Content Based Image Retrieval System: Review

Transcription

1 Content Based Image Retrieval System: Review Garvita #1, Priyanka Kamboj #2 #1 M.tech, Computer Science Department, Kurukshetra University #2 Assistant Professor in Computer Science Department, Kurukshetra University JMIT Radaur, Haryana, India Abstract: Content Based Image Retrieval (CBIR), also called as Query By Image Content (QBIC). It has been an active research field since last decades. In contrast to traditional systems, where images are retrieved on the basis of keywords but in the CBIR system images are retrieved on the basis of visual content. Content Based Image Retrieval is an approach that enables a user to extract an image based on a query, from the database containing huge amount of images. Here, the term "content" refer to color, shape, texture, or any other information that can be derived from image itself. CBIRS is becoming a necessity for various applications such as medical imaging, Geographic Information Systems (GIS), space search and many others. In this paper, we have discussed various techniques that enable users to extract images from huge database, thus improving accuracy, efficiency and precision. The main issue in designing a Content Based Image Retrieval System is to select the image features that best represent the image contents in a database. whole visual content of images in words and TBIR may end up in producing irrelevant results. In addition annotation of images is not always correct and takes a lot of time. For finding alternative way of looking and overcoming the limitations forced by TBIR systems more natural and easy to understand content based image retrieval systems (CBIR) were developed. A CBIR system uses visual contents of images described in the form of low level features like color, texture, shape and spatial locations to represent images in databases. The system retrieves comparable images when an query image or sketch is presented as input to the system. Querying in this way removes need for describing the visual content of images in words and is near to human perception of visual data. A portion of representative CBIR systems is Query by Image Content (QBIC). Keywords: CBIRS, Image databases, Color string comparison, Feature extraction, Query image, Target image. I. INTRODUCTION With the headway in internet and multimedia technologies, a immense amount of multimedia data in the form of audio, video and images has been utilized as a part of numerous fields like medical treatment, satellite data, video and still images repositories and surveillance system. This has made a progressing interest of frameworks that can store and retrieve mixed media information in a powerful way. Numerous multimedia information storage and retrieval systems have been developed till now to cater these requests. The most common retrieval systems are Text Based Image Retrieval (TBIR) systems, where search is based on automatic or manual explanation of images. A conventional TBIR searches database for the similar text surrounding the image as given in the query string. The TBIR systems are fast as the string matching is computationally less time consuming process. However, it is sometimes hard to express the Figure: Architecture of a typical CBIR system In a typical CBIR system (above Figure), image low level features like color, texture, shape and spatial locations are represented in the form of a multidimensional feature vector. The feature vectors of images in database form a feature database. The retrieval process is initiated when a user query the system using an image or sketch of the object. The query image is converted into the internal representation of feature vector using same feature extraction routine that was utilized for building the feature database. The similarity measure is utilized to ISSN: Page 129

2 calculate the distance between the feature vectors of query image and those of the objective images in the feature database. At last, the retrieval is performed using an indexing scheme which facilitates an efficient searching of the image database. Recently, user s relevance feedback is also incorporated to further improve the retrieval process in order to produce perceptually and semantically more meaningful retrieval results. II. RELATED WORK There are various methods have been proposed to extract the images from very large database. Here, we have some of the papers that use different techniques to retrieve the images: Roshi Choudhary et al. [7] proposed an approach to perform content based image retrieval. It is an integrated approach used to extract color and texture feature from images. By using single feature, correct results can never produced. So multi feature extraction is more beneficial to perform image