Presenting a method based on automatic image annotation techniques for the semantic recovery of images using artificial neural networks

With the ever-increasing growth of the Internet and the digital imaging industry, the need to organize and separate images is strongly felt. As a result, image databases with very large sizes were created. In such a situation, there is a strong need for effective tools and methods for image search and recovery. In this proposed method, an automatic image delineation method using an artificial neural network has been presented. At first, by studying reference books and articles related to the basic concepts that are necessary to start working in this field, then by studying the articles of recent years in the desired field, we will examine the strengths and weaknesses of the subject in more detail. Finally, by implementing and testing various ideas that have been expressed in the articles, and with the guidance of the supervisor and applying his ideas, he has tried to cover the weaknesses of other ideas, and finally, a method to improve the accuracy of image indexing and retrieval methods in databases based on the content of images to achieve high efficiency and reduce the semantic gap between low-level features and human perceptual concepts. The proposed color extraction method is simpler, less computationally complex, more accurate and faster. The results of evaluations and comparisons indicate the relative superiority of the proposed method over other methods.

Keywords:

Automatic image , Annotation techniques , Semantic recovery , Artificial Neural Networks