A Survey: Content Based Image Retrieval

Abstract

The field of image processing is addressed significantly by the role of CBIR. Peculiar query is the main feature on which the image retrieval of content based problems is dependent. Relevant information is required for the submission of sketches or drawing and similar type of features. Many algorithms are used for the extraction of features, which are related to similar nature. The process can be optimized by the use of feedback from the retrieval step. Analysis of colour and shape can be done by the visual contents of image. Here, neural network, Relevance feedback techniques based on image retrieval are discussed.

Headnote

ABSTRACT

The field of image processing is addressed significantly by the role of CBIR. Peculiar query is the main feature on which the image retrieval of content based problems is dependent. Relevant information is required for the submission of sketches or drawing and similar type of features. Many algorithms are used for the extraction of features which are related to similar nature. The process can be optimized by the use of feedback from the retrieval step. Analysis of colour and shape can be done by the visual contents of image. Here neural network, Relevance feedback techniques based on image retrieval are discussed.

Keywords- Colour, Database, Image, Query, Shape.

I. INTRODUCTION

CBIR is a term known as the process of Retrieving image from features such as shape information, texture or colour.[1]The researches carried on database and processing of images involves CBIR system and these vary from issues of storage that are user friendly[2] PNPA operation is applied successfully which eliminates the discontinuity of edges that formed by the image capturing from a distance[3]In radiology practices the CBIR system is becoming important due to the substantial advantage in medicine and feature analysis[4- 5].It also plays significant role in CT scan images MRI scan images and x-rays [6-10].The geometrical structural change of the body is possibility examined and processed by medical image [11].In case of breast cancer the diagnosis is possible by using digital mammography technique which is being widely applicable but in order to safe human body from the bad effects, some safe procedures are preferred which include MRI, Infrared and Biopsy [12].In medical image sequence proposed lossless fast method gives the best sequences of CE and MRI images which ultimately improves the compression rate[13].Modalities in the medical images such as US bladder ,US phantom, CT and MR images are segmented which is using different algorithms e.g. KNN, GAL,ISNN. Segmentation can be used by applying feature extraction method that is (2D-CWT), MGH and a joined version of both 2D-CWT, MGH called hybrid features. Analysis outcome show that KNN did better than GAL for image segmentation the performance of GAL is better than that of ISSN[14].The expansion of the segmentation performed by the distinguishing the each object in proper segmentation with the object in marker controlled watershed segmentation[15].In all the multimedia application the compression of image is used to save storage and transmission time[16]Layered compression technique is used to further compressed the data by using entropy or dictionary or both compression techniques[17].In the last decade the researches paid attention on image retrieval therefore mostly work explained in depth[18].In visual image the angle oriented edges are most effective as perspective of representation. The image description has been well explained by using Pythagorean theory [19].Any data that requires the image features directly from content based data in now days are becoming source of fast retrieval [20-21]. Such system in which performance optimization and retrieval process play necessary role CBIR systems are key challenges[22- 25].Multi-tiered approaches are mostly difficult as these involve microscopic images[26]In recent years the drawback for imaging contains different viewpoints In order to get better the retrieval accuracy SURF is combined among colour feature [27].To overcome such draw backs detectors and descriptors are used[28]several applications are stored and handled using management such problems can be handled by retrieval techniques which may be MULVS system[29-30].Some designed programs such as photo book, visual SEEK, video Q are handled to image video clips and audio clips.[31]Desired features of image in microscopic level are possible to be extracted by the using of technique known as microscopic feature extraction that is useful in the area of medical sciences for tracing the infected parts in the body these may be bone crackes or holes and tainted cell[32] The fast and faster results are possible to achieve in calculations where initial step of Hull quick formulation is applied [33] the maximum margin criterion (MMC) may be used to solve the problems such as small sample size (SSS).This may be stated as the form of LDA which is modified and convex hull[34] Many researchers have paid attention in retrieval of feature of textures colour and shape which is given in [35].Such combination of colours and texture along with shape have been optimized in[36]some examples of such features can be found in[37]system defined in[38]deals with the quires of audio visual[39]which also combines CBIR systems[40].Next level to the image retrieval for better image and performance were followed by the researchers. The proposed methods for retrieving the colour, texture and shapes are being used now days. Some methods can be discussed.

