A Hybrid Technique for Image Retrieval Using Canny and Neural Network

Today image retrieval is big area of research. Specially, over internet it is widely used. Edge Detection plays an important role to detect edges. An edge of the object /Image can be used to define the shape and structure of image. Shape is an important feature of image through which the images can be browsed from large image database. The novel hybrid method based on the Canny Algorithm and Neural Network is proposed for image retrieval to get the image at very fast rate. The hybrid architecture is used to take the advantage of more than one technique in one umbrella. Image pre processing is done before applying the neural network. In image processing, image Smoothing is performed using Gaussian Filter and Fuzzy Technique applied to convert Gray level image in binary image. Finally, neural network is designed and trained for detecting actual edges. Neural network is a wonderful tool to work on real world data. Error Back-Propagation classification technique of neural network is to be used to detect proper edges.