The Research And Application Of Edge Detection To Text Region In The Image

In the recent years,great progress has been made in Edge Detection technology. The edge obtains a lot of useful information,and applying an edge detector to an image may significantly reduce the amount of data to be processed,while preserving the important structural properties of an image.So the research of edge detection attracts more and more professors put a great deal of time and effort into this field.Now due to the handwriting is various and complicated,the technologies in image processing of static handwriting image are developed slower than both print hand image and real-time handwriting image.The author studied and compared some different typical detection methods, especially in Canny edge detection operator and mathematical morphology.Based on Mathematical morphology,an application is made to detect the text region edge in a image,which costs a short time to progress.Based on the research in this paper,the author develops software by VC++ to implement the edge detection.The results of experiments indicate that the effect of the software to detect text region edge is well and effective while reducing noises during progressing.Another feature of the application is allowed user to customize the parameters while the application running.The result the change would be shown directly to the user.If the user doesn't satisfy the effect of the result,he/she would keep changing these parameters until it shows perfect.This application is also applied to the text region detection of the static handwriting image,which also shows a well effect. The author suggests that sharing the research in this paper would make some contribution to the image processing and computer vision.