I. SEMANTIC GAP

Algorithms from the evolutionary stage may be called as Evolutionary programming (EP).In such type of programming's the high level feature of keyword is used to be retrieved for the images and then accordingly the use of relevant images and query keywords the retrieval by database is accomplished [41].Dealing with databases of images, semantic analysis giving us the idea that how to make retrieval of similar images [42]. It is mainly divided into two steps. firstly the explanation of the images with their complete description of meanings, are taken from an ontology(study and thought of natural existence).and secondly by calculating the comparison of two similar images based on (HSBD), and also based on natural thoughts these two steps helps us to solve the problem of semantic gap[43].

II. IMAGE DATABASE:

It is necessary to have collection of images with distinct feature where there exist a large number of images this will be easy to separately identify them [44], decision regarding the selection of image has to be taken. Not to choose the similar images with in group. Sometimes first image [45] is selected but nothing is said to process about the selecting an image as it is selected randomly or by the hand.About three million images are handled by the retrieval system of images on large scale. Collateral text and the visual features are the basics of system retrieval of images. Such process that consists of keyword of initial or image based query and results of visual appearance on the feedback .This is possible to set on large scale data [46].

III. WEB BASE DATABASE:

Efforts are required to make the image processing to web retrieval imaging. An image searching engine is proposed to retrieve the web image in improved form [47] .Heterogeneous and inter-active types of data deals with several trends of multi-model image retrieving systems. This is in search still to find the resemblance and combine different models. Iterative model for similarity propagation proposed such approaches are used to learn the semantic similarities of images in the way to leverage the relationship existing between the textual annotations and their images [48].

I. CBIR BY COLOUR

One of the important features in image retrieval is the colour feature [49] Proper selection of colour space is basic issue in local Fourier transform. Different result will be gained by using different colour spaces. [50] Professional and the other users can have good access to changing technologies of internet and colour imaging. Now a day's many sources are used for imaging such as digital cameras, scanner, internet and digital video and all these are increasing challenges for computer systems. This is a challenge to store and index such huge data in efficient manner which could be easily accessible. This corner has been investigated for more than 30 years and successful achievement gained to tackle with image compression colour scheme of human vision which requires to translate such RGB colours into some other. [51-52] The recognition of objects and images can be describe colour histogram [53].

II. CBIR BY SHAPE

In the image the natural feature of shapes of objects is used for image retrieval and image storage [54]. The extensively powerful feature is the shape but these run into two problems. Firstly that image is to be partitioned from region for which descriptors of shape can be extracted secondly the shape for natural objects is to be researched till present[55] Two types of shapes contour-based and region based are usually represented in contour based methods there is a need to extract boundary information which may not be available sometimes. Shape of boundary does not rely on the region based methods. Therefore both kinds of representations may be necessarily made. Four important types of descriptors such as ZMD, GD, FD and CSSD are used in which CSSD and FD are based on contour whereas GD and ZMD based on region [56-57].

II. CBIR BY SHAPE

In the image the natural feature of shapes of objects is used for image retrieval and image storage [54]. The extensively powerful feature is the shape but these run into two problems. Firstly that image is to be partitioned from region for which descriptors of shape can be extracted secondly the shape for natural objects is to be researched till present[55] Two types of shapes contour-based and region based are usually represented in contour based methods there is a need to extract boundary information which may not be available sometimes. Shape of boundary does not rely on the region based methods. Therefore both kinds of representations may be necessarily made. Four important types of descriptors such as ZMD, GD, FD and CSSD are used in which CSSD and FD are based on contour whereas GD and ZMD based on region [56-57].

