Medical Image Retrieval Using Content Based Image Retrieval System

DOI : 10.17577/IJERTV3IS10832

Download Full-Text PDF Cite this Publication

Text Only Version

Medical Image Retrieval Using Content Based Image Retrieval System

1Kanupriya, 2Amanpreet Kaur

1 Computer Science and Engineering RIMT-IET

Mandigobindgarh

2 Computer Science and Engineering RIMT-IET

Mandigobindgarh

Abstract. Due to the enormous increase in image database sizes, as well as its vast deployment in various applications, the need for CBIR development arose. The present Content Based Image Retrieval system is to describe the solution to the problem of retrieving the query image from the large image database using fuzzy C-Means clustering method. The CBIR can use the primitive features of an image such as texture, color, orientation and shape. These features are extracted and used as the basis for a similarity check between images stored in the database. In the proposed approach the color and texture features are used and it also uses the fuzzy c-means clustering algorithm. The database consists of various medical images of different size. The Red Green Blue color model is used. The color value of each image in the database and the query image is obtained & then Median Filtering is applied to reduce the noise. The Fuzzy C-Means Clustering can used to obtain the more features of the images and to

improve retrieval efficiency. The similarity between the query image and the images in the database is done using Euclidean Distance approach and the minimum distance image is retrieved from the database. The final result is obtained, that utilizes the features of the images as the basis for comparison and retrieval using Matlab functions.

Keywords: Content-Based Image Retrieval, Texture feature extraction, Fuzzy C-Means Clustering, Median Filtering, Matlab, RGB color space.

  1. Introduction

    Content based image retrieval (CBIR) is the application of Computer vision techniques to the image retrieval problem, that is, the problem of searching for digital images in large databases. The Content Based Image Retrieval tries to solve this problem as it provides the means to index, search and retrieve those images. It is a task of searching images from a database and retrieval of an image, which seems to be similar

    to a given example or query image. Content- based image retrieval uses the visual contents of an image such as color, shape, texture, and spatial layout to represent and index the image. In typical content-based image retrieval systems, the visual contents of the images in the database are extracted and described by multi- dimensional feature vectors. Various feature vectors can be computed by different methods that are available to the users. The CBIR system consists of following components:

    • Query image: This is the image to be

      search in the image database whether the same image is present or not or how many similar kind images exist or not.

    • Image Database: Consists of n number of images depends on the user choice.

    • Feature extraction: It extracts visual information from the image and saves them as features vectors in a features database. The feature extraction finds the image description in the form of feature value (or a set of value called a feature vector) for each pixel. These feature vectors are used to compare the query with the other images and retrieval.

    • Image matching: Here, information about each image is stored its feature vectors for computation process and these feature vectors are matched with the feature vectors of query image which helps in measuring the similarity.

    • Resultant Retrieved images: It searches the previously maintained information to find the matched images from database and the output will be images having same or very closest features as that of the query image.

        1. Existing CBIR Systems

          Some of the existing CBIR systems [3] are:

          • QBIC or Query by Image Content

            It is the first commercial content based retrieval system. This system allows users to graphically pose and refine queries based on multiple visual properties such as color, texture and shape. QBIC supports queries based on colored input images, user- constructed images, sketches, and selected color and texture patterns.

          • VisualSEEK and WebSEEK

            Virage is content based image search engine developed at Virage Inc. It supports texture and spatial location matching as well as color matching.

          • NeTra

            This system uses texture, shape, color, spatial location and texture matching, as well as image segmentation.

          • MARS (Multimedia Analysis and Retrieval System)

            This system makes use of color, spatial layout, texture and shape matching.

          • Viper (Visual Information Processing for Enhanced Retrieval)

            This system retrieves images based on color and texture matching.

          • The img (Anaktisi) is a CBIR system on the web based on various descriptors which includes powerful color and texture features. The img (Anaktisi) provides different ways to search and retrieve them.

        2. Objectives

          There are two major issues concerned while designing a CBIR system:

          • RGB Component of all images are calculated and stored.

          • Every image in the image database is to be represented efficiently by extracting significant features.

          • Resultant image is to be retrieved using similarity measure between query and every image in the image data base.

      statistical measures have been used, such as the gray level energy, discrete entropy, relative entropy, mutual information and information redundancy [1].

      A system based on the fuzzy c-means clustering algorithm, this CBIR system uses color and texture features in color image segmentation. Technique to form compound queries based on the combined features of different images is devised. This technique allows users to have a better control on the search criteria, thus a higher retrieval performance can be achieved [2].

