img(Anaktisi) is Loading...
Loading
Please Wait...
In this web-site a new set of feature descriptors is presented in a retrieval system. These descriptors have been designed with particular attention to their size and storage requirements, keeping them as small as possible without compromising their discriminating ability. These descriptors incorporate color and texture information into one histogram while keeping their sizes between 23 and 74 bytes per image.

FCTH (Acronym of the “Fuzzy Color and Texture Histogram”) constituted by 8 regions, as these are determined by a fuzzy system that takes decision with regards to the texture of the image. Each region is constituted by 24 individual regions, as these result from a second fuzzy system that takes decision with regards to the color of the image. Overall, the final histogram that results includes 8 X 24 = 192 bins. Each bin is then quantized in 3 Bits, limiting the length of the descriptor to the 72 Bytes.

CEDD (Acronym of the “Color and Edge Directivity Descriptor”) constituted by 6 regions, as these are determined by a texture unit. Each region is constituted by 24 individual regions, as these result from a fuzzy system that takes decision with regards to the color of the image. Overall, the final histogram that results includes 6 X 24 = 144 bins. Each bin is then quantized in 3 Bits, limiting the length of the descriptor to the 54 Bytes.

C.FCTH (Acronym of the “Compact Fuzzy Color and Texture Histogram”) constituted by 8 regions, as these are determined by a fuzzy system that takes decision with regards to the texture of the image. Each region is constituted by 10 individual regions, as these result from a second fuzzy system that takes decision with regards to the color of the image. Overall, the final histogram that results includes 8 X 10 = 80 bins. Each bin is then quantized in 3 Bits, limiting the length of the descriptor to the 30 Bytes.

C.CEDD (Acronym of the “Compact Color and Edge Directivity Descriptor”) constituted by 6 regions, as these are determined by a texture unit. Each region is constituted by 10 individual regions, as these result from a fuzzy system that takes decision with regards to the color of the image. Overall, the final histogram that results includes 6 X 10 = 60 bins. Each bin is then quantized in 3 Bits, limiting the length of the descriptor to the 23 Bytes

Also, an Auto Relevance Feedback (ARF) technique is introduced which is based on the proposed descriptors. This technique readjusts the initial retrieval results based on user preferences improving the retrieval score significantly.The goal of theAutomatic Relevance Feedback (ARF) algorithm is to optimally readjust the initial retrieval results based on user preferences. During this procedure the user selects from the first round of retrieved images one or more as being relevant to his/her initial retrieval expectations. Information from these selected images is used to alter the initial query image descriptor.

Primarily, the initial image query one-dimensional descriptor is transformed to a three-dimensional vector W based on the inner features of the descriptor. When the user selects an relevant image from the retrieval results, each bin of that selected image's descriptor X updates the corresponding value of the W vector in a Kohonen Self Organized Featured Map (KSOFM) manner so that it moves closer to the new value emerging from X. The final descriptor to query the image database is formed by the values of the three-dimension vector W. The above procedure is repeated every time the user selects a relevant image.

download | citation
overview video
The rapid growth of digital images through the widespread popularization of computers and the Internet makes the development of an efficient image retrieval technique imperative. Content-based image retrieval, known as CBIR, extracts several features that describe the content of the image, mapping the visual content of the images into a new space called the feature space. The feature space values for a given image are stored in a descriptor that can be used for retrieving similar images. The key to a successful retrieval system is to choose the right features that represent the images as accurately and uniquely as possible. The features chosen have to be discriminative and sufficient in describing the objects present in the image. To achieve these goals, CBIR systems use three basic types of features: color features, texture features and shape features. It is very difficult to achieve satisfactory retrieval results using only one of these feature types.

To date, many proposed retrieval techniques adopt methods in which more than one feature type is involved. For instance, color, texture and shape features are used in both IBM's QBIC and MIT's Photobook. QBIC uses color histograms, a moment-based shape feature, and a texture descriptor. Photobook uses appearance features, texture features, and 2D shape features. Other CBIR systems include SIMBA, CIRES, SIMPLIcity, IRMA, FIRE and MIRROR. A cumulative body of research presents extraction methods for these feature types.

In most retrieval systems that combine two or more feature types, such as color and texture, independent vectors are used to describe each kind of information. It is possible to achieve very good retrieval scores by increasing the size of the descriptors of images that have a high dimensional vector, but this technique has several drawbacks. If the descriptor has hundreds or even thousands of bins, it may be of no practical use because the retrieval procedure is significantly delayed. Also, increasing the size of the descriptor increases the storage requirements which may have a significant penalty for databases that contain millions of images. Many presented methods limit the length of the descriptor to a small number of bins, leaving the possible factor values in decimal, non-quantized, form.

