An Adaptive Edge Detecting Method for Satellite Imagery Based on Canny Edge Algorithm

— Satellite images are unclear and it is very difficult to get information from them. This paper deals with the detection of edges of a satellite image. Here, edge detection is the fundamental tool of image segmentation. Image segmentation is a process of dividing an digital image into set of pixels, it is used to identify the objects and boundaries un an image.A set of connected pixels which forms boundary between two disjoint regions defines an edge,which is very important to acquire information from an image.There are many segmentation techniques like threshold, clustering, PDE, ANN based techniques, of all these methods edge based technique is the most optimum one. Canny edge detection algorithm is used to detect a wide range of edges in image uses multi-stage algorithm. Low error rate,good localization and single response are the main features of this algorithm.


INTRODUCTION
Image processing is used to analyze and manipulate the images based on different purpose, the output of the image processing may be the image or any characteristic of that image. Edge detection is one of the well-developed field in image processing .This includes a variety of mathematical methods which are used to identify the points in digital image.There are many edge [3] detecting techniques like clustering, thresholding, compression based methods etc.,Most of the information about images will be found in detecting the edges. So, we detect the edges in image using filters which enhances some areas of image which contain edges, thereby sharpness of the image [1] will increase and image will appear clear.

IMAGE SEGMENTATION:
Image segmentation is a process of dividing the digital image into very smallest parts called pixels or image objects. If the pixel of an image is high, the higher the quality of the image. The three basic categories [9][10] of image segmentation are A. Structural segmentation techniques: This technique relies upon the required region which is to be segmented. B. Stochastic segmentation techniques: These techniques works on the discrete pixel values of the image C. Hybrid techniques: These techniques uses discrete pixel and structural information together [11].

BLOCK DIAGRAM DESCRIPTION:
Step 1: Computing the horizontal (Gx) and vertical (Gy) gradient of each pixel in an image.
Step 2: Using the above information the magnitude (G) and direction (of the each pixel on the image is calculated.
Step 3: In this step all non-maxima's are made zero and suppressed.
Step 4: The high and low thresholds are measured using the histogram of the gradient magnitude of the image.
Step 5: To get the proper edge map hysteresis thresholding [5][6][7] is employed which will link between the weak and strong edges. The weak are taken into consideration if and only if it is connected to one of the strong edges or else it is eliminated from the edge map.

SATELLITE IMAGES:
 Satellite images give a good representation of what is happening at every point in the world, especially overoceans where large gaps in data occur. Color images are due to the grey level [13].  The resolution of satellite images varies depending on the instrument used and the altitude of the satellite's orbit.  Earth's surface at various spectral, temporal, radiometric, and increasingly detailed spatial resolutions as is determined by each collection system's sensing device, and the orbital path of its reconnaissance platform are depicted by satellite imagery.

International Journal of Advanced Engineering Research and Science (IJAERS)
[Vol-7, Issue-4, Apr-2020] https: //dx.doi.org/10.22161/ijaers.74.37  ISSN: 2349-6495(P) | 2456-1908(O) www.ijaers.com Page | 295  /dx.doi.org/10.22161/ijaers.74.37  ISSN: 2349-6495(P) | 2456-1908(O) www.ijaers.com Page | 296  It is adaptive in nature, less sensitive to noise, resolved the problem of streaking, provides good localization and detects sharper edges as compared to others. Canny edge detection technique is used in license plate reorganization system finds practical application in traffic management, public safety and military department. Satellite images are not very clear and information can be easily obtained from detecting the edges by using canny edge detection algorithm which is a part of segmentation process.