      Query Image Images in Database Feature Extraction (Feature vectors) Image Matching/Similarity Checking (Euclidean Distance) Retrieved image. Surveyed almost 300 key theoretical and empirical contributions in the current decade related to image retrieval and

      QUERY IMAGE

      FEATURE EXTRACTION

      IMAGE IN DATABASE

      IMAGE MATCHING

      automatic image annotation, and in the process

      RETRIEVED IMAGE

      discuss the spawning of related subfields and discuss significant challenges involved in the adaptation of existing image retrieval techniques to build systems that can be useful in the real

      Fig 1. CBIR system and its components.

  2. Related Work

    Fuzzy C means clustering method with thresholding for underwater image segmentation is explained. The paper focuses on comparison of fuzzy c means clustering algorithms with proposed method for underwater images. The paper focuses on comparison of fuzzy c means clustering algorithms with proposed method for underwater images. To evaluate the nonlinear image region segmentation, quantitative

    world [3].

    The image is represented by a Fuzzy Attributed Relational Graph (FARG) that describes each object in the image, its attributes and spatial relation. Here, texture and color attributes are computed in a way that model the Human Vision System (RGB). A new approach for graph matching that resembles the human thinking process is proposed [4].

  3. Proposed System Approach and System Architecture

      1. System Architecture

        INPUT IMAGE (QUERY AND DATABASE)/p>

        COLOR & TEXTURE TRANSFORM AND MEDIAN FILTERING

        FEATURE EXTRACTION (GLCM) FEATURE EXTRACTION (FUZZY C MEANS FOR

        BETTER RESULTS

        IMAGE COMPARISION (DISTANCE COMPUTATION)

        Fig 1.

        RETRIVE AND DISPLAY IMAGE HAVING SAME FEATURES

        Fig. 2: CBIR using proposed approach

      2. Image Feature Extraction

        This process computes the image feature vectors which are then used by Euclidean distance calculation and helps in retrieval process .The RGB color model is used. To extract the color feature, color median filtering is applied recursively to deemphasize noises for preprocessing and color is calculated. The median filter is to run through the signal entry by entry, replacing each entry with median of neighboring entries. After feature extraction, the pixel features are clustered into groups using the fuzzy c-means clustering algorithm if the result is to be improved. The query images are processed and transform to different color space for better performance, then similarity matrix

        can be calculated using fuzzy clustering of images. The distance between the two images is thus found and image having minimum distance that is the images similar to query image is retrieved and displayed. The use of color features has become increasingly important. With the assistance of color features, objects in an image can be distinguished easily.

      3. Proposed Fuzzy C-Means Clustering The Fuzzy C-Means (FCM) is a clustering which allows one piece of data to belong to two or more clusters. This method is frequently used in pattern recognition. The FCM objective function and its generalizations are the most heavily studied fuzzy model in Pattern Recognition. There is an infinite range of possible fuzzy partitions. Therefore, an optimization model or objective function must be devised to search Input images (Query + Database images) Color space transformation & Median filtering Feature Extraction (GLCM) Feature Extraction (Fuzzy C-means clustering for better result) Retrieved and Display images having same features Image comparison (distance computation) for the optimal partition according to the chosen objective function. The way that most researchers have solved the optimization problem has been through an iterative locally optimal technique, called the FCM algorithm. The FCM objective function weighted the distance between a given data point and a given prototype by the corresponding degree of membership between the two. Thus, partitions that minimize this function are those

    that weight small distances by high membership values and large distances by low membership values.

  4. Result of CBIR System

    The CBIR system is working on color transformation, filtering, feature extraction, calculating distance between the feature vectors, similarity matrix .Using the feature extraction described above ,the feature vectors of query and images in the database or file are obtained and compared using the Euclidean distance Metric .The result is retrieved images matched on the basis on the feature vector values .

    Fig. 3: Resultant retrieved images by CBIR system

  5. Conclusion & Future Work

    Content based image retrieval system is using the existing inbuilt function of MATLAB software is easiest way to implement. It is not necessary that image having same color is of same domain, so there is a need of comparing texture and shape also to improve results. As

    image collections grow in size the system may take a lot of time, and eventually reduce the query-retrieval process. To increase the speed and the user's interaction with image retrieval systems, the images to be access from the web/Internet sources and the CBIR system can be implemented over the World Wide Web and applying proposed fuzzy C-means algorithm in a more efficient manner.