The Moving Picture Experts Group (MPEG) defines a standard for content-based access to multimedia data in their MPEG-7 standard.. This standard identifies a set of image descriptors that maintain a balance between the size of the feature and the quality of the retrieval results.

In this web-site a new set of feature descriptors is presented in a retrieval system. These descriptors have been designed with particular attention to their size and storage requirements, keeping them as small as possible without compromising their discriminating ability. These descriptors incorporate color and texture information into one histogram while keeping their sizes between 23 and 74 bytes per image.

High retrieval scores in content-based image retrieval systems can be attained by adopting relevance feedback mechanisms. These mechanisms require the user to grade the quality of the query results by marking the retrieved images as being either relevant or not. Then, the search engine uses this grading information in subsequent queries to better satisfy users' needs. It is noted that while relevance feedback mechanisms were first introduced in the information retrieval field, they currently receive considerable attention in the CBIR field. The vast majority of relevance feedback techniques proposed in the literature are based on modifying the values of the search parameters so that they better represent the concept the user has in mind. Search parameters are computed as a function of the relevance values assigned by the user to all the images retrieved so far. For instance, relevance feedback is frequently formulated in terms of the modification of the query vector and/or in terms of adaptive similarity metrics.

Also, in this web-site an Auto Relevance Feedback (ARF) technique is introduced which is based on the proposed descriptors. The goal of the proposed Automatic Relevance Feedback (ARF) algorithm is to optimally readjust the initial retrieval results based on user preferences. During this procedure the user selects from the first round of retrieved images one as being relevant to his/her initial retrieval expectations. Information from these selected images is used to alter the initial query image descriptor.


[ download | citation ] Close Close
Close Close [ download | citation ]
Close Close
You need to have flash enable to see the Video Preview
Close Close
For questions about the descriptors (CEDD, FCTH, etc) contact Savvas Chatzichristofis. For questions about the Relevance Feedback and generally about the img(Anaktisi) web site contact Konstantinos Zagoris.

phD student Savvas Chatzichristofis (email:savvash@gmail.com) received the diploma of Electrical Engineer in 2004 from the Democritus University of Thrace (DUTH), Greece and now is PhD candidate at the same university in the area of machine intelligent, neural networks, Fuzzy Logic and computer vision. He is an establishing member of the «Greek Open Source Adherents’ Club». He is also Member in Cyprus Scientific and Technical Chamber since 2005, licentiate in the fields of Electronics, Information Science and Electrical Mechanics. He is also registered in the Electro-mechanical Service of Cyprus since 2007.

Dr Konstantinos Zagoris (email:kzagoris@gmail.com) received the Diploma in Electrical and Computer Engineering in 2003 from Democritus University of Thrace, Greece and his phD from the same univercity in 2010. His research interests include document retrieval, color image processing and analysis, document analysis and pattern recognition. He is a member of the Technical Chamber of Greece.

Assoc. Professor Yiannis S. Boutalis (email:ybout@ee.duth.gr) received the diploma of Electrical Engineer in 1983 from the Democritus University of Thrace (DUTH), Greece and the PhD degree in Electrical and Computer Engineering (topic Image Processing) in 1988 from the Computer Science Division of National Technical University of Athens, Greece. Since 1996, he serves as a faculty member, at the Department of Electrical and Computer Engineering, DUTH, Greece, where he is currently an Associate Professor and director of the automatic control systems lab. He served as an assistant visiting professor at University of Thessaly, Greece, and as a visiting professor in Air Defence Academy of General Staff of air forces of Greece. He also served as a researcher in the Institute of Language and Speech Processing (ILSP), Greece, and as a managing director of the R&D SME Ideatech S.A, Greece, specializing in pattern recognition and signal processing applications. His current research interests are focused in the development of Computational Intelligence techniques with applications in Control, Pattern Recognition, Signal and Image Processing Problems.