retrieval. To extract the color feature, higher order of color moment is used which is the descriptor of color. To extract texture, LBP is used which is the descriptor of texture. Local binary pattern is mainly used to face recognition Vinee. V. Kawade et al. [10] announced a user based system for CBIR in which genetic algorithm is applied. The different features of color image like mean, standard deviation and the image bitmap are used for retrieval. The texture features like edge histogram of an image and the entropy of gray level co-occurrence matrix are used. Moreover, the genetic algorithm is applied to help user in identifying the images which satisfy his needs for reducing gap between the users expectation and the retrieval results. Experimental results show remarkable improvement in the performance after applying IGA. Kommineni Jenni et al. [8] presented a Content Based Image Retrieval approach based on the database classification using Support Vector Machine (SVM) and color string coding feature selection. In SVM method, the feature extraction was done based on the basis of color string coding and string comparison. Here, they succeed in transferring the images retrieval problem to strings comparison. Thus the computational complexity is decreases obviously and increased the accuracy in obtaining results for image retrieval. Using database classification we can improve the performance of the content based image retrieval than compared with normal CBIR that is without database classification. Anuja khodaskar et al. [1] proposed an advanced content based image retrieval system using topical rule based classification strategy which improve retrieval performance significantly. The proposed classification strategy used three training rules, low level, high level and expert rules which improve classification accuracy and effectiveness, ultimately encroachment in quality of classification. Experimental result shows performance evolution in precision, accuracy and retrieval time of image retrieval. Siddarth Ladhake et al. [6] provides a system for the large scale database is designed and implemented. Million images on internet is a big challenge for accurate and efficient image retrieval. Here, the proposed system exploits semantic binary code generation techniques with semantic hashing function, fine and coarse similarity measure technique, automatic and manual relevance feedback technique which improve accuracy, speed of image retrieval. In this experimental result clearly shows that performance of image retrieval is improved in terms of accuracy, efficiency and retrieval time. Y H Sharath Kumara et al. [5] presents a model for representation and indexing of flower images for purpose of retrieving flowers of interest based a query sketch. Here, they swot the correctness of Kdtree indexing scheme for flower retrieval system based on shape descriptors viz., Scale Invariant Feature Transform (SIFT), Histogram of Gradients (HOG) and Edge Orientation Histograms (EOH). To uphold the efficiency of proposed method, an experiment has been conducted on their own flower data set and from this it achieves a good accuracy with indexing approach. R. Malini et al. [4] was desired to improve the effectiveness of retrieving images on the basis of color content by Color Averaging technique. Firstly, an average mean based technique with reduced feature size is proposed. Secondly, a feature extraction technique based on central tendency is proposed. The proposed CBIR techniques are tested on Wang image database and indexed image database. In the last, results that are obtained compared with the existing technique based on memory utilization and query execution time. The experimental results show that proposed technique gives the better performance in terms of higher precision and recall values with less computational complexity than the conventional techniques. ISSN: Page 130

3 Manoharan Subramanian et al. [3] proposed a new mechanism for CBIR systems which is based on two works. The first work is based on filtering technology which includes anisotropic morphological filters, hierarchical Kaman filters and particle filters. The second work is based on the feature extraction which includes color and gray level features and after this the results were normalized. Finally, the experimental results shows that this proposed technique of CBIR using advanced filter approaches is much better than the existing system GLCM and color feature extraction for CBIR process. Priyadarshini Patil et al. [2] proposed and implemented an efficient image retrieval technique using both color and texture features of an image. In real time applications, we need for developing efficient techniques to find images form huge databases. To find an image from database, every