III. CONTENT BASED IMAGE RETRIEVAL USING NEURAL NETWORK

Several techniques which based on neural networks are used to image retrievals. The adaptive fascinating factor to be used for neural networks in CBIR is the learning capability. Significant performance can be gained by use of difficult techniques and understanding of contents. Different ideas and comments are given in papers on CBIR given in [61-62].Better performance on healthcare imaging and other medical imaging applications can be done by the help of neural networks in such applications as per described in [63-64]. To handle with database containing large image the use of artificial neural network is efficiently proven and is successfully being used in the application of this area [65]

IV. RELEVANCE FEEDBACK SYSTEM FOR CBIR

Some images such as content based images are retrieved by using relevance feedback scheme [73-74].To find about the relevancy of images a list will be needed containing candidate image 75].This idea is used behind the feedback also reflex the parameter by which modification, semantic space and space features can be found [76].According to [77] many approaches have adopted to weight positive and negative examples and for the features of online selection. Power technique in retrieval of information system is based on traditional process. The process of adjusting the query using feedback of information is better to approximate for need of user information the vector equation for optimization of problem has been used as feedback [78] the better understanding using RF model is as it combines the NE and PE in framework of probabilistic manner [79].

V. CONCLUSION

Now days internet is being used frequently which ultimately demands to increase multimedia application along with image retrieval system. In order to share the information now days the mean of digital images is playing very important role. This is much effective in the terms of expression. Visual contents of the image seldom contain the particulars of the situation against which it is meant for. This kind of work is easily handled in the situation where limited numbers of images are stored but the situation becomes worse when the imaged exceed one hundred and one million. In such situations the need of an intelligent searcher raises to find the specific image out of million images in a very short time. In this study luminous work done by the researchers has been illustrated to acknowledge the concept. Analysis of colour and shape can be done by the visual contents of image. Here we have discussed the techniques which are mostly used in CBIR and to improve the retrieval system of images along with their performance.

References

References:

[1] Mussarat Yasmin, Muhammad Sharif, Sajjad Mohsin, Use of Low Level Features for Content Based Image Retrieval: Survey, World Applied Sciences Survey, 2(11), 2013, 65-75.

[2] A. W. M. Smeulders, M. Worring, S. Santini, A. Gupta, and R. Jain, Content-Based Image Retrieval at the End of the Years, IEEE Transactions on Pattern Analysis and Machine Intelligence, 22(12) ,2000, 1349- 1380.

[3] Waqas Haider, Muhammad Sharif Malik, Mudassar Raza, Abdul Wahab, Izhar Ahmed Khan, Umar Zia, Jawad Tanveer, Hadia Bashir, A Hybrid Method for Edge Continuity Based on Pixel Neighbors Pattern Analysis (PNPA) for Remote Sensing Satellite Images, Int. J. Communications Network and System Sciences, (5),2012, 624-630.

[4] Akgül, C. B, Rubin, D. L, Napel, S, Beaulieu, C. F, Greenspan, H, &Acar, B, Content-based image retrieval in radiology current status and future directions, Journal of Digital Imaging ,24(2),2011,208-222.

[5] Ramamurthy, B, and K. R. Chandran, Content based Image Retrieval for Medical Images using Canny Edge Detection Algorithm, International Journal of Computer Applications, 17(6), 2011, 1-37.

[6] Masood, Saleha, Muhammad Sharif, Mussarat Yasmin, Mudassar Raza, and Sajjad Mohsin, Brain Image Compression: A Brief Survey, Research Journal of Applied Sciences, (5), 2013, 49-59.

[7] Yasmin, Mussarat, Sajjad Mohsin, Muhammad Sharif, Mudassar Raza, and Saleha Masood, Brain Image Analysis: A Survey, World Applied Sciences Journal, 19(10), 2012, 1484-1494.

[8] Yasmin, Mussarat, Muhammad Sharif, Saleha Masood, Mudassar Raza, and Sajjad Mohsin, Brain image enhancement: A survey,World Applied Sciences Journal, 17 (9),2012,1192- 1204.

[9] Yasmin, Mussarat, Muhammad Sharif, Saleha Masood, Mudassar Raza, and Sajjad Mohsin, Brain image reconstruction: A short survey, World Applied Sciences Journal, 19(1), 2012, 52-62.

[10] Raza, Mudassar, Muhammad Sharif, Mussarat Yasmin, Saleha Masood, and Sajjad Mohsin, Brain Image Representation and Rendering: A Survey, Research Journal of Applied Sciences, (4), 2012.

[11] Irum, Isma, Mudassar Raza, and Muhammad Sharif, Morphological techniques for medical image, Research Journal of Applied Sciences, Engineering and Technology, 4(17), 2012,2948-2962.