Professor Nikos Papamarkos (email:papamark@ee.duth.gr) received his Diploma Degree in Electrical and Mechanical Engineering from the University of Thessaloniki, Thessaloniki, Greece, in 1979 and the Ph.D. Degree in Electrical Engineering in 1986, from the Democritus University of Thrace, Greece.From 1987 to 1990 Dr. Papamarkos was a Lecture, from 1990 to 1996 Assistant Professor, 1996-2003 Associate Professor in the Democritus University of Thrace where he is currently Professor since 2003. During 1987 and 1992 he has also served as a Visiting Research Associate at the Georgia Institute of Technology, USA. His current research interests are in digital image processing, computer vision, document processing, analysis and recognition, pattern recognition, neural networks, signal processing, filter design and optimization algorithms. He has published a number of Journal and Conferences. Also, he is author of three Greek books. Professor Nikos Papamarkos is a Senior Member of IEEE, Member of IAPR, Member of IEE and Member of the Greek Technical Chamber.

Democritus University of Thrace - Department of Electrical and Computer Engineering


This project constitutes collaboration of the Automatic Control Systems (ACSL) and Image Processing and Multimedia (IPML) laboratories.
PhD candidate Savvas A . Chatzichristofis
PhD candidate Konstantinos Zagoris
Assoc. Professor Yiannis S. Boutalis
Professor Nikos Papamarkos

Descriptors References:
J[1]. S. A. Chatzichristofis, K. Zagoris, Y. S. Boutalis and Nikos Papamarkos, “ACCURATE IMAGE RETRIEVAL BASED ON COMPACT COMPOSITE DESCRIPTORS AND RELEVANCE FEEDBACK INFORMATION”, «International Journal of Pattern Recognition and Artificial Intelligence (IJPRAI) », submitted for publication.

C[1]. S. Α. Chatzichristofis and Y. S. Boutalis, “CEDD: COLOR AND EDGE DIRECTIVITY DESCRIPTOR - A COMPACT DESCRIPTOR FOR IMAGE INDEXING AND RETRIEVAL.”, « 6th International Conference in advanced research on Computer Vision Systems ICVS 2008» Proceedings: Lecture Notes in Computer Science (LNCS) pp.312-322, May 12 to May 15, 2008, Santorini, Greece.

C[2]. S. A. Chatzichristofis and Y. S. Boutalis, “FCTH: FUZZY COLOR AND TEXTURE HISTOGRAM- A LOW LEVEL FEATURE FOR ACCURATE IMAGE RETRIEVAL” «9th International Workshop on Image Analysis for Multimedia Interactive Services”, Proceedings: IEEE Computer Society , May 7 to May 9, 2008, Klagenfurt, Austria

C[3]. S. A. Chatzichristofis and Y. S. Boutalis, “A HYBRID SCHEME FOR FAST AND ACCURATE IMAGE RETRIEVAL BASED ON COLOR DESCRIPTORS”, «IASTED International Conference on Artificial Intelligence and Soft Computing (ASC 2007)» Proceedings: ACTA PRESS pp.280-285, August 29 to August 31, 2007, at Palma De Mallorca, Spain.

C[4]. Savvas A. Chazichristofis and Yiannis S. Boutalis, "CONTENT BASED MEDICAL IMAGE INDEXING AND RETRIEVAL USING A FUZZY COMPACT COMPOSITE DESCRIPTOR" Signal Processing, Pattern Recognition and Applications (SPPRA 2009), Innsbruck, Austria, February 17 – 19, 2009, To Appear.

C[5]. Savvas A. Chazichristofis, Mathias Lux and Yiannis S. Boutalis, "SELECTION OF THE PROPER COMPACT COMPOSITE DESCRIPTOR FOR IMPROVING CONTENT BASED IMAGE RETRIEVAL", Signal Processing, Pattern Recognition and Applications (SPPRA 2009), Innsbruck, Austria, February 17 – 19, 2009. To Appear.

C[6]. Savvas A. Chazichristofis and Yiannis S. Boutalis, "CONTENT BASED MEDICAL IMAGE INDEXING AND RETRIEVAL USING A NEW FUZZY COMPACT COMPOSITE DESCRIPTOR", Submited for publication

