CBIR Using Features Extracted from
Halftoning-Based Block Truncation Coding
with PRESS Algorithm
Malarvizhi.A1, Nagajothi.S2
Asst. Professor, Dept. of ECE, Vinayaka Missions Kirupananda Variyar Engineering College, Salem, India1
Asst. Professor, Dept. of ECE, ULTRA College of Engineering and Technology for Women, Madurai,India 2
ABSTRACT: CBIR technique can be establish in a number of different domains such as Data Mining, Education, Medical Imaging, Crime Prevention, Weather forecasting, Remote Sensing and Management of Earth Resources. In this paper, shape features are extracted from the database images and the same are polar raster scanned into specified intervals in both radius and angle, using the proposed Polar Raster Edge Sampling Signature (PRESS) algorithm. Counts of edge points lying in these cases are stored in the feature library. When a query image passed on to the system, the features are extracted in the similar fashion. Subsequently, similarity measure is performed between the query image features and the data-base image features based on Euclidian Distance similarity measure and the database images that are relevant to the given query im collage are retrieved. PRESS algorithm has been successfully implemented and tested in a CBIR System developed by us. This technique pre-serves rotation and scale invariance. It is evaluated by querying different images. The retrieval efficiency is also estimated by finding precision-recall values for the retrieval results.
KEYWORDS: image retrieval; polar raster edge sampling signature; image segmentation; feature extraction; K-means clustering; Canny algorithm
I. INTRODUCTION
From large scale image databases according to users' interests, has been an active and fast advancing research area since the 1990s. in the past ten years, remarkable progress has been made in both theoretical research and system development. However, there remain many challenging research problems that continue to attract researchers from multiple disciplines.Information retrieval is the process of converting a request for information into a meaningful set of references. Early work on image retrieval can be traced back to the late 1970s. In 1979, a conference on Database Techniques for Pictorial Applications was held in Florence. Since then, the application potential of image database management techniques has attracted the attention of researchers. Early techniques were not generally based on visual features and also on the textual annotation of images.In other words, images were first annotated with text and then searched using a text-based approach from traditional database management systems.
The difficulties faced by text-based retrieval became more and more severe. This need formed the driving force behind the emergence of content-based image retrieval techniques.Since 1997, the number of research publications on the techniques of visual information extraction, organization, indexing, user query and interaction, and database management has increased enormously. Similarly, a large number of academic and commercial retrieval systems have been developed by universities, government organizations, companies, and hospitals.
II. FUNDAMENTAL ASPECTSOF CBIR AND RELATED WORKS
Previous CBIR systems can be classified into two categories according to the type of queries: text query or pictorial query. In text query based systems, images are characterized by text information such as keywords and captions. Text features are powerful as a query, if appropriate text descriptions are given for images in an image database. However, giving appropriate descriptions must be done manually in general and it is time consuming. There are many ways one can pose a visual query. A good query method will be natural to the user as wellas capturing enough information from the user to extract meaningful results.
In pictorial query based systems, an example of the desired image is used as a query. To retrieve similar images with the example, image features such as colors and textures, most of which can be extracted automatically, are used.The typical CBIR system performs two major tasks. The first one is feature extraction (FE), where a set of features, called image signature or feature vector, is generated to accurately represent the content of each image in the database. A feature vector is much smaller in size than the original image, typically of the order of hundreds of elements (rather than millions). The second task is similarity measurement (SM), where a distance between the query image and each image in the database using their signatures is computed so that the top“closest” images can be retrieved.Popular knowledge claims that an image is worth 1000 words. Unfortunately, these 1000 words may differ from one individual to another depending on their perspective and/or knowledge of the image context. Thus, even if a 1000 word image description were available, it is not certain that the image could be retrieved by a user with a different description. The image retrieval is an interesting and fastest developing methodology in all fields. It is an effective and well organized approach for retrieving the image. Image retrieval techniques are splitted into two categories text and content-based categories. The text-based algorithm comprises some special words like keywords. Keywords and annotations should be dispenses to each image, when the images are stored in a database. The annotation operation is time consuming and tedious. In addition, it is subjective. Furthermore, the annotations are sometimes incomplete and it is possible that some image features may not be mentioned in annotations [1]. In a CBIR system, images are automatically indexed by their visual contents through extracted low-level features, such as shape, texture, color, size and so on [1]. However, extracting all visual features of an image is a difficult task and there is a problem namely semantic gap in the semantic gap, presenting high-level visual concepts using low-level visual concept is very hard. In order to alleviate these limitations, some researchers use both techniques together using different features.
