Research Articles Base: Multi-spectral Analysis

The use of image texture in multi-spectral analysis of satellite images for classifying land cover/land use with respect to any one field of research

Introduction Preamble Since the launch of the LANDSAT 4 and 5 satellites in the early 1980s the use of satellite images in archaeological work has become a realistic possibility. The Thematic Mapper scanner collects data on seven different bands of the electromagnetic spectrum, and because different ground features have distinctive reflective properties, analysis of these data can identify details of the earth’s surface from distant space. Certain bands have been used to identify cultural features such as roads and buildings, other bands are sensitive to different rock and soil types and yet others to different types of vegetation. Satellite images are large raster images and can be manipulated and enhanced by Image Processing techniques just like any other digital image. Each LANDSAT image covers an area of 34,000 sq km and has a spatial resolution of 30 m; each pixel in the image represents a 30-metre square on the ground. The French SPOT 12 satellite improves on this resolution with 20 m for multi spectral images and 10 m for single-band images covering an area of 48,000 sq km ( 2003).   Because of this rather poor resolution these satellite images have tended not to be used for the prospection and location of individual sites unless the sites are of a considerable size. On a LANDSAT image, for example, unless an archaeological feature fills a considerable proportion of a 30-metre pixel it is not going to influence the numeric value of that pixel. Where satellite images have proved to be very useful, however, is in providing environmental and landscape information in areas where maps are either difficult to acquire or difficult to use such as jungle and desert. An early example of the latter is the UNESCO Libyan Valleys Survey where a LANDSAT image provides essential background information for an archaeological survey placing known, and identifiable, archaeological features within a wider landscape setting. Image Processing also allows for the classification of a landscape into different land types, and land uses, according to the spectral signal, and this information can be used for locational analysis or for survey design. Again, because satellite images are spatial data, their integration with other types of landscape, environmental and archaeological information through GIS software is becoming increasingly common practice ( 2003). Satellite images are used by institutions and governments for different purposes.  The use of satellite images changes the approaches of a government in dealing with people. The paper will discuss about the use of image texture in multi spectral analysis of satellite images for classification of land cover or land use. Such topic is interesting because this study may help in solving the land problems of nations.  This also opens up other related topics.   Aims/Objectives The main aim of the study is to discuss about the use of image texture in multi spectral analysis of satellite images for classifying land cover/land use with respect to any one field of research. Another aim of the study is determine the importance of image texture in analyzing satellite images with regards to land cover/land use.   Approach, structure and outline of the essay The essay will be a combination of an analytical and theoretical structure wherein analysis of the topics in the essay will be supported by related theories. The essay will be divided into different parts. Each part will provide information that has something to do with image texture and classification of land cover/land use. The paper will also have a summary and conclusion part at the end to give further discussions about the topic.   Part 1 Texture Analysis and Multi spectral analysis Data mining The motivation behind mining data, whether commercial or scientific, is the same the need to find useful information in data to enable better decision making or a better understanding of the world. Traditionally, data analysts have turned to data mining techniques when the size of their data has become too large for manual or visual analysis. In the science and engineering domains the size of the data is only one reason why data mining techniques are gaining popularity. Science data in areas such as remote sensing, astronomy, and computer simulations is routinely being measured in terabytes and petabytes. However, what makes the analysis of these data sets challenging is not just the size, but the complexity of the data. Advances in technology have introduced complexity in scientific data, complexity that can take various forms such as multi-sensor, multi-spectral, multi-resolution data; spatiotemporal data; high dimensional data; structured and unstructured mesh data from simulations; data contaminated with different types of noise; three-dimensional data, and so on (2003). As a result of this complexity, visual data analysis, given its subjective nature and the human limitations in absorbing details, is becoming impractical even for moderate-sized scientific data sets large to massive data sets visual analysis is practically impossible. As a result, science and engineering data sets provide a very rich environment for the application of data mining, one in which the diversity of problems is matched only by the potential benefits obtained when knowledge is discovered in the data (2003).   