مطالعه آشكارسازي لبه در پردازش تصوير با استفاده از شبكه عصبي پيچشي, By Hossein Hosseinzadeh

Research

Title	A study of edge detection in image processing using convolutional neural network
Type	Thesis
Keywords	پردازش تصوير، شبكه عصبي پيچشي، آشكارسازي لبه، پردازش تصوير
Researchers	Fatemeh Foladi (Student) , Hossein Hosseinzadeh (Primary advisor) , Ahmad Shirzadi (Primary advisor) , zeinab Sedaghatjoo (Advisor)

Abstract

Edge detection is one of the basic steps in image processing that has many applications. So far, many and varied methods have been presented for edge detection, but a comprehensive and applicable method in various fields and different types of edges has not been presented. The aim of this thesis is to present a method for edge detection using convolutional neural network. The efficiency of this proposed method has been evaluated on two databases named Train and Test. The Train database consists of 10,000 color photo models along with the mask corresponding to the color photo (the masks are all manually edged photos from the color photo) for training the network.