III.EXISTING WORK
The CCF and BPF of an image are simply derived from the two ODBTC quantizers and bitmap, respectively, by involving the visual codebook. Experimental results show that the proposed method is superior to the block truncation coding image retrieval systems and the other earlier methods, and thus prove that the ODBTC scheme is not only suited for image compression, because of its simplicity, but also offers a simple and effective descriptor to index images in CBIR system.
In this paper, the ODBTC algorithm is generalized for color images in coping with the CBIR application. The main advantage of the ODBTC image compression is on its low complexity in generating bitmap image by incorporating the Look-Up Table (LUT), and free of mathematical multiplication and division operations on the determination of the two extreme quantizers. The traditional BTC derives the low and high mean values by preserving the first-order moment and second-order moment over each image block, which requires additional computational time. Conversely, ODBTC identifies the minimum and maximum values each image block as opposed to the former low and high mean values calculation, which can further reduce the processing time in the encoding stage. In addition, the ODBTC yields better reconstructed image quality by enjoying the extreme-value dithering effect.
The color distribution of the pixels in an image contains huge amount of information about the image contents. The attribute of an image can be acquired from the image color distribution by means of color co-occurrence matrix. This matrix calculates the occurrence probability of a pixel along with its adjacent neighbors to construct the specific color information. This matrix also represents the spatial information of an image.
Color Co-occurrence Feature (CCF) can be derived from the color co-occurrence matrix. In the proposed scheme, CCF is computed from the two ODBTC color quantizers. The minimum and maximum color quantizers are firstly indexed using a specific color codebook. The color co-occurrence matrix is subsequently constructed from these indexed values. Subsequently, the CCF is derived from the color co-occurrence matrix at the end of computation. In general, the color indexing process on RGB space can be defined as mapping a RGB pixel of three tuples into a finite subset (single tuple) of codebook index (the most representative codeword). LBG Vector Quantization (LBGVQ) generates a representative codebook from a number of training vectors. an image retrieval system is presented by exploiting the ODBTC encoded data stream to construct the image features, namely Color Co-occurrence and Bit Pattern features. As documented in the experimental results, the proposed scheme can provide the best average precision rate compared to various former schemes in the literature. As a result, the proposed scheme can be considered as a very competitive candidate in color image retrieval application.
IV.PROPOSED WORK
Figure1: PRESS Algorithm
The proposed effective Content Based Image Retrieval system comprises of the processes such as Image Segmentation, Feature Extraction and Image retrieval based on the query image. Here the CBIR is based on the extracted shape signature. The shape signature extraction is as follows. Initially, the image is segmented based on color, using the K – Means clustering algorithm. Canny algorithm is employed to detect the edges. The strong edges and the connected edges are identified using various techniques like double thresholding and edge tracking. The edge data in the boundary and the region are polar raster scanned in both radius and angle. Numbers of edge points identified are stored in the feature library for all the database images. When an image is queried, the system extracts shape feature for the image in the same way and then computes the similarity measure between the features of the query image and the feature existing in the feature database based on the Euclidean Distance method. Minimum distance indicates the closest match and specified number of best matched images are extracted. The proposed method is detailed in the following sections.
In this pseudo code the proposed PRESS algorithm is used for extracting the shape features of the image. The approach is that the final edge components extracted are polar raster scanned or binned into ten intervals in the Radius and the Angle. Count of edge points lying in these bins are found and stored as two vectors r and t. It is further normalized for sum of counts. The same process is repeated for all the images in the database and the r and t values are saved in the feature database.