Texture analysis It is readily apparent that in the physical geographic domain, continued rapid developments in remote sensing have dramatically increased the availability of data describing earth surface processes, such as climate, changes in land cover and land use, deforestation, and urbanization. Each of these new sources of physical geographic data are related to aspects of human spatial activity, but none of them can be thought of as materially augmenting more traditional social scientific data that describe the social characteristics of individuals and groups. It is also noteworthy that such data are rarely useful without extensive ground-truthing, a process that remains time-consuming and expensive ( 2004). Allied to the use of a GIS is remote sensing for data gathering. This deals with the detection and measurement of phenomena without being in contact with them. This data gathering method uses devices sensitive to electromagnetic energy such as light, heat, and radio waves. Remote sensing provides a unique perspective from which to observe large regions and global monitoring is possible from nearly any site on earth. In practice, satellite remote sensing has been put to use to estimate atmospheric water vapor, trace gases, aerosol particles, clouds and precipitation, and to monitor and quantify changes in territories that are otherwise not accessible. The use of remote sensing means changes can be observed and merged with the other data layers of a GIS ( 1998).   Another type of remote sensing device generating input for a GIS is videography, an advance on aerial photography. Video techniques using portable equipment produce geo referenced aerial video images in analogue or digital format. Airborne videography does not rely on special aircrafts and incorporates visual, infra-red, and thermal imaging. The advantages of this technology lie in its customization flexibility, the possibilities for data integration with other digital data such as scanned or digitized maps and satellite data, and its low cost compared with conventional aerial photography. The combination of these technologies provides important means of analysis and access to environmental information. The United Nations Environment Program (UNEP) contributes to the Global Resource ( 1998). Information Database (GRID), compiling and archiving geo referenced data. The GRID centers uses specialized ICT applications, such as remote sensing, and geographic information systems, and are building the expertise to prepare, analyze, and present environmental data. UNEP’s Environmental and Natural Resources Information Networking (ENRIN) program is supported by a satellite telecommunication system (1998).    With full Internet connectivity, the satellite communication system provides a high capacity means of delivering environment data and information for other UNEP initiatives such as GRID. The project is financed by a number of donor countries and implemented in collaboration with UNEP and the European Space Agency (ESA) ( 1998). After conducting a remote sensing, researchers use texture analysis to gather the different data they need in analyzing land cover and use. The texture analysis helps in showing an animated display of the different features of a certain area. It dissects certain features of an area. The first figure shows an example of a texture   Figure 1 Example of an image that underwent texture generation The figure shows a certain image that underwent texture modification. The figure marked (a) was generated by using the Discrete Markov random field model. The figure marked (b) was generated by using the Gaussian Markov random field model. The figure that was marked (c) was generated using the fractal model. Each method created different results for the figure depending on the procedure of each model.   Multi spectral Analysis LANDSAT has the widest spectral coverage of any remote sensing satellite. It covers seven wavelength bands ranging from the blue, green, and red of the visible spectrum down through three relatively near infrared bands which permit the identification of vegetation, some chemicals, and several types of minerals and band in the thermal infrared. Perceptive uses of the LANDSAT infrared capabilities enabled at least two different groups of investigators to identify a roughly thirty kilometer radius area around the Chernobyl explosion in which virtually all of the conifer forest had been killed or severely damaged by radiation. This was information not available in any way to Westerners in the Soviet Union and probably not even apparent to a scientist able to walk through the area ( 1989). The greater the number of color bands to which a satellite’s cameras are sensitive, and the greater the sharpness with which each color can be distinguished, the more useful the instrument is in detecting and identifying chemicals, heat emission, and chemical changes. To some extent, however, one must pay a technical price for the ability to see in color, and particularly to see in the extended color regions necessary for effective multi spectral analysis. Instruments which can see in many bands tend not to be able to see small objects or fine detail in larger ones with great clarity ( 1989).  