[12] Yasmin, Mussarat, Muhammad Sharif, and Sajjad Mohsin, Survey Paper on Diagnosis of Breast Cancer Using Image Processing Techniques, Research Journal of Recent Sciences, 2(10),2013, 88-98.

[13] Mudassar Raza, Ahmed Adnan, Muhammad Sharif, Syed Waqas Haider, Lossless Compression Method for Medical Image Sequences Using Super-Spatial Structure Prediction and Inter-frame Coding, Journal of Applied Research and Technology, (10),2012.

[14] Maria Hameed, Muhammad Sharif, Mudassar Raza, Syed Waqas Haider, Muhammad Iqbal, Framework for the Comparison of Classifiers for Medical Image Segmentation with Transform and Moment based features, Research Journal of Recent Sciences ,2(6),2013, 1-10.

[15] Amir Shahzad,Muhammad Sharif, Mudassar Raza, Khalid Hussain, Enhanced Watershed Image Processing Segmentation, journal of Information & Communication Technology ,2( 1),2008 , 01-09.

[16] Mehwish Rehman, Muhammad Sharif and Mudassar Raza, Image Compression: A Survey, Research Journal of Applied Sciences, Engineering and Technology, 7(4), 2014,656-672.

[17] Rana Muhammad Nazim, Muhammad sharif, Mudassar Raza, Aman ullah khan, Layered compression technique base on entropy or dictionary Methods, Research Journal of Applied Sciences, (5), 2013.

[18] R. Datta, D. Joshi, J. Li and J. Z. Wang, Image retrieval: Ideas, influences, and trends of the new age, ACM Computing, 40 (2), 2008 ,1-60.

[19] Mussarat Yasmin, Muhammad Sharif, Isma Irum, Sajjad Mohsin, Powerful Descriptor for Image Retrieval Based on Angle Edge and Histograms, Journal of Applied Research and Technology, (11), 2013.

[20] Liu, Ying, Dengsheng Zhang, Guojun Lu, and Wei-Ying Ma, A survey of content-based image retrieval with high-level semantics, Pattern Recognition, 40(1), 2007, 262-282.

[21] Schaefer, Gerald, Content-based image retrieval: some basics, in (First Ed), Man- Machine Interactions 2, (Springer Berlin Heidel berg, 2011) 21-29.

[22] Jain, M, & Singh, S. K , A Survey On: Content Based Image Retrieval Systems Using Clustering Techniques for Large Data sets, International Journal of Managing Information Technology, 3(4), 2011.

[23] Collins, R.T, Yanxi, L, Leordeanu, and M, Online selection of discriminative tracking features, IEEE Trans. Pattern Anal. Mach. Intell, 27(10), 2005, 1631-1643.

[24] Wei, J, Guihua, E, Qionghai, D, Jinwei, G, Similarity-based online feature selection in content-based image retrieval, IEEE Trans. Image Process, 15(3), 2006, 702-712.

[25] Akakin, HaticeCinar, and Metin N. Gurcan, Content-based microscopic image retrieval system for multi-image queries, Information Technology in Biomedicine, IEEE Transactions, 16, (4), 2012, 758-769.

[26] K. Mikolajczyk and C. Schmid, A Performance Evaluation of Local Descriptors, IEEE Transactions on Pattern Analysis and Machine Intelligence, 27(10), 2005, 1615- 1630.

[27] Velmurugan, K., and Lt Dr S. Santhosh Baboo ,Content-based image retrieval using SURF and colour moments, Global Journal of Computer Science and Technology, 11(10), 2011,1-5.

[28] S. Kiranyaz, K. Caglar, O. Guldogan, and E. Karaoglu, MUVIS: A multimedia browsing, indexing and retrieval framework, In Proc. Third Int. Workshop on Content Based Multimedia Indexing, 2003,22-24.

[29] Kiranyaz, Serkan, Kerem Caglar, Esin Guldogan, Olcay, Guldogan, and Moncef Gabbouj. MUVIS: a content-based multimedia indexing and retrieval framework, Proceedings. IEEE Seventh International Symposium. In Signal Processing and Its Applications,(1), 2003,1- 8.

