Research Topics: Multi-spectral Analysis

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...