In multi spectral analysis the focus is to gather data with regards to changes in the land cover, changes in geographical features and broadening of the area being searched. This kind of analysis best serves it purpose especially when no new geographical map is available.  The next figure shows an example of an image in a multi spectral analysis. Figure 2 Example of an image that underwent multi spectral generation The figure shows how an image would look like if generated in a multi spectral process. The figure projects different circumstances in the image. Each of the parts of the figure showed different transformation of the image depending on what the color system stands for.  The color system for each part of the figure can help researchers determine the change, the pattern of change and the effects of a change to a certain land area. Each part of the figure has different land area that put its focus on. Part 2 Land cover/ Land use Land cover Like those of GIS, the technologies and science of earth observation (EO) have been closely linked with the discipline of Geography. Since the 1960s, sensors mounted on-board Earth-orbiting satellites have offered an important source of data about the configuration and dynamics of the Earth’s surface. Among other things, EO data have serviced the mapping and monitoring requirements of the land surface, the atmosphere and the oceans (2004).  Many of the activities undertaken by the academic or professional geographers who make use of these data have, in the past, tended to be concerned with various attempts to transform them into representations of land cover, or with the change in these properties with respect to time; that is to say, the focus has been largely on aspects of image classification or, more specifically, the problems relating to the allocation of individual image pixels to discrete classes in pre-defined classification schemes (2004).   One of the problems in using EO data to provide information about urban areas is that the signal recorded by EO sensors actually only describes the amount of solar radiation reflected by the Earth’s surface, and its variation as a function of wavelength. While it is generally possible to derive representations of the spatial distribution of land cover from such data, obtaining information on land use, which is frequently of greater interest than simple land cover, is much more problematic. Put simply, there is no simple, consistent, relationship between the spectral reflectance of an individual pixel measured by a remote-sensing device and land use. This is particularly true in urban areas, where many different categories of land use may be composed of the same basic land cover types (2004).  The land cover is an important data needed by researchers to determine information about certain geographical areas that have value to them.   Land Use There are two major advantages in using a GIS to formulate wilderness evaluation databases. First, the approach is open-ended. New data may be added and current data modified. Information about access and land use is often poorly recorded and lacking in currency. Even the most recently available information may be inaccurate and out of date. This makes the compilation of a reliable database difficult, particularly because of the necessary dependence on published sources for much of the required information. Second, the process is spatially flexible, enabling scale to be matched to purpose. Furthermore, maps showing the distribution of wilderness identified in the inventory can be generated rapidly and efficiently in order to assist decision-making (2002).   The purpose of wilderness inventory has been to identify areas of wilderness quality for the possible enactment of conservation measures by government. Inventories provide a systematic means of ensuring the designation areas of high environmental quality. Recognition of wilderness is the necessary first step towards protecting, appreciating and managing wilderness areas. However, identifying an area as wilderness does not, by itself, ensure that its wilderness qualities can be maintained; this may only be done through the appropriate legislation and management. Decisions of this kind are inevitably judgmental, requiring comparative assessments of the social worth of alternative and often conflicting land use opportunities (2002). The land use is an important data to researchers because it helps them determining the next steps they will have to accomplish. It assists researchers in planning the next activities they will do on a certain area.   Part 3 Use of image texture in multi spectral analysis There is a very important use of image texture in multi spectral analysis of satellite images for classification of land cover or land use. Image texture when used in a multi spectral analysis of satellite image creates an image that can give more data and clearer visualization of the different aspects and properties of the generated image.  