Edge detection algorithms are important tools in image

processing applications for carrying out much information and

being relatively easy to produce. Sobel Canny and logarithmic

algorithms [1] are among several edge detection algorithms

used frequently nowadays. The ution of such edge detection

algorithms is an old problem. Authors [1][3] tend to use visual

uation that limits the comparison between different edge

images. In this paper, we present a new edge enhancement

method and five different measures that can be used to

statistically uate edge detection algorithms. The new edge

enhancement method is based on cooperation between different

edge detection algorithms. The new edge preserves the

advantages of each edge image. Experimental results using two

edge detection algorithms proved the efficiency of this method.-Edge detection algorithms are important tools in image

processing applications for carrying out much information and

being relatively easy to produce. Sobel　Canny　and logarithmic

algorithms [1] are among several edge detection algorithms

used frequently nowadays. The　ution of such edge detection

algorithms is an old problem. Authors [1][3] tend to use visual

uation that limits the comparison between different edge

images. In this paper, we present a new edge enhancement

method and five different measures that can be used to

statistically　uate edge detection algorithms. The new edge

enhancement method is based on cooperation between different

edge detection algorithms. The new edge preserves the

advantages of each edge image. Experimental results using two

edge detection algorithms proved the efficiency of this method.