image is represented with certain features. Color and texture are the two important visual features of an image. Here we compare and analyze performance of an image retrieval using both these features. And we see that CBIR using color features gives high precision whereas CBIR using texture feature features give high recall. On average, CBIR using texture-color features increases precision by 37% and recall by 22%. Hence, CBIR using texture-color features is efficient and boosts performance. Manish K. Shriwas et al. [11] proposed a new technique for CBIR called as Local tetra patterns (LTrPs). In this technique images are retrieving from database based on the direction of pixel that is calculated by vertical and horizontal derivatives. By combining LTrPs with Gabor transform (GT) the effectiveness of proposed method will also be analyzed. By considering the diagonal pixel for calculation the derivation of vertical and horizontal direction the result of proposed method will be improved. Accuracy and efficiency of proposed approach will reaches a level of 93% and it can also be suitable for various applications such as biomedical research, educational, crime prevention, web images etc. Sushant Shrikant Hiwale et al. [12] proposed a CBrn system which extracts the features of digital image to retrieve similar images from huge databases. We have used Color Histogram, Color Auto-Correlogram, Color Moment, Gabor Wavelet and Discrete Wavelet transform to extract image features. The images are classified using Support Vector Machine (SVM) classifier which effectively distinguishes between relevant and irrelevant images. The results depict that proposed method has better precision and recall rate compared to other methods. A. A. Khodaskar et al. [13] proposed innovative framework for effective, intelligent and efficient content based image retrieval is based on three soft computing techniques such as Artificial Neural Network, Fuzzy Logic and support vector machine. In this framework, the SVM based relevance feedback is introduced, that will provide feedback to the system that whether the retrieved image is relevant or not. The main purpose is to avoid nonrelevant images and to reduce the un-necessary overhead of the system. Content based image retrieval based on soft computing techniques like ANN, fuzzy logic and support vector machine improved retrieval performance in term of accuracy, precision and efficiency by using immense power complementary nature of soft computing techniques. Devyani Soni et al. [14] proposes an efficient color space Based Approach for Image Retrieval Using fusion of Color Histogram and color correlogram. During experimentation, both HSV color model as well as RGB color model is used for the same process of retrieval. She developed this mechanism for image retrieval based on color features of image with the help of MATLAB tool. The retrieval accuracy is close to 100% in the proposed algorithm and it has also been observed that HSV color space gives more accurate result as compared to RGB color space. This work can be enhanced by taking other features such as texture and shape of image into consideration. Ammar Huneiti et al. [9] proposed a CBIR method by extracting both color and texture feature vectors using the Discrete Wavelet Transform (DWT) and the Self Organizing Map (SOM) artificial neural networks. At query time texture vectors are compared using a similarity measure which is the Euclidean distance and the most similar image is retrieved. In addition, other relevant images are also retrieved using the neighborhood of the most similar image from the clustered data set via SOM. Results showed that the proposed method is able to retrieve images with higher average precision values than other methods proposed in literature by just comparing the texture similarity and without any need to compare color similarities. ISSN: Page 131

4 Table No.1 Comparative study of recent Image Retrieval Techniques. S.No. Author s Name Technique Used Results 1. Roshi Choudhary Color Moment (CM) and Local Binary Pattern (LBP) Extract color and texture feature, Improves performance 2. Vinee. V. Kawade Genetic Algorithm (GA) Retrieves images efficiently and precisely 3. Kommineni Jenni Support Vector Machine (SVM) and color string coding feature selection Increases the performance and provide better results 4. Anuja khodaskar Topical rule based classification Improves image retrieval accuracy and Retrieval time 5. Siddarth Ladhake Semantic Binary Code Generation Enhanced accuracy, efficiency and retrieval time 6. Y H Sharath Kumara Kd-tree based indexing approach Achieves good accuracy 7. R. Malini Feature Vector Extraction Provides better retrieval results and performance 8. Manoharan Subramanian Anisotropic morphological