The image texture also creates mathematical equations that help in determining which among the features of the image can give the data needed for the accomplishment of the goals. With a better image and more data acquired classification of the land cover and land use can be easily done and with lesser errors. Classification of land cover and land use is important for governments because it helps them decides what actions, rules, services and changes should be given to a certain land area. Case study One case study related to the different discussions involves Kentucky and its changing landscapes. Kentucky’s landscapes are rapidly changing from agrarian, silvicultural, and coal-extractive dominance to more urbanized forms. Farmland is under increased development pressure; new homes, businesses, and shopping areas are being constructed on land that once produced crops and grazed cattle and horses. To be able to make good land-management and planning decisions for the future one needs good information on trends in landscape-use changes. The Kentucky Governor’s Office for Technology (GOT) was recently awarded a .3 million grant from the National Aeronautics and Space Administration (NASA) to improve the understanding of the Commonwealth’s forest and urban and rural landscapes (2003).    This grant money funds the Kentucky Landscape Snapshot Project, which will provide an updated, statewide land-cover map, a forest-cover map, and a high-resolution land-use map for many urban areas. The maps will reflect changes in Kentucky’s landscapes and will be useful to decision makers and land planners (2003).  Kentucky needed something that will help the state in making sure that it has a proper land management system. In this system the land cover and the land use will be classified and determined. The state believes that by knowing details about their land they can achieve the objectives set by the people and the government. Summary  The texture analysis helps in showing an animated display of the different features of a certain area. It dissects certain features of an area. The multi structural analysis helps in of the parts of the figure showed the different circumstances in the image. The multi structural analysis shows a more animated different transformation of the image depending on what the color system stands for. Image texture when used in a multi spectral analysis of satellite image creates an image that can give more data and clearer visualization of the different aspects and properties of the generated image.   Conclusion Traditionally, data analysts have turned to data mining techniques when the size of their data has become too large for manual or visual analysis. One method use in data mining is remote sensing for data gathering. This deals with the detection and measurement of phenomena without being in contact with them.  After conducting a remote sensing, researchers use texture analysis to gather the different data they need in analyzing land cover and use. The texture analysis helps in showing an animated display of the different features of a certain area.  If used with multi spectral analysis, texture analysis helps in providing an image that can be highly reliable in providing information on classification of land cover and land use. The combination of the two analysis gives a better chance of having accurate data that are needed by the government and other people involved in such study. Credit:ivythesis.typepad.com

The use of image texture in multi-spectral analysis of satellite images for classifying land cover/land use with respect to any one field of research

Introduction Preamble Since the launch of the LANDSAT 4 and 5 satellites in the early 1980s the use of satellite images in archaeologic...

Utilisation of Image Texture in Multi-spectral Analysis of satellite images for classifying land cover

Utilisation of Image Texture in Multi-spectral Analysis of satellite images for classifying land cover             Technology is rapidly advancing over the past years, various applications and discoveries have been obtained and studied. Multi-spectral analysis of satellite images is one of the new discoveries in this century and it stimulates new applications. Multi-spectral analysis poses a huge potential in the field of forestry, geology, and even agriculture. According to  (2007) multi-spectral analysis is the study of factual information in various spectral bands. (1984) discussed that in multispectral analysis, every Landsat scenes of 85×85 mi is separated into thousands of pixels which is around 1 acre in size wherein each of the pixel has an equivalent location within the scene. In addition, Wenger stated that multispectral analysis cannot be anticipated to stand alone and must be utilised with other pieces of information sources like the topographic maps and aerial photographs, moreover multispectral analysis gives reliable basic information for areas that are large in just a short span of time.             The objectives of this paper are the following: <ul style="margin-top: 0in; margin-bottom: 0in" type="disc"> <li class="MsoNormal" style="text-align: justify; line-height: 200%"> The main objective of this paper is to examine image texture in multi-spectral analysis of the satellite images of land cover. </li> <li class="MsoNormal" style="text-align: justify; line-height: 200%"> This paper also gives discussion on texture analysis and the various measures used in the image texture. </li> <li class="MsoNormal" style="text-align: justify; line-height: 200%"> And finally, a conclusion will be made.</li> </ul>   Image Texture             (1998) defined image texture as a function of the spatial differentiation in the intensities of the pixels; it is also very helpful in numerous applications and has been a subject of intense study and scrutiny by a number of researchers. ICVL (1999) discussed that texture models differentiate local spatial information in an image, moreover texture can be a useful tool in identifying things particularly in natural scenes such as the land cover, and however the texture of an image relies on factors such as scene geometry and illumination conditions.                