[30] Kiranyaz, Serkan, and Moncef Gabbouj, Hierarchical cellular tree: an efficient indexing scheme for content-based retrieval on multimedia databases, IEEE Transactions, 9(1), 2007, 102-119.

[31] TianYumin and Mei Lixia, Image Retrieval based on Multiple Features using Wavelet, Proc. Fifth International Conference on Computational Intelligence and Multimedia Applications, 2003, 137-142.

[32] Muhammad Sharif, Mudassar Raza, Sajjad Mohsin, Jamal Hussain Shah, Microscopic Feature Extraction Method, Int. J. Advanced Networking and Applications, 4 (5), 2013, 1700-1703.

[33] Sharif, Muhammad,Syeda Zilley Zahra Naqvi, Mudassar Raza, Waqas Haider, A new approach to compute convex hull Innovative Systems, Design and Engineering ,2(3),2011,186-192.

[34] Murtaza, Marryam, Muhammad Sharif, Mudassar Raza, and J. Shah, Face Recognition using Adaptive Margin Fisher's Criterion and Linear Discriminant Analysis, International Arab Journal of Information Technology, 11(2),2014,1-11.

[35] Priya, R, and VasanthaKalyani David, Optimized Content based Image Retrieval System based on Multiple Feature Fusion Algorithm, International Journal of Computer Applications, 31(8), 2011.

[36] Qianni Zhang and EbroulIzquierdo, Histology Image Retrieval in Optimized Multi feature Spaces, IEEE Journal of Biomedical and Health Informatics,17(1), 2013 ,240-249.

[37] Roland Kwitt, Peter Meerwald and Andreas Uhl, Efficient Texture Image Retrieval Using Copulas in a Bayesian Framework,IEEE transactions on image processing, 2011,1- 20.

[38] Naphade, Milind R, and Thomas S. Huang, Extracting semantics from audio-visual content: the final frontier in multimedia retrieval Neural Networks, IEEE Transactions on, 13 (4),2002,793-810.

[39] Yixin Chen Member, IEEE, James Z. Wang, Member, IEEE and Robert Krovetz, CLUE, Cluster-Based Retrieval of Image by Unsupervised Learning, IEEE transactions on image Processing, 2005,1-14.

[40] Singh, Sanjay T, and Trimbak Ramchandra Sontakke, Neutralizing the Semantic Gap in the Content Based Image Retrieval (CBIR) System, European Journal of Scientific Research, 75 (4), 2012, 536-549.

[41] Kurtz, Camille, Christopher F. Beaulieu, Sandy Napel, and Daniel L. Rubin, A hierarchical knowledge-based approach for retrieving similar medical images described with semantic annotations, Journal of biomedical informatics, 2014, 1-18.

[42] Rehman, Mehwish, Muhammad Iqbal, Muhammad Sharif, and Mudassar Raza, Content Based Image Retrieval: Survey, World Applied Sciences Journal, 19 (3),2012, 404-412.

[43] Li, Zhixin, Zhongzhi Shi, Weizhong Zhao, Zhiqing Li, and Zhenjun Tang, Learning semantic concepts from image database with hybrid generative/discriminative approach, Engineering Applications of Artificial Intelligence, 26(9), 2013, 2143-2152.

[44] Müller, Henning, Stéphane Marchand- Maillet, and Thierry Pun, The truth about corel-evaluation in image retrieval, in (First Ed), Image and Video Retrieval, (Springer Berlin Heidelberg, 2002) ,38-49.

[45] Shyu, Mei-Ling, Shu-Ching Chen, Min Chen, and Chengcui Zhang, A unified framework for image database clustering and contentbased retrieval, Proceedings. 2nd ACM international workshop. on Multimedia databases, 2004, 19-27.

[46] Quack, Till, Ullrich Mönich, Lars Thiele, and B. S. Manjunath, Cortina: a system for largescale, content-based web image retrieval, Proceedings.12th annual ACM international .conference on Multimedia, 2004 ,508-511.

[47] Lin, Wei-Hao, Rong Jin, and Alexander Hauptmann, Web image retrieval re-ranking with relevance model , Proceedings. IEEE/WIC International Conference .on Web Intelligence, 2003,242-248.