filters, hierarchical Kaman filters and particle filters, also Feature extraction which includes color and gray level features High level filters with feature extraction gives normalized results 9. Priyadarshini Patil Texture-color feature based CBIR using trace transform and Color Histogram High precision, High recall, more efficient and boosts performance 10. Manish K. Shriwas Local tetra patterns (LTrPs) Accuracy and efficiency is close to 93% 11. Sushant Shrikant Hiwale Color Histogram, Color Auto- Correlogram, Color Moment, Gabor Wavelet and Discrete Wavelet transform and Support Vector Machine (SVM) classifier 12. A. A. Khodaskar Artificial Neural Network, Fuzzy Logic and support vector machine 13. Devyani Soni Color space Based Approach using fusion Color Histogram and Color Correlogram 14. Ammar Huneiti Discrete Wavelet Transform (DWT) and the Self Organizing Map (SOM) Artificial Neural Networks Provide better precision and recall rate than other methods Avoids non-relevant images and Improves accuracy, precision and efficiency by using these soft computing techniques Retrieval accuracy is close to 100% Retrieve images with higher average precision values than other methods ISSN: Page 132

5 III. CONCLUSION AND FUTURE SCOPE Nowadays, CBIR has been an active research field and is used in various fields like Medical, Geographic Information System (GIS), and Space Research etc. In this paper, we have discussed various techniques that enable users to extract the relevant image from huge database improving the accuracy, precision and reducing retrieval time. In future work, we will work on feature extraction and classification to get the better results for image retrieval. [13] A. A. Khodaskar and S. A. Ladhake, A Novel Approach for Content Based Image Retrieval in Context of Combination S C Techniques, International Conference on Computer Communication and Informatics (ICCCI -2015), Jan , 2015, Coimbatore, INDIA [14] Devyani Soni, K. J. Mathai, An Efficient Content Based Image Retrieval System based on Color Space Approach Using Color Histogram and Color Correlogram, Fifth International Conference on Communication Systems and Network Technologies 2015 REFERENCES [1] Anuja khodaskar and Siddarth Ladhake, Advanced Image Retrieval with Topical Classification Strategy, International Conference on Intelligent Computing, Communication & Convergence (ICCC-2014) Conference Organized by Interscience Institute of Management and Technology,Bhubaneswar, Odisha, India [2] Priyadarshini Patil and Bhagya Sunag, Analysis of Image Retrieval Techniques Based On Content, IEEE International Advance Computing Conference (IACC) [3] Manoharan Subramanian and Sathappan Sathappan, An Efficient Content Based Image Retrieval using Advanced Filter Approaches, The International Arab Journal of Information Technology, Vol. 12, No. 3, May 2015 [4] R.Malini and C.Vasanthanayaki, An Enhanced Content Based Image Retrieval System using Color Features, International Journal Of Engineering And Computer Science ISSN: Volume 2 Issue 12 Dec, 2013 Page No [5] Y H Sharath Kumara, D S Gurub, Retrieval of Flower Based on Sketches, International Conference on Information and Communication Technologies (ICICT 2014) [6] Anuja khodaskar, Siddarth Ladhake, Promising Large Scale Image Retrieval by using Intelligent Semantic Binary Code Generation Technique, International Conference on Intelligent Computing Communication & Convergence (ICCC-2014) Conference Organized by Interscience Institute of Management and Technology, Bhubaneswar, Odisha, India [7] Roshi Choudhary, Nikita Raina, Neeshu Chaudhary, Rashmi Chauhan, Dr. R H Goudar, An Integrated Approach to Content Based Image Retrieval, International Conference on Advances in Computing, Communications and informatics (lcacc1) 2014 [8] Kommineni Jenni, Satria Mandala, Mohd Shahrizal Sunar, Content Based Image Retrieval Using Colour Strings Comparison, 2nd International Symposium on Big Data and Cloud Computing (ISBCC 15) [9] Ammar Huneiti, Maisa Daoud, Content-Based Image Retrieval Using SOM and DWT, Journal of Software Engineering and Applications, 2015, 8, Published Online February 2015 [10] Vinee. V. Kawade and Arti. V. Bang, Content Based Image Retrieval Using Interactive Genetic Algorithm, Annual IEEE India Conference (INDICON) 2014 [11] Manish K. Shriwas, Prof. V. R. Raut, Content Based Image Retrieval: A Past, Present and New Feature Descriptor, IEEE International Conference on Circuit, Power and Computing Technologies [ICCPCT] 2015 [12] Sushant Shrikant Hiwale, Dhanraj Dhotre, Dr. G.R.Bamnote, Quick Interactive Image Search in Huge Databases Using Content-Based Image Retrieval, IEEE Sponsored 2nd International Coriference on Innovations in Iriformation Embedded and Communication Systems ICIIECS'15 ISSN: Page 133