Multi-spectral analysis uses various types of multi-spectral scanners in determining the vegetation mapping, fire mapping, and monitoring soil moisture. enumerated and discussed the instruments and these are: Multispectral Scanners- it is an optical-mechanical electronic device that observes the scene under an aircraft or satellite platform in a number of discrete bands of the ultraviolet, visible, and reflected, near and middle infrared portions of the electromagnetic spectrum. Wegner discussed that a normal multispectral scanner is composed of a rotating mirror and a telescope in order to emphasis radiation reflected from a small part of the surface of the earth on an array of detectors that are sensitive to energy. Every detector in the system observes the same element of resolution of the scene below however in different bands of wavelength.     Types of Multispectral Scanners             Airborne Multispectral Scanners- the commercial airborne multispectral scanners gather electromagnetic energy in 5-12 bands. The five most appropriately chosen bands are the most efficient wherein the two are the most visible, one is near the infrared, one middle infrared, and one thermal infrared. The airborne multispectral scanners could deliver spatial and spectral resolutions necessary to precisely map wildland resources; on the other hand this type of multispectral scanner is not economical. Landsat Multispectral Scanner- the launching of the Landsat-I in the year 1972 and Landsat-2 and 3 in the year 1975 and 1978 simultaneously, have given the resource managers a comparatively inexpensive multispectral scanner (MSS) data on a customary basis which is every 9 to 18 days. The data from the Landsat are duplicated in either on the digital form or the photographic form on magnetic tapes that are compatible in computers. The data from the Landsat Multispectral scanners are specified in five distinct bands of the electromagnetic spectrum. Wenger stated that the Landsat satellite multispectral data is helpful in the following areas: <ul style="margin-top: 0in; margin-bottom: 0in" type="disc"> <li class="MsoNormal" style="text-align: justify; line-height: 200%"> Vegetation mapping in wider classes.</li> <li class="MsoNormal" style="text-align: justify; line-height: 200%"> Stratification for sampling in broad renewable natural resource inventories. </li> <li class="MsoNormal" style="text-align: justify; line-height: 200%"> Mapping defoliation.</li> <li class="MsoNormal" style="text-align: justify; line-height: 200%"> Mapping and estimating clear-cut areas. </li> <li class="MsoNormal" style="text-align: justify; line-height: 200%"> Mapping the surface water. </li> </ul>   Aside from the Landsat Multispectral Scanner, another Landsat instrument is being utilised by geographers, researchers, etc. and this is the Landsat Thematic mapper (TM) which is a multispectral satellite that measures the electromagnetic energy in seven spectral bands that is composed of seven spectral bands that extends from visible to thermal infrared and each pixel indicates an area of 30 m by 30 m in six out of the seven bands and the pixels in the thermal band indicates an area of 120 m by 120 m ( 2001). Figure 1: Landsat Thematic mapper image of Jackson Purchase area of western Kentucky (2001). Texture Analysis             Texture analysis is the segmentation or classification of textural features with consideration to the shape of the small element, density and direction of prevalence ().              (1998) discussed that the model based texture analysis techniques’ foundations are on the development of an image model that could be utilised to describe and synthesise the texture. Motivation              discussed that texture analysis is an important aspect in the study in machine vision. Texture perception is important in the view of human vision and view of practical applications of machine vision. Human Vision- a number of psychologist studies the texture perception of the human vision because the performance of different texture algorithms is examined against the performance of the visual system of human beings. A number of researches in psychophysiology have pointed out that a multi-channel, frequency and analysis of the location of the image that is shaped on the retina and is completed through the brain.             One of the most widely recognised studies in texture perception of humans is from  due to the fact that he focused on the spatial statistics of the grey levels of the images. The concept of firs and second order spatial statistics according to  (1981): First-order statistics- it measures the possibility of noticing a grey value at the haphazardly selected location in the image, it can be calculated from the histogram of pixel concentration in the image. The average concentration in an image is an example of the first-order statistics. Second order statistics- it is the possibility of noticing a pair of grey values that takes place at the endpoints of a dipole of random length in the image at a random location and position.   