Image Database References:
  • Jia Li, James Z. Wang, ``Automatic linguistic indexing of pictures by a statistical modeling approach,'' IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 25, no. 9, pp. 1075-1088, 2003.
  • James Z. Wang, Jia Li, Gio Wiederhold, ``SIMPLIcity: Semantics-sensitive Integrated Matching for Picture LIbraries,'' IEEE Trans. on Pattern Analysis and Machine Intelligence, vol 23, no.9, pp. 947-963, 2001.
  • G. Schaefer and M. Stich, “UCID - An Uncompressed Colour Image Database”, 9th International Workshop on Image Analysis for IASTED International Conference on Artificial SPIE, Storage and Retrieval Methods and Applications for Multimedia, San Jose, USA, 2004, pp.472-480.
  • D. Nistér and H. Stewénius. Scalable recognition with a vocabulary tree. In IEEE Conference on Computer Vision and Pattern Recognition (CVPR), volume 2, pages 2161-2168, June 2006
  • IRMA 2005 database is courtesy of TM Deserno, Dept. of Medical Informatics, RWTH Aachen, Germany
  • M. J. Huiskes, M. S. Lew (2008). The MIR Flickr Retrieval Evaluation. ACM International Conference on Multimedia Information Retrieval (MIR'08), Vancouver, Canada
overview contact Links credits technology
Calculating the Descriptors. Please Wait...
Loading
Image Guide
 
Original Image
Query Image Information
Image Name:
DataSet:
NMRR:
Recall:
Descriptor:
Relev. Feedback:
CPU Exec. Time:
Choose Action
Start Another Search
Show Search History
Clear Search History
Total Images:
Search by :
  • DataSet:
  • Descriptor:
  •  Previous
  •   Page:
  • Next 
125.jpg
Search Similar Images
Show Annotated Keywords
animal: 45,46%
bird: 0%
face: 47%
boat: 0%
buildings: 57,32%
car: 51,95%
water: 71,48%
mountain: 58,48%
night: 0%
field: 38,53%
sky: 62,25%
sunset: 0%
plants: 45,66%
Original Image
327.jpg
Search Similar Images
Show Annotated Keywords
animal: 39,31%
bird: 0%
face: 59,69%
boat: 0%
buildings: 52,48%
car: 88,42%
water: 49,33%
mountain: 56,35%
night: 0%
field: 35,26%
sky: 46,73%
sunset: 0%
plants: 41,04%
Original Image
251.jpg
Search Similar Images
Show Annotated Keywords
animal: 21,14%
bird: 0%
face: 35,75%
boat: 0%
buildings: 65,82%
car: 42,01%
water: 34,6%
mountain: 38,32%
night: 0%
field: 19,92%
sky: 33,03%
sunset: 0%
plants: 28,96%
Original Image
558.jpg
Search Similar Images
Show Annotated Keywords
animal: 33,15%
bird: 0%
face: 31,07%
boat: 0%
buildings: 30,8%
car: 32,15%
water: 37,46%
mountain: 37,11%
night: 0%
field: 27%
sky: 42,1%
sunset: 0%
plants: 44,2%
Original Image
400.jpg
Search Similar Images
Show Annotated Keywords
animal: 49,23%
bird: 0%
face: 42,72%
boat: 0%
buildings: 39,91%
car: 37,8%
water: 46,89%
mountain: 43,38%
night: 0%
field: 36,84%
sky: 50,85%
sunset: 0%
plants: 56,3%
Original Image
616.jpg
Search Similar Images
Show Annotated Keywords
animal: 58,49%
bird: 0%
face: 61,58%
boat: 0%
buildings: 47,22%
car: 56,99%
water: 42,02%
mountain: 43,34%
night: 0%
field: 48,77%
sky: 41,24%
sunset: 0%
plants: 50,06%
Original Image
703.jpg
Search Similar Images
Show Annotated Keywords
animal: 65,1%
bird: 0%
face: 47,44%
boat: 0%
buildings: 42,73%
car: 40,26%
water: 42,88%
mountain: 41,41%
night: 0%
field: 64,99%
sky: 42,05%
sunset: 0%
plants: 82,62%
Original Image
967.jpg
Search Similar Images
Show Annotated Keywords
animal: 27,17%
bird: 0%
face: 41,72%
boat: 0%
buildings: 15,4%
car: 20,02%
water: 17,68%
mountain: 19,12%
night: 0%
field: 25,49%
sky: 18,72%
sunset: 0%
plants: 21,24%
Original Image
55.jpg
Search Similar Images
Show Annotated Keywords
animal: 34,34%
bird: 0%
face: 73,02%
boat: 0%
buildings: 44,62%
car: 71,58%
water: 40,04%
mountain: 46,45%
night: 0%
field: 30,13%
sky: 39,38%
sunset: 0%
plants: 36,8%
Original Image
  • DataSet:
  • Descriptor:
  • Previous Page Previous
  •   Page:
  • Next Next Page