V.EXPERIMENTAL RESULTS
Figure2: Query Image and Retrieved Images Precision refers to the percentage of retrieved pictures that are relevant to the query. Recall pertains to the percentage of all the relevant pictures in the search database which are retrieved. Precision and Recall follow an inverse relationship. Precision falls while recall increases as the number of retrieved pictures, often termed as scope, increases. Hence, it is typical to have a high numeric value for both precision and recall. Traditionally, results are summarized as precision-recall curves or precision-scope curves.
Figure 3:In precision and recall, crossover is the point on the graph where both the precision and recall curves meet. The crossover point can also be used to measure the accuracy. Higher the number of crossover points better will be the performance of the system.
VI.CONCLUSION
Euclidean Distance method. Image segmentation is done using K-means clustering algorithm, which groups the image pixels under color. Canny algorithm is then used to extract edges. Proposed PRESS algorithm has been successfully applied on the edge components to extract shape features. The same is tested in a CBIR System developed by us. This technique preserves rota-tion and scale invariance. It is evaluated by querying different images. Precision-Recall values are used for the comparison of retrieval results. The implementation results illustrates that this novel image retrieval process effectively retrieves the images that are very close to the query image from the database. Precision - Recall table and Precision comparison plot with existing CBIR techniques prove the effectiveness of the system.
REFERENCES
[1] Li, X., Shou, L., Chen, G., Hu, T. and Dong, J. (2008). Modelling Image Data for Effective Indexing and Retrieval In Large General Image Database. IEEE Transaction on Knowledge and Data Engineering. 20(11), 1566-1580.
[2] S.-F. Chang, A. Eleftheriadis, and R. McClintock, Next-generation content representation, creation and searching for new media applications in education, IEEE Proceedings, 1998, to appear.
[3] G. Qiu, “Color image indexing using BTC,” IEEETrans. Image Process., vol. 12, no. 1, pp. 93–101, Jan. 2003.
[4] M. R.Gahroudi and M.R. Sarshar, “Image retrieval based on texture and color method in BTC-VQ compressed domain,” in Proc. 9th Int. Symp. Signal Process. Appl., Feb. 2007, pp. 1–4.
[5] F.-X. Yu, H. Luo, and Z.-M. Lu, “Colour image retrieval using pattern co-occurrence matrices based on BTC and VQ,” Electron. Lett., vol. 47, no. 2, pp. 100–101, Jan. 2011.
[6] V. Udpikar and J. Raina, “BTC image coding using vector quantization,” IEEE Trans. Commun., vol. 35, no. 3, pp. 352– 356, Mar. 1987. [7] Y. Wu and D. C. Coll, “BTC-VQ-DCT hybrid coding of digital images,” IEEE Trans. Commun., vol. 39, no. 9, pp. 1283– 1287, Sep. 1991. [8] C. S. Huang and Y. Lin, “Hybrid block truncation coding,” IEEE Signal Process. Lett., vol. 4, no. 12, pp. 328–330, Dec. 1997.
[9] Y.-G. Wu and S.-C. Tai, “An efficient BTC image compression tech- nique,” IEEE Trans. Consum. Electron., vol. 44, no. 2, pp. 317–325, May 1998.
[10] J.-M. Guo and M.-F. Wu, “Improved block truncation coding based on the void-and-cluster dithering approach,” IEEE Trans. Image Process., vol. 18, no. 1, pp. 211–213, Jan. 2009.
[11] J.-M. Guo, “High efficiency ordered dither lock truncation coding with dither array LUT and its scalable coding application,” Digit. Signal Process., vol. 20, no. 1, pp. 97–110, Jan. 2010.
[12] J.-M. Guo, M.-F. Wu, and Y.-C. Kang, “Watermarking in conjugate ordered dither block truncation coding images,” Signal Process., vol. 89, no. 10, pp. 1864–1882, Oct. 2009.