Machine Vision- The techniques in texture analysis have been used in different fields, in older domains texture analysis already plays a huge role. Texture analysis plays a huge role in automated inspection, medical image processing, document processing, and remote sensing. Automated inspection- the applications of texture processing in automated inspection are limited which includes detection of deficiency in the images of textile, carpet wear, and paints in automobiles. In determining the defects in the texture of the images the applications focused more on inspection of the textile. An example is in the study of  (1988) the researchers’ utilised signal processing methods to identify the defects in the texture images. In carpet wear  (1988) suggested that to assess the carpet wear simple texture features calculated from the second-order grey level dependency statistics must be utilised. And from this technique the numerical texture feature acquired can characterise the wear of the carpet in a successful manner. Medical Image Analysis- the methods in image analysis have a very significant role in medical use which includes the automatic extraction of features from the image wherein it is very useful in various tasks. The selected features seize the morphological properties of the images such as colour and certain textures. A number of medical studies have utilised texture analysis in identifying the images such as the study of (1972) wherein the researchers were able to classify the pulmonary diseases by the features of the texture of the image. Moreover, the study of  and  (1978) selected different first-order statistics and second order statistics to distinguish various types of white blood cells.  Document Processing- one of the most useful applications of image analysis is one the document processing which ranges from recognition of the postal address to the interpretation and investigation of maps. The basis of image analysis methods suggested for document processing are the characteristics of the printed documents and tries to identify the benefits could be gain from it. Remote Sensing- in remote sensing, texture analysis has been used widely such classification of lands to identify the various forms of terrains. In the study of  the researchers used grey level  co-occurrence features of the images to study the remotely sensed images.             There are three types of measures utilised by researchers, geographers, agriculturists, etc. in the image texture and these are the statistical measures, spectral measures, and structural measures. Statistical Measures- according to Hogg in the statistical methods, the grey level-histogram, statistics based in grey-level occurrence matrix are calculated in order to make a clear distinction on the different textures of the image. In the Statistical Approach have four types according to Hogg and these are: First Order Histogram- According to the first order histogram describes the regularity of presence of each grey level in a local area. The most widely used statistical measures are the average, variance, entropy, and coefficient of variation.  Average:                                                                   Variance:                               Entropy:                                                                    Co-efficient of variation:                 fi =  frequency of grey level i that occurs in a pixel window quantk= quantization level of the image W= total number of pixels in a window.             Second-order grey level-co-occurrence matrix- the second-order set of texture measures which is based on the brightness value spatial-dependency grey-level co-occurrence matrices (1973). The six most widely used measures are the angular second moment, the contrast, the correlation, the entropy, the inverse difference moment, and the covariance. Angular second moment:                                                   Contrast: Correlation:                                                                             Entropy: Inverse difference moment:                                              Covariance:      quantk= quantization level of the image hc(i,j)= the (i,j)th entry in one of the grey level co-occurrence matrix             Third order grey level co-occurrence matrix- it is a higher-order of texture measures that was based on third-order histograms, the measures used for this technique are the angular second moment contrast, correlation, entropy, the inverse and the covariance difference moment. Angular second moment:                        Contrast: Correlation: Entropy: Inverse:   Covariance: P(i, j, k) = the (i, j, k)th entry of the third order grey level co-occurrence matrix quantk=quantization level of the image             Grey level difference vector- it is the sum of the diagonals of the grey-level co-occurrence matrices. The measures used for this technique are angular second moment, contrast, correlation, entropy, inverse difference moment, and variance. Angular Second Moment:                                          Contrast:                                          Correlation:                                                            Entropy:                                              Inverse difference moment:                                    Variance: hc=(m) = the probability that a pair of brightness values having an absolute difference of m occurs at separation c.             