Technological advances in digital imaging, broadband networking, and data storage are motivating millions of people to communicate with one another and express themselves by sharing images, video, and other forms of media online. Acquiring, storing, and transmitting photos is now trivial, but to manipulate, index, sort, filter, summarize, or search through them is significantly harder. The Statistical Visual Computing Laboratory at the University of California, San Diego, has been considering the problem of content-based image retrieval for years. This effort explores many issues in image representation and intelligent system design, including the evaluation of image similarity, automatic annotation of images with descriptive captions, and the ability to understand user feedback during image search.

/pdf/from-pixels-to-semantic-spaces-advances-in-content-based-fl15ylzxov.pdf

From pixels to semantic spaces: Advances in content-based image retrieval - eScholarship

ICTEDCT-CBIR: Integrating curvelet transform with enhanced dominant colors extraction and texture analysis for efficient content-based image retrieval

The paper discusses novel image retrieval methods based on shape features extracted using gradient operators and slope magnitude technique with Block Truncation Coding (BTC). Four variations of proposed „Mask-Shape-BTC‟ image retrieval techniques are proposed using gradient masks like Robert, Sobel, Prewitt and Canny. The proposed image retrieval techniques are tested on generic image database with 1000 images spread across 11 categories. In all 55 queries (5 from each category) are fired on the image database. The average precision and recall of all queries are computed and considered for performance analysis. In all the considered gradient operators for shape extraction, „Mask-ShapeBTC‟ CBIR techniques outperform the „Mask-Shape‟ CBIR techniques. The performance ranking of the masks for proposed image retrieval methods can be listed as Robert (best performance), Prewitt, Sobel and lastly the Canny.

/pdf/image-retrieval-with-shape-features-extracted-using-gradient-48u2809zhv.pdf

Image Retrieval with Shape Features Extracted using Gradient Operators and Slope Magnitude Technique with BTC

Always the thrust for better and faster image retrieval techniques has nourished the research in content based image retrieval. The paper presents 32 novel image retrieval techniques using the feature vectors obtained by applying Walsh transform on row mean and column mean of full image, four fragments, sixteen fragments and 64 fragments of image. All the proposed CBIR techniques are tested on generic image database of size 1000 with 11 image classes. From the average precision and average recall values obtained by firing 55 queries on the image database it is found that use of row mean and column mean with image fragmentation improves the performance resulting in better image retrieval. In all these techniques to speed up the image retrieval process notion of energy compaction is introduced and tested for 100%, 95%, 90% and 85% of energy of feature vectors using Walsh transform.

Walsh Transform over Row Mean and Column Mean using Image Fragmentation and Energy Compaction for Image Retrieval

The paper presents improved content based image retrieval (CBIR) techniques based on multilevel Block truncation coding using multiple threshold values. Block truncation Coding based features is one of the CBIR methods proposed using color features of image. The approach basically considers red, green and blue planes of image together to compute feature vector. The color averaging methods used here are BTC Level-1, BTC Level-2, BTC Level-3.Here the feature vector size per image is greatly reduced by using mean of each plane and find out the threshold value then divide each plane using threshold value, color averaging is applied to calculate precision and recall to calculate the performance of the algorithm. Instead of using all pixel data of image as feature vector for image retrieval, these six feature vectors can be used, resulting into better performance and if increased the no of feature vector get better performance .The proposed CBIR techniques are tested on generic image database having 1000 images spread across 11 categories. For each proposed CBIR technique 55 queries (5 per category) are fired on the generic image database To compare the performance of image retrieval techniques average precision and recall are computed of all queries. The results have shown the performance improvement (higher precision and recall values) with proposed methods compared to BTC Level-1.

Improved CBIR using Multileveled Block Truncation Coding

Content-based image retrieval (CBIR) is a technique to retrieve images on the basis of image specific features such as colour [5, 17, 19], texture [11, 14, 16, 23] and shape [2, 4]. CBIR operates on a totally different principle from keyword indexing. Initially, features are computed for both stored and query images, and used to identify images most closely matching the query. The need to find a desired image from a collection is shared by many professional groups, including journalists, design engineers [24] and art historians. While the requirements of image users can vary considerably, it can be useful to characterize image queries into three levels of abstraction: primitive features such as colour or shape, logical features such as the identity of objects shown and abstract attributes such as the texture of image depicted [12, 15, 18]. Users needing to retrieve images from a collection come from a variety of domains, including crime prevention [13, 20], medicine, architecture [6, 9, 10, 21, 22], fashion and publishing. Many techniques have been proposed for content based image retrieval, but still there is thirst for better performance [1, 3] and faster retrieval [1, 7, 8].

Performance evaluation of image retrieval using energy compaction and imagetiling over DCT row mean and DCT column mean

Kekre Transform over Row Mean, Column Mean and Both Using Image Tiling for Image Retrieval

How to search appropriate data from huge image pool has become vital issue. Because of easy availability of imaging devices, millions of images are being added to image pool every day. Image retrieval deals with searching relevant images from large image database. The paper presents novel image retrieval techniques based on discrete cosine transform applied on row mean, column mean and combination for feature extraction. Further the concept of image fragmentation is added to these to get total of 26 novel CBIR techniques. The proposed image retrieval techniques are applied to image database of 1000 images spread across 11 categories. Experimentation shows that taking row mean, column mean and combination improves the performance of image retrieval as compared to taking DCT of full image. Also fragmentation slightly helps in improving the image retrieval techniques.

Image retrieval using DCT on row mean, column mean and both with image fragmentation

In the world of computer networks, viruses pose a major threat not only to the data, systems, and networks but also to the economy. The main factors that make viruses so dangerous is their rate of propagation and their replicating speed. These features of the viruses take them to a high priority level as far as network security is concerned. Many antivirus software patches are incapable in securing the network from a completely new virus. Most prevention techniques used today are implemented only after the virus has attacked the system and spread out in the network. This makes the security measures very static. In today's world it is very necessary to block an incoming new virus at the entry level itself and prevent its passage in the network in order to curb its harmful action. The security measures implemented must be memory, cost and time efficient. We present the architecture to vaccinate a computer and computer networks not only from existing viruses but also from a newly developed/encountered virus, making the computer and computer networks more secure.

Suraj Shirke

Papers

Walsh Transform over Row Mean and Column Mean using Image Fragmentation and Energy Compaction for Image Retrieval

Performance evaluation of image retrieval using energy compaction and imagetiling over DCT row mean and DCT column mean

Kekre Transform over Row Mean, Column Mean and Both Using Image Tiling for Image Retrieval

Image retrieval using DCT on row mean, column mean and both with image fragmentation

Network vaccination architecture