) by enhancing class discrimination and decreasing computational time. This method is also explicit and direct, with no need to use iterative optimization processes.

Research and Verification of a Remote Sensing BIF Model Based on Spectral Reflectance Characteristics Abstract Banded iron formations (BIFs) represent the most important and widely distributed iron ore resources in the world. Monitoring BIF resources is of great significance for tracking reserves and for the orderly exploitation of resources. Previous remote sensing (RS) recognition methods for ground objects in mining areas have been developed based on satellite sensor data, but those methods do not incorporate the spectral characteristics of ground objects. Consequently, these modeling methods are relatively blind because they do not consider the real spectra of ground objects in advance, affecting the accuracy of RS recognition to a certain degree. In the present study, the visible to near-infrared spectra of two types of typical iron ore (magnetite and hematite) and of the surrounding rocks in a BIF deposit were first recorded by a field-portable spectrometer in a mining area. Then, the differences in the spectral characteristics among the two types of iron ore and surrounding rocks were analyzed. Based on the different spectral characteristics in different wavelength ranges, RS extraction and classification models for iron ore were constructed and applied to Landsat 8 data for the actual recognition of iron ore in an open pit. This study yielded the following results. There were remarkable differences in the spectral characteristics among the two types of iron ore and surrounding rocks. In addition to distinguishing the iron ore from the surrounding rocks, magnetite and hematite were further differentiated based on the constructed inversion model. The accuracies of distinguishing iron ore from the surrounding rocks, identifying hematite, and identifying magnetite were 83.5%, 83.3%, and 85%, respectively. The results demonstrated that iron ore zones can be identified automatically by the RS model based on the measured spectral characteristics and the inversion model. Accordingly, this model can increase the identification accuracy and provide a new method for BIF deposit detection and exploitation monitoring.