In the study of (1989) in classifying the land cover using the statistical measure there were three difference statistics and these are the intersample spacing distance, Band, and Features. stated that the statistical measures are capable for identifying the information regarding the texture of the land cover; on the other hand the simple introduction of the information on texture by statistical measures in the spectral information could not make the classification accurate.       Spectral Measures             Aside from the statistical measures, researchers also utilize the spectral approaches in analysis of the texture. In this approach, Hogg discussed that the textured image is changed into frequency domain, afterwards the extraction of texture features could be done through analysis of the power spectrum.  The spectral measures have three classifications and these are the Fourier Transform, Gabor filter bank, and wavelet decomposition. A tuned bandpass filter bank exhibit similarities to the pattern of the neural receptive fields in the visual system of human beings (1982).             Spectral methods have feature extraction in to the domain of the spatial frequency wherein it has a number of advantages such as it enhances only definite features while subdues others and the cyclicity pattern of a texture could symbolise explicitly in fixed spatial frequency in the spectral.              (2001) discussed that in structural measures of analysis of the image texture filter banks and image pyramids are often utilised to change an image from the spatial area into the area of frequency and conversely. Just like in the statistical method, the spectral method also uses measures such as the wavelength coefficient for the transformation of the wavelet.             Fourier Transform- (1995) it is a calculation needed to see a wave not only in the time domain but also in the frequency domain.  (1985) stated that the fourier transform is a generalization of the complicated Fourier series in the limit as . The common Fourier transform pairs according to Arfken are the following: Function Fourier transform–1 1 Fourier transform–Cosine Fourier transform–Delta function Fourier transform–Exponential function Fourier transform–Gaussian Fourier transform–Heaviside step function Fourier transform–Inverse function Fourier transform–Lorentzian function Fourier transform–Ramp function Fourier transform–Sine   Gabor Filter Bank-  (2005) discussed that the Gabor filter is acquired through modulation of a sinusoid with a Gaussian. A Gabor filter is a straight, narrow filter that has its impulse reaction regulated by a Gaussian envelope (1946). Wavelet decomposition- it can be interpreted as signal decomposition in a set of independent, spatially oriented frequency channels ( 1989). Structural Measures             It is utilised to create more complicated texture pattern through grammar rules that indicate the generation of the pattern of the texture. Structural measures are one of the main methods utilised to develop syntactic models for textures. The structural approaches try to obtain geometrical representation which is founded on the idea that the texture is seen as a spatial organisation of texture elements. The structure and the spatial organisation of the image texture is one of the important elements in the outcome of models. In addition, the structural measures use different spatial analytical methods in discovering the cyclicity and the regularity of the textures so that the geometric patterns and placement rules of the components of the texture.             A number of studies have utilised more complex instruments in mapping and classification of the land cover. In the study of  (2000) the addition of image texture in the spectral classification of the eco regions in Canada. The results have shown that by adding image texture in the spectral classification it has increased the accuracy of classification in high spatial detail imagery.             (2003) also found out in his study that the texture enhances the classification accuracy and the measures that performed the best were the mean intensity, variance, weighted-rank fill ration and semivariogram.             Moreover,  (1999) found out that the textural measure that is based in the grey level co-occurrence matrix of the statistical measures have shown more efficient spatial forms of the landcover and landuse in multispectral images.   Conclusion             Adding the image texture in multi-spectral analysis of images captured by satellites is a breakthrough especially in the field of geosciences wherein aerial photographs are important in order to identify the land cover which will aid in the development of the area. However, there are image texture has its own share of disadvantages on multi-spectral analysis. Like in the study of , if the window size is not appropriate it will have a negative impact on the textural feature of the image. Yet, a number of studies have proven that by adding image texture in the satellite images the classification of land covers become more accurate, therefore image texture could be a useful tool in analysing and classifying the satellite images of land cover. Credit:ivythesis.typepad.com

Utilisation of Image Texture in Multi-spectral Analysis of satellite images for classifying land cover

Utilisation of Image Texture in Multi-spectral Analysis of satellite images for classifying land cover Technology is rapidly...

Top