The Importance Of Image Retrieval

In the last few years, the fast growth of the Internet has largely increased the number of image collections available. The accumulation of these image collections is attracting more and more users in various professional fields [Rui Y et al., 1999]. Conventional methods are used to retrieve the desired images by matching keywords manually that are assigned to each image . However, as a result of the large number of images in collection, manual processing has become practically impossible. Since users are unlikely to foresee all the query keywords that will be used in a matching process, assigning keywords to every image is impractical, which makes the effectiveness of the classic image retrieval very limited.

An image retrieval system is

The era has witnessed unparalleled growth in the number, availability and importance of images in all directions of life. As the large diversity and huge size of digital image collections have grown exponentially, an efficient image retrieval method is becoming increasingly significant and important too. From large image databases it is difficult to search and retrieve images with traditional text searches because the process of user based annotation has become very tedious and time consuming, as the text often fails to convey the rich structure of images. A prominent content-based retrieval system solves this problem significantly, where the retrieval is based on the automatic matching of the feature of the query image with that of the image database through some image-image similar parameters evaluation. Therefore, images will be indexed based to their own visual content parameters, such as texture, shape and color or any other feature or a combination of set of visual

Existing image search engines, such as Google and Yahoo image search, are based only on relevant text given to images. But the synonymy and polysemy of the relevant texts lead to the serious limitation. To solve these problems, the visual features are extracted and used in Content Based Image Retrieval (CBIR) [Matei Dobrescu et al., 2010]. Average of Red, Green and blue color channels (Average RGB), Global Color Histogram (GCH), Local Color Histogram (LCH) and Color Moment of Hue, Saturation and brightness Value (HSV) are some of the techniques that have been used in CBIR method to extract the visual features for query image. The results of each technique are used for comparison; the significance of this comparison is to find out which one of these techniques gives more accurate and efficient