[48] Wang, Xin-Jing, et al,Multi-model similarity propagation and its application for web image retrieval, Proceedings .12th annual ACM international conference. On Multimedia, 2004,944-951.

[49] Schaefer, Gerald, and Michal Stich, UCID: an uncompressed color image database, Proc. SPIE 5307 .on Storage and Retrieval Methods and Applications for Multimedia, 2004, 472-480.

[50] Yu, Hui, Mingjing Li, Hong-Jiang Zhang, and JufuFeng, Color texture moments for content-based image retrieval In Image Processing, Proceedings. International Conference on IEEE, (3), 2002,929-932.

[51] Kekre, H. B, and Sudeep D. Thepade, Boosting Block Truncation Coding with Kekre's LUV Color Space for Image Retrieval, International Journal of Electrical, Computer & Systems Engineering,2(3),2008.

[52] Gagaudakis, George, and Paul L. Rosin, Incorporating shape into histograms for CBIR,Pattern Recognition, 35 (1),2002,81- 91.

[53] Zhang, Dengsheng, and GuojunLu, Content- Based Shape Retrieval Using Different Shape Descriptors: A Comparative Study, ICME, 2001, 1-4.

[54] Mussarat Yasmin, Muhammad Sharif, Sajjad Mohsin, Image Retrieval Techniques Using Shapes of Object: A Survey, Sci.Int (Lahore), 25(4), 2013, 723-729.

[55] Yasmin, Mussarat, Sajjad Mohsin, Isma Irum, and Muhammad Sharif, Content based image retrieval by shape, color and relevance feedback, Life Science Journal 10 (4s), 2013, 593-598.

[56] Prasad, B. G., K. K. Biswas, and S. K. Gupta,Region-based image retrieval using integrated color, shape, and location index, Computer vision and image understanding ,94 (1),2004,193-233.

[57] Singh, Chandan, Improving image retrieval using combined features of Hough transform and Zernike moments, Optics and Lasers in Engineering ,49(12),2011,1384-1396.

[58] Cheng, Heng-Da, X. H. Jiang, Ying Sun, and Jingli Wang, Color image segmentation advances and prospects, Pattern recognition, 34 (12), 2001,2259-2281.

[59] Qin, Chanchan, Guoping Zhang, Yicong Zhou, Wenbing Tao, and Zhiguo Cao, Integration of the saliency-based seed extraction and random walks for image segmentation, Neurocomputing, (129), 2013, 378-391.

[60] Zhang, Jun, and Lei Ye, Series feature aggregation for content-based image retrieval, Computers & electrical engineering ,36(4), 2010, 691-701.

[61] Seetharaman, K, and S. Sathiamoorthy,Color image retrieval using statistical model and radial basis function neural network, Egyptian Informatics Journal ,15 (1),2014,59-68.

[62] Sánchez-Cruz, Hermilo, and Ernesto Bribiesca, Polygonal approximation of contour shapes using corner detectors, Journal of applied research and technology, 7(3), 2009,275-290.

[63] Tao, Dacheng, Xiaoou Tang, Xuelong Li, and Yong Rui, Direct kernel biased discriminant analysis: a new content-based image retrieval relevance feedback algorithm, Multimedia IEEE Transactions ,8 (4) ,2006,716-727.

[64] Yasmin, Mussarat, Muhammad Sharif, and Sajjad Mohsin, Neural Networks in Medical Imaging Applications: A Survey, World Applied Sciences Journal, 22 (1) ,2013.

[65] Noda, Kuniaki, Hiroaki Arie, Yuki Suga, and Tetsuya Ogata, Multimodal integration learning of robot behavior using deep neural networks, Robotics and Autonomous Systems ,62 (6), 2014.721-736.

[66] Kurtz, Camille, Christopher F. Beaulieu, Sandy Napel, and Daniel L. Rubin, A hierarchical knowledge-based approach for retrieving similar medical images described with semantic annotations, Journal of biomedical informatics,2014,1-18.

[67] Gao, Chao, Dongguo Zhou, and YongcaiGuo, Automatic iterative algorithm for image segmentation using a modified pulse-coupled neural network, Neurocomputing,(119),2013,332-338.