  6. References

  1. Content-based image retrieval systems. IEEE Computer, 28(9), 1995.

  2. M. Flickner et al. Query by image and video content: The QBIC system. IEEE Computer, 28(9):2332, 1995.

  3. V. Ogle and M. Stonebraker. Chabot: Retrieval from a relational database of images. IEEE Computer, 28(9):4048, 1995.

  4. Pentland, R. Picard, and S. Sclaroff. Photobook: Content based manipulation of image databases. International Journal of Computer Vision, 18(3):233 254, 1996.

  5. Jiang, J., A. Armstrong and G. C. Feng, Direct content access and extraction from JPEG compressed images, Patten Recognition, pp.1-9, 2001.

  6. Lay, J. A. and L. Guang, Image retrieval based on energy histograms of the low frequency DCT coefficients, Proc. of

    the IEEE International Conference on Acoustics, Speech, and Signal Processing, vol.6, 3009-3012, 1999.

  7. Sim, D. G., H. K. Kim and R. H. Park, Fast texture description and retrieval of DCT-based compressed images, Electronic Letters, vol.37, no.1, pp.18- 19, 2001.

  8. Huang, Y. L. and R. F. Chang, Texture features for DCT-coded image retrieval and classification, Proc. of the IEEE International Conference on Acoustics, Speech, and Signal Processing, vol.6, pp.3013-3016, 1999.

  9. M. Flickner et al. Query by image and video content: The QBIC system. IEEE Computer, 28(9):23-32, September 1999.

  10. Huang, Y. L. and R. F. Chang, Texture features for DCT-coded image retrieval and classification, Proc. of the IEEE International Conference on Acoustics, Speech, and Signal Processing, vol.6, pp.3013-3016, 1999.

  11. M. Flickner et al. Query by image and video content: The QBIC system. IEEE Computer, 28(9):23-32, September 1995.

  12. Wynne Hsu, T. S. Chua, and H.

    K. Pung. An integrated color-spatial approach to content-based image

    retrieval. In ACM Multimedia Conference, pages 305-313, 1995.

  13. Virginia Ogle and Michael Stonebraker. Chabot: Retrieval from a relational database of images.IEEE Computer, 28(9):40-48, September 1995.

  14. P. Berman and L. G. Shapiro. Efficient content based retrieval: Experimental results. In IEEE Workshop on Content Based Access of Image and Video Libraries, 1999.

  15. C.Carson, M. Thomas, S.Belongie, J.Hellerstein, and J. Malik. Blobworld: A system for region based image indexing and retrieval. In Third International Conference on Visual Information Systems. SpringerVerlag, June 1999.

  16. J. S. De Bonet and P. Viola. Structure driven image database retrieval. Advances in Neural Information Processing, 10,1997.

  17. M. Flickner, H. Sawhney, W. Niblack, J. Ashley, Q. Huang, B. Dom,

    M. Gorkani, J. Hafner, D. Lee, D. Petkovic, D. Steele, and P. Yanker. Query by image and video content

    :TheQBIC system. IEEE Computer,28(9):2332, september1995.

  18. J. Zhang and T. Tan, Brief

    review of invariant texture analysis methods, Pattern Recognit 35 (2002),

    735747.

  19. G.M. Haley and B.S. Manjunath, Rotation invariant texture classification using a complete space frequency model, IEEE Trans Image Process 8 (1999), 255269.

  20. K. Jafari-Khouzani and H. Soltanianzadeh, Radon transform orientation estimation for rotation invariant texture analysis, IEEE Trans Pattern Anal Machine Intell 27 (2005), 10041008.

  21. Haralick RM. Statistical and Structural Approaches to Texture[J].Proceeding of the IEEE,1979,67(5):786-804.

  22. S. Arivazhagan, and L. Ganeshan, Texture classification using wavelet transform Pattern Recognition Letters, Vol.24, pp. 1513-1521, 2003.

  23. R. M. Haralick, statistical and structural approaches to texture, Proceedings of IEEE 67:786-804, 1979.

  24. Ritendra Datta, Dhiraj Joshi, Jia Li and James Wang, Image Retrieval: Ideas, Influences, and Trends of the New Age, Proceedings of the 7th ACM SIGMM international workshop on Multimedia information retrieval,

November 10-11, 2005, Hilton, Singapore.

Leave a Reply