Multi Spectral Image Enhancement in Satellite Imagery - Acsijci.org-Ag

ACS-International Journal on Computational Intelligence, Vol-1, Issue-1 Aug 2010 13

Multi Spectral Image Enhancement in Satellite Imagery1J.ASHOK, 2E.G.RAJAN

1Professor and Head , Department of Information Technology, GCET, Hyderabad,India.2President,Pentagram Research Center, Hyderabad,India.E-mail:[email protected],[email protected]

ABSTRACT

This paper describes a GUI based multi spectral image enhancement used to achieve highly real-

digital image populates only a small portion of the available range of digital values. Image enhancementinvolves changing the original values so that more of the available range is used; this then increases thecontrast between features and their backgrounds. It consists of reading the binary image on the basisof pixels taking them byte wise and displaying it, calculating the statistics of an image, automaticallyenhancing the color of the image based on statistics calculation using algorithms and working with RGBcolor bands. Finally the enhanced image is displayed along with image histogram.

Keywords: Image enhancement, digital satellite imagery, histogram

1. INTRODUCTION

For every digital image the pixel value represents the magnitude of an observed characteristic suchas brightness level. An image histogram is a graphical representation of the brightness values that com-prise an image[1]. The brightness values are displayed along the x-axis of the graph. The frequency ofoccurrence of each of these values in the image is shown on the y-axis.

The enhancement techniques are employed in order to increase the contrast and brightness of an im-age using RGB color bands. Therefore, the distinction of features in the scene can be easily performed

an image can be enhanced by spreading out the range of scene illumination.

2. ANALYSIS

The analysis of this paper has been done mainly on

Reading the binary image on the basis of pixels, taking them byte wise and displaying the raw image.

The image is displayed along with the histogram.

Calculating the statistics of the image

It includes calculating mean, median, mode and standard deviation for the image based on the pixel values.

Automatically enhancing the color of the image based on statistics calculation and working with RGB color bands.

for later usage.

14 ACS-International Journal on Computational Intelligence, Vol-1, Issue-1 Aug 2010

3. DIGITAL IMAGE

In a generalized way, a digital image is an array of numbers describing spatial distribution of a certain

data or topographical elevation or radar return. It is a two dimensional array of brightness value whereeach number in the array represents the value of the parameter at that point or over a certain unit area(Fig). The smallest unit that constitutes the picture is a picture element called pixel. Each point in theimage or scene is represented by an integer digital number (DN) usually 256 (0 to 255), normally in thedynamic range of the display device the lowest intensity of the parameter where no signal is assigned toDN value zero and the highest intensity or largest signal to 255, the intermediate intensities receivingappropriate intermediate DN value[2].

3.1 DATA FORMATS OF DIGITAL SATELITE IMAGERY

Digital data from the various satellite systems supplied to the user in the form of DVD or CD-ROM.As no worldwide standard for the storage and transfer of remotely sensed data has been agreed upon,

Fig. Structure of a digital image

J.ASHOK, E.G.RAJAN


through the CEOS (Committee on Earth Observation Satellites) format is becoming accepted as thestandard. Digital remote sensing data are often organized using one of the three common formats usedto organize image data. For instance an image consisting of four spectral channels, which can be visual-ized as four superimposed images, with corresponding pixels in one band registering exactly to those inthe other bands. These common formats are:

Band Interleaved by Line (BIL)

Band Interleaved by Pixel (BIP)

Band Sequential (BSQ)

Digital image analysis is usually conducted using Raster data structures – each image is treated as anarray of values. It offers advantages for manipulation of pixel values by image processing system, as it

represent the array of pixels as discrete patches or regions, where as Vector data structures uses polygonalpatches and their boundaries as fundamental units for analysis and manipulation though vector formatis not appropriate for digital analysis of remotely sensed data.

Images are used in various applications such as remote sensing, bio-medical engineering, forensicsciences, astronomy, non-destructive testing etc. These images can be obtained from the data collectedfrom various satellites, scanners etc. In all these applications, image processing is required for easy inter-pretation of images and also to extract more information. Image Processing in its general form pertainsto the alteration and analysis of pictorial information of image data[3].

together with lists of typical processing routines. Image restoration compensates for data errors, noise,and geometric distortions introduced during the scanning, recording, and playback operations.

Restoring periodic line dropouts.

Restoring periodic line striping.

Filtering of random noise.

Correcting for atmospheric scattering.

Correcting geometric distortions.

Image enhancement alters the visual impact that the image has on the interpreter in fashion that im-proves the information content.

Contrast enhancement.

Intensity, hue and saturation transformations.

Density slicing.

Edge enhancement.

Making digital mosaics.

Producing synthetic stereo images.

Multi Spectral Image Enhancement in Satellite Imagery


Information extraction utilizes the decision-making capability of the computer to recognize andclassify pixels on the basis of their digital signatures. The common events in image processing are thecapture, storage, enhancement and interpretation of information in an image.

Producing principal-component images.

Producing ratio images.

Producing change-detection images.

The term digital image processing generally refers to processing of a two-dimensional picture by acomputer. In a border context, it implies digital processing of any two-dimensional data. A digital image

This form of Image Processing was brought about by the advent of the computer. The data converted-

ciple advantage of digital image processing methods is its versatility, repeatability and the preservationof original data precision.

4. IMPLEMENTATION

for the image reading. These inputs are number of pixels, number of lines, and number of bands of theselected image.

4.2 Image Reading

Image reading is a big process where each and every pixel of the image is read byte wise. the main

meaning of the image will not be changed.