[68] Son, Le Hoang, Nguyen Duy Linh, and Hoang Viet Long, A lossless DEM compression for fast retrieval method using fuzzy clustering and MANFIS neural network, Engineering Applications of Artificial Intelligence, 2013,33-42.

[69] Humberto Sossn, ElizabethGuevara, Efficient training for dendrite morphological neural networks, Neurocomputing, 2013, 132-142.

[70] Sudipta.Mukhopadhyay, A new contentbased image retrieval technique using fuzzy class membership,Neur°Computing, 34(6), 2013,646-654.

[71] Abrahams, Alan S. Eloise Coupey, Eva X. Zhong, Reza Barkhi, and Pete S. Manasantivongs, Audience targeting by B-to- B advertisement classification: A neural network approach, Expert Systems with Applications, 40,(8),2013,2777-2791.

[72] Milanova, Mariofanna, Roumen Kountchev, Stuart Rubin, Vladimir Todorov, and Roumiana Kountcheva,Content Based Image Retrieval Using Adaptive Inverse Pyramid Representation, ,In (First Ed) ,Human Interface and the Management of Information and Interaction ,(Springer Berlin Heidelberg, 2009)304-314.

[73] Yasmin, Mussarat, Sajjad Mohsin, and Muhammad Sharif, Intelligent Image Retrieval Techniques A Survey, Journal of Applied Research and Technology,12 (1),2013,87-103.

[74] Seo, Kwang-Kyu , An application of oneclass support vector machines in contentbased image retrieval, Expert Systems with Applications, 33 (2), 2007,491-498.

[75] Su, Zhong, Hongjiang Zhang, Stan Li, and Shaoping Ma, Relevance feedback in contentbased image retrieval Bayesian framework, feature subspaces and progressive learning, Image Processing IEEE Transactions, 12 (8),2003,924-937.

[76] Lew, Michael S, Nicu Sebe, Chabane Djeraba, and Ramesh Jain, Content-based multimedia information retrieval State of the art and challenges, ACM Transactions on Multimedia Computing, Communications, and Applications (TOMCCAP), 2 (1), 2006,1- 19.

[77] Rahimi, M, M.H. Alizadeh, R. Rajabi and N. Mehrshad, The Comparison of Innovative Image Processing and Goniometer Methods in Q Angle Measurement ,World Applied Sciences Journal,18(2),2001,226-232.

[78] Rui, Yong. Relevance feedback a power tool for interactive content-based image retrieval, Circuits and Systems for Video Technology, IEEE Transactions, 1998, 644-655.

[79] MacArthur, S. D, Brodley, C. E, Kak, A. C, & Broderick, L. S, Interactive content-based image retrieval using relevance feedback, Computer Vision and Image Understanding,88(2), 2002,55-75 ,2002.

[80] Arevalillo-Herráez, Miguel, and Francesc J.Ferri, An improved distance-based relevance feedback strategy for image retrieval, Image and Vision Computing,31(10),2013,704-713.

[81] da Silva, André Tavares, Jefersson Alex dos Santos, Alexandre Xavier Falcao, Ricardo da S. Torres, and Léo Pini Magalhães, Incorporating multiple distance spaces in optimum-path forest classification to improve feedback-based learning, Computer Vision and Image Understanding ,116 (4),2012,510- 523.

[82] Piras, Luca, and Giorgio Giacinto, Synthetic pattern generation for imbalanced learning in image retrieval, Pattern Recognition Letters ,33(16),2012,2198-2205.

[83] Metternich, Michael J, and Marcel Worring, Track based relevance feedback for tracing persons in surveillance videos, Computer Vision and Image Understanding ,117 (3),2013,229-237.

[84] Saneifar, Hassan, Stéphane Bonniol, Pascal Poncelet, and Mathieu Roche, Enhancing passage retrieval in log files by query expansion based on explicit and pseudo relevance feedback, Computers in Industry,2014,1-15.

[85] Wang, Xiang-Yang, Yong-Wei Li, Hong- Ying Yang, and Jing-Wei Chen, An image retrieval scheme with relevance feedback using feature reconstruction and SVM reclassification, Neurocomputing,(127),2014, 214-230