外文翻譯---基于模糊邏輯技術圖像上邊緣檢測（英文）

1、Edge Detection in Digital Images Using Fuzzy Logic Technique Abdallah A. Alshennawy, and Ayman A. AlyAbstract—The fuzzy technique is an operator introduced in order to simulate at a mathematical level the compensatory

2、 behavior in process of decision making or subjective evaluation. The following paper introduces such operators on hand of computer vision application. In this paper a novel method based on fuzzy logic reasoning str

3、ategy is proposed for edge detection in digital images without determining the threshold value. The proposed approach begins by segmenting the images into regions using floating 3x3 binary matrix. The edge pixels ar

4、e mapped to a range of values distinct from each other. The robustness of the proposed method results for different captured images are compared to those obtained with the linear Sobel operator. It is gave a permane

5、nt effect in the lines smoothness and straightness for the straight lines and good roundness for the curved lines. In the same time the corners get sharper and can be defined easily. Keywords—Fuzzy logic, Edge detect

6、ion, Image processing, computer vision, Mechanical parts, Measurement. I. INTRODUCTION VER the last few decades the volume of interest, research, and development of computer vision systems has increased enormously.

7、Nowadays they appear to be present in almost every sphere of life, from surveillance systems in car parks, streets, and shopping centers, to sorting and quality control systems in the majority of food production. Thu

8、s, introducing automated visual inspection and measurement systems are necessary, specially for the two dimensional mechanical objects, [1:8]. In part due to the substantial increase in digital images that are pr

9、oduced on a daily basis (e.g., from radiographs to images from satellites) there is an increased need for the automatic processing of such images, [9,10,11]. Thus, there are currently many applications such as comput

10、er-aided diagnosis of medical images, segmentation and classification of remote sensing images into land classes (e.g., identification of wheat fields, and illegal marijuana plantations, and estimation of crop growth

11、), optical character recognition, closed loop control, content-based retrieval for multimedia applications, image manipulation for the film industry, identification of registration details from car number plates, and

12、 a host of industrial inspection tasks (e.g., detecting defects in textiles, rolled steel, plate glass, etc.). Historically much data has been generated as images to Abdallah A. Alshennawy, Assistant Professor, Desig

13、n and prod. Eng. Dept. Tanta University, Egypt (e-mail: abd_alshennawy@yahoo.com). Ayman A. Aly, Associate Professor, is with Mechatronics section, Assiut University, 71516, Egypt (e-mail: ayman_aly@yahoo.com). facili

14、tate human analysis (it is much easier to understand an image than a comparable table of numbers), [12]. And so this has encouraged the use of image analysis techniques over other possible methods of data processing.

15、In addition, since humans are so adept at understanding images, image based analysis provides some aid in algorithm development (e.g., it encourages geometric analysis) and also helps informally validate results. Wh

16、ile the role of computer vision can be summarized as a system for the automated (or semi- automated) analysis of images, many variations are possible, [9,13]. The images can come from different modalities beyond norma

17、l gray-scale and colour photographs, such as infrared, X-ray, as well as the new generation of hyper-spectral satellite data sets. Second, many diverse computational techniques have been employed within computer visio

18、n systems such as standard optimization methods, AI search strategies, simulated annealing, genetic algorithms, [14,15, 16]. Usage of specific linear time-invariant (LTI) filters is the most common procedure applied

19、 to the edge detection problem, and the one which results in the least computational effort. In the case of first-order filters, an edge is interpreted as an abrupt variation in gray level between two neighbor pixels.

20、 The goal in this case is to determine in which points in the image the first derivative of the gray level as a function of position is of high magnitude. By applying the threshold to the new output image, edges in a

21、rbitrary directions are detected. In other ways the output of the edge detection filter is the input of the polygonal approximation technique to extract features which to be measured, [1]. A very important role is p

22、layed in image analysis by what are termed feature points, pixels that are identified as having a special property. Feature points include edge pixels as determined by the well-known classic edge detectors of PreWitt

23、, Sobel, Marr, and Canny [17:21]. Recently there has been much revived interest [22,23] in feature points determined by “corner“ operators such as the Plessey, and interesting point operators such as that introduced b

24、y Moravec. [24,25] Classical operators identify a pixel as a particular class of feature point by carrying out some series of operations within a window centered on the pixel under scrutiny. The classic operators wor

25、k well in circumstances where the area of the image under study is of high contrast. In fact, classic operators work very well within regions of an image that can be simply converted into a binary image by simple thr

26、esholding as shown in Fig.1. To be definite as to the failings of classic operators: classic edge detector tends to give poor results for labeling edge pixels, when an edge, although O World Academy of Science, Enginee

27、ring and Technology 51 2009178“white”. The adopted membership functions for the fuzzy sets associated to the input and to the output were triangles, as shown in Fig.3. (a) (b) Fig. 3 Membership functions of the fuzzy

28、sets associated to the input and to the output ERule1 ERule3 ERule2 Rule1 If {(i-1, j-1 ) & (i-1, j) & (i-1, j+1) } are whites If {(i, j-1) & (i, j) & (i, j+1) } are whites If {(i+1, j-1)

29、 & (i+1, j) & (i+1, j+1)} are blacks checked pixel is Edge Rule2 If {(i-1, j-1 ) & (i-1, j) & (i-1, j+1) } are blacks If {(i, j-1) & (i, j) & (i, j+1) } are whites If {(i+1, j-1

30、) & (i+1, j) & (i+1, j+1)} are whites checked pixel is Edge Rule3 If {(i-1, j-1 ) & (i, j-1) & (i+1, j-1) } are blacks If {(i-1, j) & (i, j) & (i+1, j) } are whites If {(i-1, j+

31、1) & (i, j+1) & (i+1, j+1)} are whites checked pixel is Edge Rule4 If {(i-1, j-1 ) & (i, j-1) & (i+1, j-1) } are whites If {(i-1, j) & (i, j) & (i+1, j) } are whites If {(i-1, j

32、+1) & (i, j+1) & (i+1, j+1)} are blacks checked pixel is Edge (a) ERule4 ERule5 ERule7 ERule6 Rule5 If {(i-1, j) & (i-1, j-1) & (i, j-1) & (i+1, j-1)} are blacks If {(i-1, j+1) & [i, j+1] &

33、amp; (i+1, j+1) &(i+1, j)} are whites If (i, j) is white checked pixel is Edge Rule6 If {(i-1, j) & (i-1, j-1) & (i, j-1) & (i+1, j-1)} are whites If {(i-1, j+1) & [i, j+1] & (i+1, j+1) &

34、 (i+1, j)} are blacks If (i, j) is white checked pixel is Edge Rule7 If {(i-1, j-1) & (i, j-1) & (i+1, j-1) & (i+1, j)} are blacks If {(i-1, j) & (i-1, j+1) & (i, j+1) & (i+1, j+1)} are whit

35、es If (i, j) is white checked pixel is Edge Rule8 If {(i-1, j) & (i-1, j+1) & (i, j+1) & (i+1, j+1)} are blacks If {(i-1, j-1) & (i, j-1) & (i+1, j-1) & (i+1, j)} are whites If (i, j) is whi