4.3 Displaying the Image

4.4 Histogram of Raw Image

The histogram is a graphical representation of an image in which, x-axis consists of intensity and y-axis comprising of frequency of the brightness values. Histogram for raw image is obtained by taking a

initialized to zeros and are incremented for every repetition of the intensity value. In this way, the histogram

J.ASHOK, E.G.RAJAN


array will be obtained. This histogram is displayed by using graphics in java, using the methods paintand repaint and by considering the scale. In this process, the maximum value of the pixels is calculated

displayed according to the scale[5].

4.5 Statistics

Statistics include calculating mean, median, mode, standard deviation, minimum and maximum val-ues according to the pixel values. These statistics are very important as the enhancement is done basedon these values.

4.6 Min-Max Enhancement

The enhancement is done based on the minimum and maximum values of the pixels. Where the zerovalue of the pixels are replaced by the minimum value and the highest value becomes the maximumvalue there by increasing the intensity ranges making the image more clear.

4.7 Standard Deviation Enhancement:

In standard deviation enhancement, the minimum value is standard deviation subtracted from themean and maximum value is standard deviation subtracted from maximum value. The enhancementdone using standard deviation is more accurate as the zero probable values are replaced by the calculatedrange of standard deviation values[6].

4.8 Enhanced Histogram

The enhanced histogram values are obtained similar to the previous procedure and it is displayedusing the same methods with changed parameters.

4.9 Image Properties Window

The image properties of window are a frame containing the image name, number of pixels, lines,bands. It also consists of statistics of the image.

4.10 Raw Pixel Values

The raw pixel values is a frame containing a table of two-dimensional image represented in the formof pixel table for every band i.e. red, green and blue respectively.

4.11 Enhanced Pixel Values

Similarly, the enhanced pixel values frame consists of two-dimensional representation of image inthe form of pixels for every red, green and blue bands.

5. TESTING

5.1 Real Data Testing

If we use real, live customer data in your testing and development of applications, we may want tothink twice about the risks of exposing the data.

Organizations that use live data in their testing do so basically because it makes the testing more realworld and better puts the application through its paces. Trouble is, it also can expose sensitive data to



real thing in application testing and

5.2 Validation Testing

Test validation is a procedure that demonstrates that a test is job-related and correlates to on-the-jobperformance [8]. There are three types of validations

a) Content Validation

Content Validation is a procedure where the content of the job in question is analyzed, matched andcompared, (function by function) to the abilities measured by the test. This shows that the test is relatedto the job.

b) Concurrent Validation

Concurrent Validation is a procedure, which tests individuals currently in the position, and then sta-tistically compares their results with their current performance.

c) Predictive Validation

Predictive Validation is a procedure, which tests individuals when they are hired, and then statisti-cally compares the test results, after a certain time, to on-the-job performance. A successful comparisonor correlation suggests that the test is predictive of on-the-job success. For a good predictive study, alarge application pool and considerable time and effort are required. A suggested scenario would be totest all applicants as they are hired but not use test results for selection[9].

At the end of a period, say one year, a comparison is made between the test results and on-the-jobperformance criteria.

Existing System:

The existing system in NRSC is not automatic and the images are enhanced manually which is timeconsuming.

Proposed System:

The present module is automatic in which images are enhanced with in no time and the enhancementis correct as it uses statistics for enhancement [10].

6. RESULTS

The results of this work are carried out in several phases. The following is the sequence of the proce-dure involved in getting the results.

J.ASHOK, E.G.RAJAN


Step 1:

As a result we obtain the raw image and histograms which are shown below.

Step 2: Min-Max Enhancement is done and enhanced image along with enhanced histograms are dis-played as shown below.

Fig 1. Raw image and its Histogram

Fig.2 Enhanced image and its Histogram

Step 3 : Standard deviation Enhancement is done and enhanced image along with enhanced histogramsare displayed as shown below.



7. CONCLUSION

The system facilitates processing of raw images which are obtained from satellites. This module isdeveloped in order to attain highly realistic visualizations of raw satellite imagery with all the featuresmentioned earlier. In this whole process, the user is also provided with an option of manipulating the im-age manually. Raw pixel values before enhancing the image and enhanced pixel values after enhancingthe image are displayed in tabular format for reference.

8. REFERENCES1. Digital Image Processing-An Algorithmic Introduction Using Java by Wilhelm urger and Mark J.Burge

2. Image Processing-Principles and Applications by Tinku Acharya and Ajoy K.Ray

3. Digital Image Processing by Rafael C.Gonzalez and Paul Wintz

4. http://en.wikipedia.org/wiki/Image_Processing

7. http://www.owlnet.rice.edu/~elec539/projects97/morphjrks/moredge.html

8. http://www.pages.drexel.edu/~weg22/can_tut.html

9. http://javastuff.dev.java.net/code/operations.Binarize.html

10. http://forums.sun.com/thread.jspa?threadID=5378450&tstart=0

IJCI-2K10-02

Fig 3. Enhanced image and its histogram using Standard deviation

J.ASHOK, E.G.RAJAN

http://en.wikipedia.org/wiki/Image_Processing

http://www.owlnet.rice.edu/~elec539/projects97/morphjrks/moredge.html

http://www.pages.drexel.edu/~weg22/can_tut.html

http://javastuff.dev.java.net/code/operations.Binarize.html

http://forums.sun.com/thread.jspa?threadID=5378450&tstart=0

Documents

Multi Spectral Image Enhancement in Satellite Imagery - Acsijci.org-Ag